What would you do if you knew?

Transcription

1 What would you do if you knew? Teradata Database Support Utilities Release B K December 2016

2 The product or products described in this book are licensed products of Teradata Corporation or its affiliates. Teradata, Applications-Within, Aster, BYNET, Claraview, DecisionCast, Gridscale, MyCommerce, QueryGrid, SQL-MapReduce, Teradata Decision Experts, "Teradata Labs" logo, Teradata ServiceConnect, Teradata Source Experts, WebAnalyst, and Xkoto are trademarks or registered trademarks of Teradata Corporation or its affiliates in the United States and other countries. Adaptec and SCSISelect are trademarks or registered trademarks of Adaptec, Inc. Amazon Web Services, AWS, [any other AWS Marks used in such materials] are trademarks of Amazon.com, Inc. or its affiliates in the United States and/or other countries. AMD Opteron and Opteron are trademarks of Advanced Micro Devices, Inc. Apache, Apache Avro, Apache Hadoop, Apache Hive, Hadoop, and the yellow elephant logo are either registered trademarks or trademarks of the Apache Software Foundation in the United States and/or other countries. Apple, Mac, and OS X all are registered trademarks of Apple Inc. Axeda is a registered trademark of Axeda Corporation. Axeda Agents, Axeda Applications, Axeda Policy Manager, Axeda Enterprise, Axeda Access, Axeda Software Management, Axeda Service, Axeda ServiceLink, and Firewall-Friendly are trademarks and Maximum Results and Maximum Support are servicemarks of Axeda Corporation. CENTOS is a trademark of Red Hat, Inc., registered in the U.S. and other countries. Cloudera, CDH, [any other Cloudera Marks used in such materials] are trademarks or registered trademarks of Cloudera Inc. in the United States, and in jurisdictions throughout the world. Data Domain, EMC, PowerPath, SRDF, and Symmetrix are registered trademarks of EMC Corporation. GoldenGate is a trademark of Oracle. Hewlett-Packard and HP are registered trademarks of Hewlett-Packard Company. Hortonworks, the Hortonworks logo and other Hortonworks trademarks are trademarks of Hortonworks Inc. in the United States and other countries. Intel, Pentium, and XEON are registered trademarks of Intel Corporation. IBM, CICS, RACF, Tivoli, and z/os are registered trademarks of International Business Machines Corporation. Linux is a registered trademark of Linus Torvalds. LSI is a registered trademark of LSI Corporation. Microsoft, Active Directory, Windows, Windows NT, and Windows Server are registered trademarks of Microsoft Corporation in the United States and other countries. NetVault is a trademark or registered trademark of Dell Inc. in the United States and/or other countries. Novell and SUSE are registered trademarks of Novell, Inc., in the United States and other countries. Oracle, Java, and Solaris are registered trademarks of Oracle and/or its affiliates. QLogic and SANbox are trademarks or registered trademarks of QLogic Corporation. Quantum and the Quantum logo are trademarks of Quantum Corporation, registered in the U.S.A. and other countries. Red Hat is a trademark of Red Hat, Inc., registered in the U.S. and other countries. Used under license. SAP is the trademark or registered trademark of SAP AG in Germany and in several other countries. SAS and SAS/C are trademarks or registered trademarks of SAS Institute Inc. Simba, the Simba logo, SimbaEngine, SimbaEngine C/S, SimbaExpress and SimbaLib are registered trademarks of Simba Technologies Inc. SPARC is a registered trademark of SPARC International, Inc. Symantec, NetBackup, and VERITAS are trademarks or registered trademarks of Symantec Corporation or its affiliates in the United States and other countries. Unicode is a registered trademark of Unicode, Inc. in the United States and other countries. UNIX is a registered trademark of The Open Group in the United States and other countries. Other product and company names mentioned herein may be the trademarks of their respective owners. The information contained in this document is provided on an "as-is" basis, without warranty of any kind, either express or implied, including the implied warranties of merchantability, fitness for a particular purpose, or non-infringement. Some jurisdictions do not allow the exclusion of implied warranties, so the above exclusion may not apply to you. In no event will Teradata Corporation be liable for any indirect, direct, special, incidental, or consequential damages, including lost profits or lost savings, even if expressly advised of the possibility of such damages. The information contained in this document may contain references or cross-references to features, functions, products, or services that are not announced or available in your country. Such references do not imply that Teradata Corporation intends to announce such features, functions, products, or services in your country. Please consult your local Teradata Corporation representative for those features, functions, products, or services available in your country. Information contained in this document may contain technical inaccuracies or typographical errors. Information may be changed or updated without notice. Teradata Corporation may also make improvements or changes in the products or services described in this information at any time without notice. To maintain the quality of our products and services, we would like your comments on the accuracy, clarity, organization, and value of this document. Please teradata-books@lists.teradata.com Any comments or materials (collectively referred to as "Feedback") sent to Teradata Corporation will be deemed non-confidential. Teradata Corporation will have no obligation of any kind with respect to Feedback and will be free to use, reproduce, disclose, exhibit, display, transform, create derivative works of, and distribute the Feedback and derivative works thereof without limitation on a royalty-free basis. Further, Teradata Corporation will be free to use any ideas, concepts, know-how, or techniques contained in such Feedback for any purpose whatsoever, including developing, manufacturing, or marketing products or services incorporating Feedback. Copyright by Teradata. All Rights Reserved.

3 Preface Purpose This book, Support Utilities, describes utility programs most often used to support Teradata Database. These utilities are used primarily by Teradata Support personnel and field engineers. Notice: These utilities manipulate Teradata Database at a low level. Incorrect use of these utilities can damage your system. Before using these utilities, consult with the Teradata Support Center. Chapter names reflect the utility common name followed by the name of the executable utility program enclosed in parentheses, for example, Control GDO Editor (ctl). Use the executable program name to start the utility from the command line or from Database Window. Some utilities run only from Database Window (DBW). For information, see the Database Window (xdbw) chapter in Utilities. Audience This book is intended for database administrators and other technical personnel who use Teradata Database. Supported Software Releases and Operating Systems This book supports Teradata Database Teradata Database is supported on: SUSE Linux Enterprise Server (SLES) 11 SP1 SUSE Linux Enterprise Server (SLES) 11 SP3 Teradata Database client applications support other operating systems. Prerequisites You should be familiar with basic relational database management theory and technology. To become familiar with concepts specific to Teradata Database, read Introduction to Teradata and SQL Fundamentals. Support Utilities, Release

4 Preface Changes to This Book Changes to This Book Release Utility Description Teradata Database December 2016 Filer New CLEAREIOCACHE command New DIRECTSTORAGE option to DUMPDISK command New SNAPMI command and snapshot MI feature for faster file system startups. SCANDISK uses full cylinder reads by default when scanning [W]DBs if the scanned AMP is offline (down/fatal) or if the database is not running. Appendix: Starting the Utilities Removed discussion of HUTCNS from Starting the Utilities appendix. The HUTCNS utility was discontinued, starting with Teradata Databaserelease Additional Information URL Teradata Community - Developer Zone Teradata Downloads Description Use the Teradata Information Products Publishing Library site to: View or download a manual: 1. Under Online Publications, select General Search. 2. Enter your search criteria and click Search. Download a documentation CD-ROM: 1. Under Online Publications, select General Search. 2. In the Title or Keyword field, enter CD-ROM, and click Search. The Teradata home page provides links to numerous sources of information about Teradata. Links include: Executive reports, white papers, case studies of customer experiences with Teradata, and thought leadership Technical information, solutions, and expert advice Press releases, mentions and media resources Teradata Customer Education delivers training that builds skills and capabilities for our customers, enabling them to maximize their Teradata investment. Use Your Service to access Orange Books, technical alerts, and knowledge repositories, view and join forums, and download software patches. Developer Zone provides new ideas and developer information. Provides publicly available downloads from Teradata. 4 Support Utilities, Release 16.00

5 Preface Product Safety Information To maintain the quality of our products and services, we would like your comments on the accuracy, clarity, organization, and value of this document. Please Product Safety Information This document may contain information addressing product safety practices related to data or property damage, identified by the word Notice. A notice indicates a situation which, if not avoided, could result in damage to property, such as equipment or data, but not related to personal injury. Example Notice: Improper use of the Reconfiguration utility can result in data loss. Teradata Database Optional Features This book may include descriptions of the following optional Teradata Database features and products: In-Memory Optimization Teradata Columnar Teradata Row-Level Security Teradata Secure Zones Teradata Temporal Teradata Virtual Storage (VS) You may not use these features without the appropriate licenses. The fact that these features may be included in product media or downloads, or described in documentation that you receive, does not authorize you to use them without the appropriate licenses. Contact your Teradata sales representative to purchase and enable optional features. Support Utilities, Release

6 Preface Teradata Database Optional Features 6 Support Utilities, Release 16.00

7 CHAPTER 1 Teradata Database Utilities Overview This chapter lists the Teradata Database utilities that are documented in Utilities and Support Utilities. Alphabetical Listing of Utilities Utility Abort Host (aborthost) CheckTable (checktable) CNS Run (cnsrun) Configuration Utility (config) Purpose Aborts all outstanding transactions running on a failed host, until the system restarts the host. Checks for inconsistencies between internal data structures such as table headers, row identifiers, and secondary indexes. Allows running of database utilities from scripts. Defines AMPs, PEs, and hosts, and describes their interrelationships for Teradata Database. Note: Configuration is documented in Support Utilities. Control GDO Editor (ctl) Cufconfig Utility (cufconfig) Database Initialization Program (DIP) DBS Control (dbscontrol) Dump Unload/Load (DUL, DULTAPE) Ferret Utility (ferret) Displays the fields of the PDE Control Parameters GDO, and allows modification of the settings. Displays configuration settings for the user-defined function and external stored procedure subsystem, and allows these settings to be modified. Executes one or more of the standard DIP scripts packaged with Teradata Database. These scripts create a variety of database objects that can extend the functionality of Teradata Database with additional, optional features. Displays the DBS Control Record fields, and allows these settings to be modified. Saves system dump tables to tape, and restores system dump tables from tape. Defines the scope of an action, such as a range of tables or selected vprocs, displays the parameters and scope of the action, and performs the action, Support Utilities, Release

8 Chapter 1: Teradata Database Utilities Alphabetical Listing of Utilities Utility Filer Utility (filer) Purpose either moving data to reconfigure data blocks and cylinders, or displaying disk space and cylinder free space percent in use of the defined scope. Finds and corrects problems within the file system. Note: Filer is documented in Support Utilities. Gateway Control (gtwcontrol) Gateway Global (gtwglobal) Lock Display (lokdisp) Query Configuration (qryconfig) Query Session (qrysessn) Reconfiguration Utility (reconfig) Modifies default values in the fields of the Gateway Control Globally Distributed Object (GDO). Monitors and controls the Teradata Database workstation-connected users and their sessions. Displays a snapshot capture of all real-time database locks and their associated currently-running sessions. Reports the current Teradata Database configuration, including the Node, AMP, and PE identification and status. Monitors the state of selected Teradata Database sessions on selected logical host IDs. Uses the component definitions created by the Configuration Utility to establish an operational Teradata Database. Note: Reconfiguration is documented in Support Utilities. Reconfiguration Estimator (reconfig_estimator) Estimates the elapsed time for reconfiguration based upon the number and size of tables on the current system, and provides time estimates for the following phases: Redistribution Deletion NUSI building Note: Reconfiguration Estimator is documented in Support Utilities. Recovery Manager (rcvmanager) Show Locks (showlocks) Displays information used to monitor progress of a Teradata Database recovery. Displays locks placed by Archive and Recovery and Table Rebuild operations on databases and tables. 8 Support Utilities, Release 16.00

9 Chapter 1: Teradata Database Utilities For More Information Utility System Initializer (sysinit) Purpose For details Archive and Recovery, see Teradata Archive/Recovery Utility Reference. For details on Table Rebuild, see Utilities. Initializes Teradata Database. Creates or updates the DBS Control Record and other Globally Distributed Objects (GDOs), initializes or updates configuration maps, and sets hash function values in the DBS Control Record. Note: System Initializer is documented in Support Utilities. Table Rebuild (rebuild) Teradata Locale Definition Utility(tdlocaledef) Tpareset (tpareset) Update DBC (updatedbc) Update Space (updatespace) Vproc Manager (vprocmanager) Rebuilds tables that Teradata Database cannot automatically recover, including the primary or fallback portions of tables, entire tables, all tables in a database, or all tables in an Access Module Processor (AMP). Table Rebuild can be run interactively or as a background task. Converts a Specification for Data Formatting file (SDF) into an internal, binary format (a GDO) for use by Teradata Database. The SDF file is a text file that defines how Teradata Databaseformats numeric, date, time, and currency output. Resets the PDE and database components of Teradata Database. Recalculates the PermSpace and SpoolSpace values in the DBASE table for the user DBC, and the MaxPermSpace and MaxSpoolSpace values of the DATABASESPACE table for all databases based on the values in the DBASE table. Recalculates the permanent, temporary, or spool space used by a single database or by all databases in a system. Manages the virtual processors (vprocs). For example, obtains status of specified vprocs, initializes vprocs, forces a vproc to restart, and forces a Teradata Database restart. For More Information For more information on... starting the utilities See... the "Starting the Utilities" appendix. Support Utilities, Release

10 Chapter 1: Teradata Database Utilities For More Information For more information on... utilities related to the PDE (operating system) level of Teradata Database utilities related to Teradata Database security Archive and Recovery, FastExport, FastLoadd, MultiLoad, and TPump See... the "PDE Tools" appendix. Security Administration the following client utility books: Teradata Archive/Recovery Utility Reference Teradata FastExport Reference Teradata FastLoad Reference Teradata MultiLoad Reference Teradata Parallel Data Pump Reference 10 Support Utilities, Release 16.00

11 CHAPTER 2 Configuration Utility (config) Overview The Configuration and Reconfiguration utilities, config and reconfig, are used to define the AMPs and PEs that operate together as a Teradata Database. (The Reconfiguration utility is described in Reconfiguration Utility (reconfig).) Use Configuration to define the system: the interrelationships among the AMPs, AMP clusters (logical groupings of AMPs), PEs, and hosts that are connected as a Teradata Database system. (The Reconfiguration utility uses information from Configuration to configure the Teradata Database components into an operational system.) You must access a system console to use either utility. Use the Configuration commands described in this chapter to add, delete, list, modify, move, or show AMPs and PEs. Other steps in configuring a system involve the Parallel Database Extensions (PDE) portions of Teradata Database, and are configured with the PUT utility. For more information on PUT configuration steps, see the Parallel Upgrade Tool (PUT) Reference, available from Runs From The Configuration utility runs from Database Window or comparable interface to the Teradata Database console subsystem, such as cnsterm. For general information on starting the utilities from different interfaces, see Starting the Utilities Physical Processors The Configuration and Reconfiguration utilities are not responsible for the maintenance of the physical environment in which the Teradata Database configuration is defined. The AMPs and PEs exist within a previously defined physical configuration. Use the Parallel Upgrade Tool (PUT) to configure parts of the physical configuration, such as creating Logical Units (LUNs) on disk arrays. For more information, see Parallel Upgrade Tool (PUT) Reference. Configuration Maps Although the Configuration and Reconfiguration utilities are functionally independent, normally they are used together to change the contents of working areas, called configuration maps, in the node. A configuration map: Support Utilities, Release

12 Chapter 2: Configuration Utility (config) Hosts Stores the identification and status of each vproc in the Teradata Database Identifies the AMPs that constitute each AMP cluster Identifies each PE and its associated host The node contains these configuration maps: The current configuration map, which describes the current arrangement and status of vprocs in the system The new configuration map, which includes changes and additions to the configuration If you want to list or show information about or to delete a vproc, first you must have added it to the applicable (new or current) configuration map. If you want to add a vproc to the new configuration map, you must have defined the AMP or PE through the use of the PUT utility. These component types constitute a Teradata Database configuration: Hosts (or clients) PEs AMPs The following sections describe these components. For more information, see Parallel Upgrade Tool (PUT) Reference. Hosts You can attach more than one host simultaneously to the Teradata Database. The hosts are identified in the configuration by: A host number A logical host ID An associated host type designator You assign the host a number between 1 and The host type designator specifies the host data format. The system generates the16-bit logical host ID based on the assigned host number and the host type. The value is equal to the host number + the host type value. Connection Type IBM 370 (Mainframe) Communications Processor Motorola 68000, RISC (Solaris, HP) Honeywell 6000 (Mainframe) Host Type Designator Host Type Value IBM COP ATT3B BULLHN Logical Host ID Range (Host Number + Host Type Value) 12 Support Utilities, Release 16.00

13 Chapter 2: Configuration Utility (config) PEs and AMPs Connection Type Unisys OS1100 (Mainframe) Host Type Designator Host Type Value OS Logical Host ID Range (Host Number + Host Type Value) PEs and AMPs You can associate PEs defined in a Teradata Database configuration with one or more host computers or local area networks that also are defined in the configuration. Typically, AMPs in a configuration are related to other AMPs through cluster assignments. The placement of AMPs and PEs in a configuration is critical to the overall performance of the Teradata Database. System Configuration and Reconfiguration When Teradata Database is initialized, System Initializer (sysinit) procedures build a default configuration map. The default configuration depends on sysinit options specified by the user, and may include one AMP or all AMPs in the system. This configuration is stored in both the current and new configuration maps. When the Teradata Database is operational, Configuration describes the complete system in the new configuration map area. As the system grows and changes, use Configuration to revise the new configuration map to reflect these types of changes to the system: Addition and deletion of vprocs and hosts Changes to cluster assignments Changes to host assignments When changing cluster assignments without adding AMPs, make sure that ample disk space is available on all AMPs. If the system determines that ample space is not available, the system stops. To recover, perform a SYSINIT on the Teradata Database, which results in loss of all data. Teradata recommends that currentperm space should be less than 53% of the total maxperm space before starting a change of clusters without adding AMPs. After Configuration builds a new configuration map, the Reconfiguration utility redefines the system configuration according to the new map. Reconfiguration copies the new configuration map to the current configuration map. Message Types The application window running Configuration can contain the types of messages displayed in the output subwindow. Support Utilities, Release

14 Chapter 2: Configuration Utility (config) Message Types Message Information Prompt Error Description Indicates the status of a command or the result of an operation. OK indicates that a command has been accepted or an operation has completed successfully. Prompts for a response to a request or for confirmation of an action. Composed of a message code and text. To view all error messages issued by Configuration, see Messages. 14 Support Utilities, Release 16.00

15 Configuration Utility Command List The Configuration utility presents a command-line environment that allows the entry of Configuration commands. The following table lists these commands and functions based on activity. The commands are described in more detail in the sections that follow. Activity Command Function AMP operations ADD AMP Adds an AMP to the configuration map. Displaying system attributes DEFAULT CLUSTER Allows Configuration to assign AMPs to clusters when a new configuration is defined. DEL AMP MOD AMP MOVE AMP LIST LIST AMP LIST CLUSTER LIST HOST LIST PE SHOW CLUSTER SHOW VPROC SHOW HOST Deletes an AMP from the configuration map. Modifies the attributes of an AMP. Moves an AMP or group of AMPs to another AMP or group of AMPs. Lists the vprocs and hosts described in the specified configuration. Lists the attributes of every AMP in the specified configuration. Lists the attributes of every AMP in the specified configuration, ordered by cluster number. Lists all hosts in the specified configuration. Lists the attributes of every PE in the specified configuration. Shows all AMPs in a specified cluster. Shows the attributes of a specified vproc. Shows information about a specified host. Host operations ADD HOST Adds a host to the configuration map. DEL HOST MOD HOST Deletes a host from the configuration map. Modifies a host in the configuration map. PE operations ADD PE Adds a PE to the configuration map. DEL PE MOD PE MOVE PE Deletes a PE from the configuration map. Modifies the attributes of a PE in the configuration map. Moves a PE or group of PEs to another PE or group of PEs. Session control BEGIN CONFIG Begins a configuration session. END CONFIG STOP Ends a configuration session and stores the new configuration map. Stops Configuration. Note: You can access Help information on Configuration by pressing the F7 key. Chapter 2: Configuration Utility (config) Configuration Utility Command List Support Utilities, Release

16 Chapter 2: Configuration Utility (config) ADD AMP ADD AMP Purpose The ADD AMP command adds an AMP or a range of AMPs to the new configuration map. Syntax mmmlist CN = nnnn Usage Notes the range of vproc numbers that is being added to the configuration. AMP vprocs are numbered consecutively starting from 0 to a maximum number of the cluster number with which the AMP is to be associated. The cluster number ranges from 0 to You can add an AMP or a range of AMPs to the Teradata Database in a well-defined order. The AMP being added must have a vproc number equal to the number of AMPs currently entered in the new configuration map. That is, given a system with a new configuration map that describes n AMPs (whose vproc numbers are 0 through n-1), the only AMP that you can add to the system now is the AMP with vproc number n. Configuration verifies that the vproc number specified for the new AMP does not exist in the new configuration map. When the ADD AMP command is accepted, the specified AMP is added to the vproc list, and the system displays: The AMP is added. If you type the ADD AMP command before the BEGIN CONFIG command, the command is ignored, and the system displays: BEGIN CONFIG is not entered, and ADD AMP is ignored. Example: Add an AMP to a cluster To add an AMP as vproc eight in cluster three to the new configuration map, type: add amp 8, cn = 3 Example: Add a range of AMPs to a cluster To add AMPs from vproc four to vproc seven in cluster three to the new configuration map, type: add amps 4-7, cn = 3 16 Support Utilities, Release 16.00

17 Chapter 2: Configuration Utility (config) ADD AMP Note: The maximum number of AMPs that you can add to a cluster is 8. To assign AMPs to clusters automatically, use DEFAULT CLUSTER. Support Utilities, Release

18 Chapter 2: Configuration Utility (config) ADD HOST ADD HOST Purpose The ADD HOST command adds a host to the host list in the new configuration map. Syntax HN = nnnn HT = Usage Notes the host number to be added. Assign the host a number from 1 to the host type to be added. For more information on hosts and host numbers, see Hosts. This command creates the host group. When PEs are configured, they are assigned to a host number. This grouping provides a host number used to refer to all the PEs. Configuration checks for a valid host number and host type. You must type this command before ADD PE commands that specify PEs to be associated with this host. Example: Add a host to a configuration To add host four to the configuration, type: add host hn = 4, ht = ibm 18 Support Utilities, Release 16.00

19 ADD PE Chapter 2: Configuration Utility (config) ADD PE Purpose The ADD PE command adds a PE or a range of PEs to the new configuration map. Syntax mmmlist HN = nnnn Usage Notes the PE (or range of PEs) that is being added to the new configuration map. PE vproc numbers start at the high end of their range (30719 or 16383, depending on the system) and decrease consecutively. the host number with which the PE (or PEs) is to be associated. The host number ranges from 1 to Configuration validates the vproc number and host number. You must add the host number to the new configuration map (through the ADD HOST command) before the PE associated with the host can be described in the new configuration map. After the ADD PE command is accepted, the specified PE is added to the vproc list, and the system displays: The PE is added. If you type the ADD PE command before the BEGIN CONFIG command, the command is ignored, and the system displays: BEGIN CONFIG is not entered, and ADD PE is ignored. Example: Adding a PE to a host To add PE to host 10, type: add PE 30715, hn = 10 Example: Adding a range of PEs to a host To add PEs from vproc to vproc in host 282, type: add PE , hn = 282 Support Utilities, Release

20 Chapter 2: Configuration Utility (config) BEGIN CONFIG BEGIN CONFIG Purpose The BEGIN CONFIG command begins the configuration session. Changes caused by subsequent commands are recorded in the new configuration map. Syntax Usage Notes When the BEGIN CONFIG command is accepted, you can type commands to describe the new configuration map. The new configuration map is copied from main memory to disk when a subsequent END CONFIG command is executed. If you type the BEGIN CONFIG command when a configuration session is in progress, the utility prompts whether to end the configuration session that is in progress: The previous BEGIN CONFIG has not been ended. Do you want to abort it? -- Answer Y(es) or N(o): To abort the previous BEGIN CONFIG command, type Y. To continue the configuration session in progress, type N. The system displays: BEGIN CONFIG is ignored. When a configuration session ends prematurely, the new configuration map is not updated, and all changes are lost. A configuration session ends prematurely if one of these events occurs: The configuration session is aborted as described above. A STOP command is executed before an END CONFIG command. A Teradata Database restart occurs. 20 Support Utilities, Release 16.00

21 DEFAULT CLUSTER Chapter 2: Configuration Utility (config) DEFAULT CLUSTER Purpose The DEFAULT CLUSTER command indicates that Configuration is to assign AMPs to clusters automatically for the specified configuration. An AMP cluster is a collection of AMPs that are grouped together logically to provide fallback capability for each other for tables that are created with the FALLBACK option. Each AMP in a cluster contains a backup copy of a portion of the primary data for every other AMP in the cluster. Syntax c the default cluster size. The maximum value of c is 8. If you do not specify a default cluster size, the default size is 2. Usage Notes Typically, you type the DEFAULT CLUSTER command at the end of a session during which AMPs were added to the system. When the DEFAULT CLUSTER command is executed, AMPs are assigned to clusters automatically. In a system with n AMPs, each cluster contains c AMPs and n/c clusters. If nis not divisible by c, the last cluster will be no smaller than 50% less and not larger than 50% more than any other cluster. The number of clusters is adjusted accordingly. After the DEFAULT CLUSTER command is accepted, AMPs are assigned to clusters, and the system displays: The cluster assignment is complete. Execution of the DEFAULT CLUSTER command overrides all previous cluster assignments. AMP cluster assignments should be defined with respect to the hardware configuration. Usually, AMP failures result from hardware-related problems. AMP clustering assignments should be defined as closely as possible to the following fault-tolerant strategy: No two or more AMPs of a cluster reside in the same node cabinet. No two or more AMPs of a cluster are serviced by the same disk array cabinet. No two or more AMPs of a cluster are serviced by the same disk array controller. Examples: Specifying a default cluster size Typing DEFAULT CLUSTER 4 for a system of 15 AMPs results in a cluster size of 4. Three clusters have four AMPs each, and a fourth cluster has three AMPs. Typing DEFAULT CLUSTER 3 for a system of 16 AMPs results in cluster size of 3. Four clusters include 3 AMPs each, and 1 cluster includes 4 AMPs. Support Utilities, Release

22 Chapter 2: Configuration Utility (config) DEFAULT CLUSTER Typing DEFAULT CLUSTER 8 for a system of 16 AMPs results in a cluster size of 8. Two clusters include 8 AMPs each. 22 Support Utilities, Release 16.00

23 DEL AMP Chapter 2: Configuration Utility (config) DEL AMP Purpose The DEL AMP command specifies that an AMP or a range of AMPs is to be deleted from the new configuration map. Syntax mmmlist one or more vproc numbers to be removed from the configuration. AMP vprocs are numbered consecutively starting from 0 to a maximum number of Usage Notes If you delete an AMP, the AMP must have a vproc number equal to the highest number for any AMP currently entered in the new configuration map. That is, given a system with a new configuration map that describes n AMPs (whose vproc numbers are 0 through n-1), the only AMP that you can remove from that configuration map is the AMP with vproc number n-1. Configuration verifies that the vproc number to be deleted exists in the new configuration map. After the DEL AMP command is accepted, the specified AMP is deleted from the vproc list in the new configuration map, and the system displays: The AMP is deleted. If you type the DEL AMP command before the BEGIN CONFIG command, the command is ignored, and the system displays: BEGIN CONFIG is not entered, and DEL AMP is ignored. Examples: Deleting AMPs from a configuration To delete an AMP that is vproc seven of the new configuration, type: da 7 To delete AMPs that are vproc four to vproc six of the new configuration, type: da 4-6 Support Utilities, Release

24 Chapter 2: Configuration Utility (config) DEL HOST DEL HOST Purpose The DEL HOST command specifies that a host is to be deleted from the new configuration map. Syntax HN = nnnn Usage Notes the host number to be deleted. The host number ranges from 1 to Configuration verifies that a host with the specified number exists. After the command is accepted, the specified host is deleted from the new configuration map, and the system displays: The host is deleted from new configuration map. If you type the DEL HOST command before the BEGIN CONFIG command, the command is ignored, and the system displays: BEGIN CONFIG is not entered, and DEL HOST is ignored. You cannot delete a host from a configuration unless the host is no longer associated with any PE that is defined in the configuration. For information, see DEL PE. Example: Deleting a host from a configuration To delete host 52, type: dh hn = Support Utilities, Release 16.00

25 DEL PE Chapter 2: Configuration Utility (config) DEL PE Purpose The DEL PE command specifies that a PE or a range of PEs is to be deleted from the new configuration map Syntax mmmlist the PE (or PEs) to be deleted from the new configuration map. PE vproc numbers start at the high end of their range (30719 or 16383, depending on the system) and decrease consecutively. Usage Notes Configuration verifies that the vproc number (or numbers) to be deleted exists in the new configuration map. When all PEs that are associated with a host have been deleted from the new configuration map, you also must delete the host from the map. After the DEL PE command is accepted, the specified PEs are deleted from the vproc list in the new configuration map, and the system displays: The PE is deleted. If you type the DEL PE command before the BEGIN CONFIG command, the command is ignored, and the system displays: BEGIN CONFIG is not entered, and DEL PE is ignored. Examples: Deleting PEs from a configuration To delete PE 16380, type: del pe To delete PEs with vproc numbers to 16351, type: del pe Support Utilities, Release

26 Chapter 2: Configuration Utility (config) END CONFIG END CONFIG Purpose The END CONFIG command validates and updates the new configuration map and terminates a configuration session. Syntax Usage Notes When the END CONFIG command is executed, Configuration validates the new configuration map. As a result, one of the these events can occur: The session is terminated, the new configuration is accepted when the configuration changes are validated, and the new configuration map is updated on disk. The system displays: The session is terminated and the new configuration is stored. The Teradata Database is ready for reconfiguration. The END CONFIG command is ignored because one of the following occurred: A BEGIN CONFIG command was not entered, and, therefore, no configuration session is in progress. The Teradata Database has not completed startup. The configuration changes were not valid; an error message that describes the problem is displayed. When the END CONFIG command is not executed successfully, the system displays: The new configuration was not stored due to error(s) detected. Please update the configuration and try END CONFIG again. 26 Support Utilities, Release 16.00

27 LIST Chapter 2: Configuration Utility (config) LIST Purpose The LIST command lists the vprocs and hosts in the current or new configuration. Syntax SHORT LONG FROM CURRENT NEW the compact configuration map. This is the default. It displays the contents of the following commands: LIST CLUSTER LIST PE Usage Notes LIST HOST the detailed configuration map. that you will choose the type of configuration map to be displayed. the configuration map that the Teradata Database is using currently. This is the default. If you do not specify the FROM keyword, the default map type is CURRENT, and the current configuration map is displayed. the proposed configuration map. When you type the LIST command without any options, the system displays the short form of the default (current) configuration map. When you type the FROM option, you must specify the map type. No default type is supplied for this form. Example: Listing of the vprocs and hosts in a configuration An example output generated by the LIST command is shown below: AMP Array Support Utilities, Release

28 Chapter 2: Configuration Utility (config) LIST Vproc-Id Range Status Cluster Vproc-Id Range Status Cluster , 16, 32, 48 Online 0 1, 17, 33, 49 Online 1 2, 18, 34, 50 Online 2 3, 19, 35, 51 Online 3 4, 20, 36, 54 Online 4 5, 21, 37, 53 Online 5 14, 30, 46, 62 Online 14 15, 31, 49, 63 Online PEs ARRAY Vproc-Id Range Status Online Down Online Down Online Down Online HOSTs ARRAY HostNo Logical HostID Type Connected PE Range COP IBM IBM 16353, 16355, 16357, , 16363, 16365, , 16362, 16364, In response to the LIST command, Configuration lists all AMPs, PEs, and hosts that are included in the current configuration map. The following table lists the possible status of the vproc in the previous example. Status Online Down Hold Newready Description The vproc is participating in the current configuration. The vproc is not participating in the current configuration. The vproc was offline during the preceding system execution and is in the process of being recovered to the online state. The vproc has been initialized but has not been added to the configuration. 28 Support Utilities, Release 16.00

29 LIST AMP Chapter 2: Configuration Utility (config) LIST AMP Purpose The LIST AMP command displays the attributes of every AMP in the current or new configuration map. Syntax SHORT LONG FROM CURRENT NEW Usage Notes the compact configuration map of the AMP vprocs. This is the default. This option contains only two fields: vproc-id range status the detailed configuration map of the AMP vprocs. This option contains three fields: vproc status cluster that you will choose the type of configuration map to be displayed. the configuration map that the Teradata Database is using currently. This is the default. The default map type is CURRENT. If you do not specify the map type, the AMP configuration is displayed from the current map. the proposed configuration map. The LIST AMP command displays the AMP configuration map. Example: Compact listing of AMP attributes An example output generated by the LIST AMP command is shown below: AMPs ARRAY Vproc-Id Range Status Vproc-Id Range Status Support Utilities, Release

30 Chapter 2: Configuration Utility (config) LIST AMP 0-63 Online Note: As indicated in the previous example, all the AMP vprocs are online, so the vproc-id is blocked from the beginning vproc 0 to the ending vproc 63. If the system displays AMP vprocs that have status other than online, the system lists their vproc-ids and status, as shown in the following example. AMPs ARRAY Vproc-Id Range Status Vproc-Id Range Status 0 Online 1-2 Fatal Online 15 Fatal Online 63 NewReady Example: Detailed listing of AMP attributes An example output generated by the LIST AMP LONG command is shown below: AMPs ARRAY Vproc Status Cluster Vproc Status Cluster 0 Online 0 1 Fatal 1 2 Fatal 2 3 Online 3 4 Online 4 5 Online Online Online Support Utilities, Release 16.00

31 LIST CLUSTER Chapter 2: Configuration Utility (config) LIST CLUSTER Purpose The LIST CLUSTER command displays the attributes of every AMP in the current or new configuration map, ordered by cluster number. Syntax SHORT LONG FROM CURRENT NEW Usage Notes the compact display of the attributes of every AMP in the current or new configuration map, ordered by cluster number. This option contains three fields: vproc-id range status CN This is the default. the detailed display of the attributes of every AMP in the current or new configuration map, ordered by cluster number. This option contains three fields: vproc status cluster that you will choose the type of configuration map to be displayed. the configuration map that the Teradata Database is using currently. This is the default. If you do not specify a map type, the current configuration map is displayed. the proposed configuration map. The LIST CLUSTER command displays the cluster configuration map. This map shows a list of AMPs ordered by cluster number. Support Utilities, Release

32 Chapter 2: Configuration Utility (config) LIST CLUSTER Example: List AMP configuration attributes, ordered by cluster number Example output generated by the LIST CLUSTER command is shown below: AMPs ARRAY Vproc-Id Range Status CN Vproc-Id Range Status Cluster 0, 16, 32, 48 Online 0 1, 17, 33, 49 Online 1 2, 18, 34, 50 Online 2 3, 19, 35, 51 Online 3 4, 20, 36, 54 Online 4 5, 21, 37, 53 Online 5 14, 30, 46, 62 Online 14 15, 31, 49, 63 Online 15 AMPs ARRAY Vproc Status Cluster Vproc Status Cluster 0 Online 0 1 Online 1 2 Online 2 3 Online 3 4 Online 4 5 Online 5 62 Online Online Support Utilities, Release 16.00

33 LIST HOST Chapter 2: Configuration Utility (config) LIST HOST Purpose The LIST HOST command displays all hosts in a current or new configuration. Syntax SHORT LONG FROM CURRENT NEW Usage Notes the compact display of all hosts in a current or new configuration map. This is the default. This option contains these fields: HostNo Logical HostID Type Connected PE range the detailed display of all hosts in a current or new configuration map. This option contains these fields: HostNo Logical HostID Type Connected PE that you will choose the type of configuration map to be displayed. the configuration map that the Teradata Database is using currently. This is the default. If you do not specify the map type, the current configuration map is displayed. the proposed configuration map. The LIST HOST command displays the host configuration map. Example: Displaying all hosts in a configuration An example output generated by the LIST HOST command is shown below: HOSTs ARRAY Support Utilities, Release

34 Chapter 2: Configuration Utility (config) LIST HOST HostNo LogicalHostID Type Connected PE Range COP IBM IBM HOSTs ARRAY HostNo LogicalHostID Type Connected PE Range COP IBM IBM For information about the ranges of logical host IDs for all supported hosts, see Hosts. 34 Support Utilities, Release 16.00

35 LIST PE Chapter 2: Configuration Utility (config) LIST PE Purpose The LIST PE command displays the attributes of every PE in the current or new configuration map. Syntax SHORT LONG FROM CURRENT NEW Usage Notes the compact configuration map of the PE vprocs. This is the default. This option contains only two fields: vproc-id range status the detailed configuration map of the PE vprocs. This option contains three fields: vproc status hostno that you will choose the type of configuration map to be displayed. the configuration map that the Teradata Database is using currently. This is the default. If you do not specify the map type, the current PE configuration map is displayed. the proposed configuration map. The LIST PE command displays the PE configuration map. Example: Compact listing of PE attributes An example output generated by the LIST PE command is shown below. PEs ARRAY Vproc-Id Range Status Vproc-Id Range Status Online Support Utilities, Release

36 Chapter 2: Configuration Utility (config) LIST PE Note: As indicated in the previous example, all the PE vprocs are online, so the vproc-id is blocked from the beginning vproc to the ending vproc If the system displays PE vprocs that have status other than online, the system lists their vproc-id and status, as shown in the following example. PEs ARRAY Vproc-Id Range Status Vproc-ID Range Status Online Down Online Down Online Down Online Example: Detailed listing of PE attributes An example output generated by the LIST PE LONG command is shown below. PEs ARRAY Vproc Status HostNo Vproc Status HostNo Online Online Online Online Online Online Online Online Support Utilities, Release 16.00

37 MOD AMP Chapter 2: Configuration Utility (config) MOD AMP Purpose The MOD AMP command specifies that an AMP or a range of AMPs is to be moved from one cluster to another. Syntax mmmlist TO CN = nnnn Usage Notes one or more vproc numbers to be moved from one cluster to another. AMP vprocs are numbered consecutively starting from 0 to a maximum number of the new cluster number entry. new cluster number with which the AMP or AMPs are to be associated. The cluster number ranges from 0 to If you do not specify a cluster number, it remains unchanged. Configuration verifies the vproc number(s) and cluster number. After the MOD AMP command is accepted, the attributes of the specified AMP are modified, and the system displays: The AMPs attribute is modified in the new configuration. If you type the MOD AMP command before the BEGIN CONFIG command, the command is ignored, and the system displays: BEGIN CONFIG is not entered, and MOD AMP is ignored. Example: Changing the cluster assignment of a range of AMPs To modify the cluster assignment of AMPs four through eight to be part of cluster one, type: ma 4-8 to cn = 1 Support Utilities, Release

38 Chapter 2: Configuration Utility (config) MOD HOST MOD HOST Purpose The MOD HOST command changes the host type in the new configuration map. Syntax HN = nnnn HT = Usage Notes the host number whose type is to be changed. The host number ranges from 1 to the host type. After the MOD HOST command is accepted, the type of the specified host is modified in the new configuration map, and the system displays: The host is changed in the new configuration map. If you type the MOD HOST command before the BEGIN CONFIG command, the command is ignored, and the system displays: BEGIN CONFIG is not entered, and MOD HOST is ignored. Example: Changing the host type in a configuration To change the host type for host 52 to cop (communications processor), type: mh hn = 52, ht = cop 38 Support Utilities, Release 16.00

39 MOD PE Chapter 2: Configuration Utility (config) MOD PE Purpose The MOD PE command specifies that a PE or a range of PEs is to be moved from one host to another. Syntax mmmlist HN = nnnn Usage Notes one or more vproc numbers to be assigned to a different host. PE vproc numbers start at the high end of their range (30719 or 16383, depending on the system) and decrease consecutively. the new host number with which the PE (or PEs) is to be associated. The host number ranges from 1 to Configuration validates the vproc number (or numbers) assigned to the PE (or PEs) and the host number with which the PE (or PEs) is to be associated. After the command is accepted, the specified PE (or PEs) is modified in the vproc list in the new configuration map, and the system displays: The PE attributes are modified in the new configuration map. If you type the MOD PE command before the BEGIN CONFIG command, the command is ignored, and the system displays: BEGIN CONFIG is not entered, and MOD PE is ignored. Example: Changing the host assignment of a range of PEs To move PEs with vproc numbers through to host 286, type: mod PE to hn = 286 Support Utilities, Release

40 Chapter 2: Configuration Utility (config) MOVE AMP MOVE AMP Purpose The MOVE AMP command specifies that all data rows from an AMP or a group of AMPs are to be moved to another AMP or group of AMPs. Syntax mmmlist nnnlist the range of AMP vproc numbers being moved in the configuration. the range of AMP vproc numbers that mmmlist is being moved to in the configuration. AMP vprocs are numbered consecutively starting from 0 to a maximum number of Usage Notes The number of AMPs specified in mmmlist and in nnnlist must be the same. The following applies. AMPs in the mmmlist nnnlist Must be AMPs defined in the current configuration. new AMPs only. After the MOVE AMP command is accepted, Configuration displays: The AMP range is moved. If you type the MOVE AMP command before the BEGIN CONFIG command, the command is ignored. Configuration displays: BEGIN CONFIG is not entered, and MOVE AMP is ignored. If you type the MOVE AMP command and the ADD AMP command, and if the MOVE AMP range of vproc numbers is before the ADD AMP range of vproc numbers, Configuration displays: Moved AMPS must follow added AMPs. If you type the MOVE AMP command and either the MODIFY AMP command or the DEL AMP command, the commands are ignored. Configuration displays: Addition, Deletion, and Moving of AMPS are mutually exclusive. 40 Support Utilities, Release 16.00

41 Note: Run the Reconfig utility to do the following: Redistribute all rows from the moved AMPs Change the configuration so that the moved AMPs are properly associated with the correct nodes and/or disk arrays For more information on MOVE AMP, see MOVE AMP Operation. Examples: Moving data from existing AMPs to new AMPs To move the data on AMP 0 in an existing eight-amp system to a new AMP 8, type: Chapter 2: Configuration Utility (config) MOVE AMP move amp 0 to 8 To move the data from AMPs 4 and 5 in an existing eight-amp system to new AMPs 8 and 9, type: move amp 4-5 to 8-9 Support Utilities, Release

42 Chapter 2: Configuration Utility (config) MOVE PE MOVE PE Purpose The MOVE PE command specifies that a PE or group of PEs is to be moved to another PE or group of PEs. Syntax mmmlist nnnlist the range of PE vproc numbers being moved in the configuration. the range of PE vproc numbers that mmmlist is being moved to in the configuration. PE vproc numbers start at the high end of their range (30719 or 16383, depending on the system) and decrease consecutively. Usage Notes The number of PEs specified in mmmlist and in nnnlist must be the same. The following applies. PEs in the mmmlist nnnlist Must be PEs defined in the current configuration. new PEs only. Note: You also can move PEs that are restricted to specific nodes, such as mainframe-connected PEs, to other nodes. After the MOVE PE command is accepted, Configuration displays: The PE range is moved. If you type the MOVE PE command before the BEGIN CONFIG command, the command is ignored. Configuration displays: BEGIN CONFIG is not entered, and MOVE PE is ignored. Note: Run the Reconfig utility to change the configuration so that the moved PEs are properly associated with the correct nodes and/or disk arrays. Examples: Moving existing PEs to new PEs To move PE in an existing eight-pe system (PEs ) to a new PE (PE 16375), type: move pe to Support Utilities, Release 16.00

43 Chapter 2: Configuration Utility (config) MOVE PE To move PEs in an existing eight-pe system (PEs ) to new PEs (PEs ), type: move pe to Support Utilities, Release

44 Chapter 2: Configuration Utility (config) SHOW CLUSTER SHOW CLUSTER Purpose The SHOW CLUSTER command displays all AMPs in a specified cluster for a current or new configuration map. Syntax CN = nnnn the cluster number for which the appropriate AMPs are to be displayed. The cluster number ranges from 0 to FROM that you will choose the type of configuration map to be displayed. CURRENT the configuration map that the Teradata Database is using currently. This is the default. If you do not specify a map type, the current cluster configuration map is displayed. NEW the proposed configuration map. Usage Notes The SHOW CLUSTER command displays the cluster configuration map. Example: To display all AMPs in a specified cluster To display the current configuration map for cluster three and its associated AMPs, type: show cluster cn = 3 An example output generated by the SHOW CLUSTER command is shown below. OK, AMPs of cluster 3 are listed. Vproc Status Cluster Vproc Status Cluster 3 Online 3 7 Online 3 11 Online 3 15 Online 3 44 Support Utilities, Release 16.00

45 SHOW HOST Chapter 2: Configuration Utility (config) SHOW HOST Purpose The SHOW HOST command shows information for a specified host. Syntax HN = nnnn FROM CURRENT NEW Usage Notes the host number. The host number ranges from 1 to that you will choose the type of configuration map to be displayed. the configuration map that the Teradata Database is using currently. This is the default. If you do not specify a map type, the current configuration map is displayed. the proposed configuration map. If you use FROM, you must specify a map type. The SHOW HOST command displays the host number, logical host ID, host type, and vproc numbers of connected PEs. Example: Displaying information about a specified host To display information from the current configuration map about host 10, type: show host hn = 10 An example output generated by the SHOW HOST command is shown below: HOSTs ARRAY HostNo Logical HostID Type Connected PE IBM IBM IBM Support Utilities, Release

46 Chapter 2: Configuration Utility (config) SHOW VPROC SHOW VPROC Purpose The SHOW VPROC command displays the attributes of a specified AMP or PE vproc in a current or new configuration. Syntax mmmm the vproc number of an AMP or PE. Note: FROM CURRENT NEW Usage Notes AMP vproc number ranges from 0 to PE vproc numbers start at the high end of their range (30719 or 16383, depending on the system) and decrease consecutively. that you will choose the type of configuration map to be displayed. the configuration map that the Teradata Database is using currently. This is the default. The default map type is CURRENT. If you do not specify a map type, the current configuration map is displayed. the proposed configuration map. The following table lists what the SHOW VPROC command displays. Type of vproc Information displayed PE Vproc number Status Host number with which the PE is associated AMP Vproc number Status Cluster number 46 Support Utilities, Release 16.00

47 Examples: Displaying attributes of specified vprocs To display the attributes of PE vproc for the current configuration map, type: Chapter 2: Configuration Utility (config) SHOW VPROC show vproc An example output generated by the SHOW VPROC command is shown below: Vproc Status HostNo Online 821 To display the attributes of AMP vproc six for the current configuration map, type: show vproc 6 An example output generated by the SHOW VPROC command is shown below: Vproc Status Cluster Online 2 Support Utilities, Release

48 Chapter 2: Configuration Utility (config) STOP STOP Purpose The STOP command stops Configuration. Syntax Usage Notes When you type the STOP command, all AMP tasks created for Configuration are aborted, and all storage allocated for Configuration is released. Typing this command does not automatically update the new configuration map for the current session. The updating of the new configuration map is accomplished by the END CONFIG command. The STOP command displays: Config is about to be stopped. Notice: The STOP command does not warn you if the END CONFIG command has not been executed. Be sure to type END CONFIG before using the STOP command. 48 Support Utilities, Release 16.00

49 Configuration Utility Examples This section shows the process of adding vprocs to a Teradata Database. Chapter 2: Configuration Utility (config) Configuration Utility Examples Current configuration New configuration Four AMPS: 0, 1, 2, 3 Eight AMPS: 0, 1, 2, 3, 4, 5, 6, 7 Two PEs: 16383, Four PEs: 16383, 16382, 16381, Cluster 0: (0, 1, 2, 3) Cluster 0: 0, 1, 4, 5 Cluster 1: 2, 3, 6, 7 Activities Performed The configuration procedure described below performs three activities: Adds AMPs Adds PEs Changes cluster assignments For this demonstration, do the following in the Database window: 1. In the Database Window, select the Supervisor (Supvr) icon. The Supervisor window appears. Note: The PDE must be up and the Supvr window must display the status as Reading before you can enter commands in the command input line. 2. In the Enter a command subwindow of the Supervisor window, type the following and press Enter: start config The Supervisor window displays: Started config in window 1. The window number represents the application window in which Configuration is running. The Configuration window appears. 3. To begin the session, type the following in the Configuration window and press Enter: begin config 4. To add AMPs numbered four through seven, type: add amp 4 add amp 5 add amp 6 add amp 7 5. To add PEs numbered and to host number one, type: add pe 16381, hn = 1 Support Utilities, Release

50 Chapter 2: Configuration Utility (config) Configuration Utility Examples add pe 16380, hn = 1 6. To assign AMPs to clusters automatically, type: default cluster 4 7. To end the session, type: end config 8. To verify the new configuration, type: list from new 9. To stop configuration, type: stop You used Configuration to define the new system. Run the Reconfiguration utility next to configure the defined Teradata Database components into an operational system. For an example of how to run the Reconfiguration utility following the use of Configuration, see Reconfiguration Utility (reconfig). 50 Support Utilities, Release 16.00

51 CHAPTER 3 Filer Utility (filer) Overview The Filer utility, filer, is a file system utility that enables trained personnel to display or modify data structures related to the Teradata Database File System, including the Write-Ahead Logging (WAL) log. Use Filer to find and correct problems within the file system and the WAL log. : Filer works at a low level of the Teradata Database File System. Commands that manipulate File System structures are potentially dangerous to the operation of your system if used improperly. To use Filer, you should have a thorough understanding of the Teradata Database File System. Runs From Filer runs from Database Window or comparable interface to the Teradata Database console subsystem, such as cnsterm. For general information on starting the utilities from different interfaces, see Starting the Utilities The Teradata Database File System The Teradata Database File System is not a general-purpose file system. It helps isolate the Teradata Database from hardware platform dependencies, and supports the creation and maintenance of database tables under the direction of Teradata Database. The file system stores data in physical rows. A physical row is a general data structure that consists of a row header (metadata that includes a unique row ID) and stored data. Each physical row stores one of several kinds of data. For example: A table row A series of one or more column partition values A table header (metadata for a table; every table has a header) Index data structures This list is not comprehensive; physical rows also store other types of data used by Teradata Database. To the file system, the kind of data stored in the physical row is immaterial. The interpretation and differentiation is made by the higher-level database software that requests and receives the data from the file system. In this chapter, the term row generally refers to a physical row in the file system, irrespective of the kind of data the physical row stores. Support Utilities, Release

52 Chapter 3: Filer Utility (filer) Write Ahead Logging (WAL) In the Teradata Database File System, a data block (DB) is a disk-resident structure that contains one or more physical rows from the same table. Any single physical row is fully contained within a DB, and every DB is fully contained within a cylinder. A cylinder is a unit of contiguous physical memory that is allocated as a single unit. At the beginning of each cylinder is a cylinder index (CI) that identifies the data blocks (DBs) the cylinder contains. The Master Index (MI) is a per-amp global structure that is the top-level file system internal index structure. It is periodically written to disk as a snapshot to speed up system startup times. If the snapshot copy of the MI has problems, the system can reconstruct the MI by inspecting all CIs allocated to an AMP, however, this takes significantly more time than reading the MI from a snapshot. For more information about data blocks, cylinders, and the physical row layouts for different data structures, see Database Design. Write Ahead Logging (WAL) WAL is a log-based file system recovery scheme in which modifications to permanent data are written to a log file, the WAL log. The log file contains change records (Redo records) which represent the updates. At key moments, such as transaction commit, the WAL log is forced to disk. In the case of a reset or crash, Redo records can be used to transform the old copy of a permanent data block on disk into the version that existed in memory at the time of the reset. By maintaining the WAL log, the permanent data blocks that were modified no longer have to be written to disk as each block is modified. Only the Redo records in the WAL log must be written to disk. This allows a write cache of permanent data blocks to be maintained. WAL protects all permanent tables and all system tables but is not used to protect the Transient Journal (TJ), since TJ records are stored in the WAL log. WAL also is not used to protect spool or volatile tables. The WAL log is maintained as a separate logical file system from the normal table area. Whole cylinders are allocated to the WAL log, and it has its own index structure. The WAL log data is a sequence of WAL log records and includes the following: Redo records, used for updating disk blocks and insuring file system consistency during restarts. TJ records used for transaction rollback. The WAL log is subject to many of the same problems as the normal file system, such as data corruption from hardware, software, and operational problems. The Filer utility provides special diagnostic startup options and commands that can display and modify the WAL log and its index, and can be used by support personnel to repair such problems. Teradata Database File System Startup File system startup is part of the normal sequence of events that are part of Teradata Database system startup sequence: 1. File system startup 2. Start the AMPs 3. Start the PEs 4. Voting for Transaction Recovery 5. Transaction Rollback 52 Support Utilities, Release 16.00

53 A normal startup of the Teradata Database file system starts the WAL log prior to starting the normal file system. The key phases of Teradata file system startup include the following: Read the snapshot MI from disk and examine it for structural consistency. If it is consistent, it can be used, avoiding the necessity of reading CIs and constructing the MI, which takes longer. Build the WAL Master Index (WMI). This is the index structure for the WAL log. The WAL log contains the Redo records and the Transient Journal (TJ) records. Perform Redo analysis and replay. This scans the WAL log for Redo records and analyzes which records should be applied, and replays the necessary modifications, to bring the CI structures up-to-date. Because the WAL protocol allows for the delayed writing of modified blocks to their home disk addresses, the blocks on disk are only completely valid after the Redo processing is complete. When the processing is finished, the CIs have been updated to reflect any cylinder migrates or other modifications that were performed. If the MI snapshot read from disk was valid, redo processing updates the MI. If the snapshot or in-memory MI was inconsistent or corrupt, REDO processing ignores the MI and the system rebuilds the MI from an inspection of CIs on the AMP. The specific File System startup sequence on each AMP is as follows: 1. Get the list of cylinders for this AMP from Teradata Virtual Storage (VS). 2. Validate Segment Zero (SEG0), the File Information Block (FIB), and MI. 3. Get the FSYS Trace segment. 4. Process the DEPOT blocks. 5. Process the WAL log: a. Build the WMI. b. Perform Redo analysis. c. Perform Redo replay. 6. Start the File System. a. Clean up cylinder index descriptors (CIDs) with special flags. b. Rebuild the free sector list. c. Clean up pending migrate flags. d. Validate cylinders in FSYS structures against Teradata VS list. 7. Start File System background tasks. Chapter 3: Filer Utility (filer) Teradata Database File System Startup When a system is corrupted, some or all of the above steps may be difficult to complete. The Filer diagnostic options allow trained personnel to start Filer even when errors are encountered during file system startup. The options also provide the ability to control which phases of the file system startup sequence are executed. This allows users to skip the startup phases which have errors and to use Filer to troubleshoot and repair the problems. Errors encountered during WAL log startup or during Redo processing will generally abort the startup. You can use the Filer diagnostic options to analyze the problems and possibly recover from the errors. Support Utilities, Release

54 Chapter 3: Filer Utility (filer) Starting Filer Starting Filer Standard Filer Startup Syntax Start Filer from the Supervisor window of Database Window (DBW). Use the DBW START command. For more information on the options available with the start command, see the Database Window (xdbw) chapter in Utilities. Syntax Element Description -l Starts Filer with file system locking disabled. Use this option to avoid hang or deadlock situations caused by lock conflicts. Because the file system locking protocols are not observed, Filer may see inconsistent information. This option should be used only when Teradata Database is in the Debug-Stop state. Use of the -l option at other times can cause a database reset. -i Same as the -l option. -v Starts Filer on a single vproc. This is known as running Filer in vprocspecific mode or local mode. If the -v vproc option of the DBW START command is specified together with the -v option of Filer, Filer is started on the vproc identified by vproc, otherwise, Filer is started on the control vproc, normally the AMP that is running with the lowest vproc number. For more information on DBW commands and options, see the Database Window (xdbw) chapter in Utilities. Note: The SCOPE and VPROC commands are not allowed when Filer is run in vproc-specific mode. For more information, see SCOPE (or VPROC). -m [percent] Causes the Teradata Database file system to allocate a percentage of available memory for Filer to use for output buffers. This can help prevent Filer from running out of buffer space when running commands that generate large amounts of output. percent specifies the percentage of memory per AMP that the file system allocates for Filer. The valid range is from 1 through 75%. If -m is used without a percent specification, 25% of available memory is allocated. -h Displays the online help for Filer. 54 Support Utilities, Release 16.00

55 Chapter 3: Filer Utility (filer) Starting Filer Syntax Element Description? Displays the online help for Filer. Usage Notes Teradata Database and the Teradata Database File System need not be running when Filer is started. Notice: Do not leave Filer running unattended because locks on segments (if any exist) could block other tasks. Either quit Filer, or use the DROP command to release these locks. Running Filer from the System Debugger and Coroner The System Debugger and Coroner are used by qualified personnel to analyze and repair serious system problems. Filer can be run from within System Debugger and Coroner. When run from System Debugger, Filer does not observe file system locking protocols. Because these protocols are normally used to avoid hang or deadlock situations, and to prevent access to file system segments that are in use by other processes, improper use of Filer from the System Debugger can result in file system corruption. When run from Coroner, Filer presents a subset of the normal Filer commands that are suitable to working with crashdump data. The Filer HELP command displays the Filer commands that are available in the current context within which Filer is running. Diagnostic Filer Startup Syntax Note: Filer diagnostic options should be used only by, or under the direction of, trained Teradata Support personnel. The Filer diagnostic options are used to repair a corrupted file system or WAL log. Users of the diagnostic options should have thorough understanding of the Teradata file system internal structures, operation, and startup sequence. To use the Filer diagnostic startup options, include them on the command line when starting Filer from the Supervisor window of Database Window. Support Utilities, Release

56 Chapter 3: Filer Utility (filer) Starting Filer The diagnostic options are described below. For an explanation of the non-diagnostic Filer startup options see Standard Startup Syntax. Syntax Element Description -u Does not build the WMI, does not perform any Redo processing, but loads the MI snapshot from disk, or, if the snapshot is invalid, builds the MI. If a problem exists such that the WAL log cannot be built successfully, first start Filer with the -x option. (If Filer does not present a command prompt, try restarting Filer using both the -x and -w options.) Repair the log, and restart Filer to perform Redo processing. If Filer does not present a command prompt when started with the options above, restart Filer using the -u option to skip building the WAL log and performing Redo operations. Examine and repair the problem WAL cylinder indexes (WCIs) and WAL data blocks (WDBs), and restart Filer again to perform Redo processing. If the WAL log is beyond repair, use the -u option again and run SCANDISK to determine the health of the file system before any decision is made on how to proceed next (such as deleting the complete WAL log). -w Executes the normal file system startup process, but ignores most errors. Redo processing continues even if errors are encountered during the Redo phase. Always builds the MI from reading CIs, rather than reading the MI from the disk snapshot copy. 56 Support Utilities, Release 16.00

57 Chapter 3: Filer Utility (filer) Starting Filer Syntax Element Description If the -w option is used with another diagnostic option, the -r option need be specified only once. For example: start filer -w -y -r 2 Use this option to clean up the WAL log and the file system as much as possible. This option skips certain file system startup errors, such as software errors related to the Redo process and its prerequisite conditions. Read errors and other hardware-related errors are not skipped. The Redo process attempts to process as much of the WAL log as possible, and then the rest of the normal file system startup logic is executed. It is highly recommended to run SCANDISK after the Redo process to determine the structural integrity of the file system. See SCANDISK. In cases where errors are encountered before the Redo phase, (for example, while building the WAL log), you can use the -w option together with another diagnostic option such as -z to skip both the errors and the Redo phase. This allows you to start Filer and possibly repair the WAL log so that subsequent file system startup and Redo processing attempts will succeed. A log file is created for each vproc where Filer -w is invoked. You can check these files for errors and warnings. See Filer -w Error Log. -x Builds the WAL Master Index (WMI), but does not perform Redo processing, and does not read the MI snapshot from disk or build the MI. Use this option to examine and repair a corrupted WAL log. The option is useful when errors are encountered during Redo processing or when building the MI. The option provides the ability to examine and repair the WMI. It allows you to inspect the WAL log without trying to process the log or build the file system. It is highly recommended to run SCANDISK WAL to determine the structural integrity of the WAL index structure. See SCANDISK. Once repairs are completed, you can run Filer again to process the Redo records. -y Builds the WMI and performs Redo processing, and does not read the MI snapshot from disk or build the MI. Use this option when the file system index structure is corrupt, and building the MI causes Filer to abort. This option gives you the ability to examine and repair the index structure. Once repairs are completed, run Filer again to build the MI. -z Builds the WMI, does not perform Redo processing, but loads the MI snapshot from disk, or, if the snapshot is invalid, builds the MI. Use this option when errors are encountered during Redo processing, and the WAL log may be damaged beyond repair. Support Utilities, Release

58 Chapter 3: Filer Utility (filer) Starting Filer Syntax Element Description You can examine the file system to determine whether it can be salvaged. You can determine whether it is possible to repair the WAL log sufficiently so that Redo processing using the -w option can complete, or to ignore the WAL log and just repair the file system index structure. It is highly recommended to run SCANDISK WAL to determine the structural integrity of the WAL log and the file system. See SCANDISK. -n Does not build the WMI, does not perform Redo processing, and does not read the MI snapshot from disk or build the MI. Use this option when serious problems exist in the file system which cause file system startup to fail in the early stages. This option executes as little file system startup logic as possible to bypass errors so that Filer can be started. You can examine the file system to determine whether it can be salvaged. Since key global file system structures are not rebuilt in this mode, some commands, including a full SCANDISK, will be disabled. -r vprocs Starts Filer in the -w, -u, -x, -y, -z, or -n diagnostic mode on the specified vprocs. The -r option must be used in conjunction with the other diagnostic options to limit the scope of these diagnostic modes. vprocs can be one of the following: A single vproc number A comma-separated list of vproc numbers A range of vproc numbers in the form of start:end. For example, 2:5 represents vprocs 2, 3, 4, and 5. A combination of vproc numbers and ranges of vproc numbers separated by commas The syntax for vprocs is as follows: Filer will start using the specified diagnostic option on the vprocs specified by the vprocs argument. Filer will start normally on the other vprocs. Note: If Filer is started with the -v option to run Filer only on a single vproc, the number of the vproc where Filer is running must be included in the -r vprocs specification. -t Causes Filer invocation to create diagnostic files (dbgfil* and dbgftr*), which can be used to debug problems related to the Teradata Database File System. 58 Support Utilities, Release 16.00

59 Diagnostic Option Usage Notes Running Filer with the diagnostic options is only allowed on vprocs that are in a fatal state or where the Teradata Database is not running. The -r option must be used with the diagnostic options to specify the AMPs on which Filer should run diagnostically. On other AMPs, Filer will run normally. The -u, -x, -y, -z, and -n options are mutually exclusive; however, the -w option can be used in combination with the -u, -x, -y, -z, or -n options. Note the distinction between the -v and the -r vprocs options of Filer: -v runs Filer on a single AMP vproc. To specify the vproc on which Filer runs, the -v option of Filer is used together with the -v vproc option of the Database Window START command. If the START command -v vproc option is not specified, Filer runs on the control vproc of the node on which Filer was invoked. -r vprocs runs Filer on all AMP vprocs in the system, but runs it in a diagnostic mode on the AMP vprocs specified by vprocs. The -r vprocs option must be used in conjunction with one or more of the other diagnostic Filer options. Filer runs its normal, non-diagnostic mode on the other AMPs in the system, which were not specified by -r vprocs. When using any of the diagnostic options, SCANDISK should be run at the level that is available to the option. Initial results from SCANDISK will indicate whether the file system structures are coherent. If repairs are made based on the SCANDISK result, SCANDISK should be run again, either immediately after the repairs, or after quitting and restarting Filer with the appropriate options. This will insure that the repairs resulted in a structurally correct file system. See SCANDISK. Some Filer commands are not available with certain diagnostic modes because these commands depend on specific file system structures which may not be built when Filer is started with those diagnostic options. The following table shows the commands that are disabled for the various diagnostic options. Commands not listed in the table are valid in all modes of Filer. -u -w -x -y -z -n CID Disabled Disabled Disabled FIND Disabled Disabled Disabled MI Disabled Disabled Disabled MODIFY Disabled ROW Disabled Disabled Disabled SCANDISK (Full) Disabled Disabled Disabled Disabled SCANDISK WAL Disabled Disabled SCANDISK WMI Disabled TABLE Disabled Disabled Disabled WABC Disabled WAL Disabled Disabled WCID Disabled Disabled WFIND Disabled WFLUSH Disabled Disabled Chapter 3: Filer Utility (filer) Starting Filer Support Utilities, Release

60 Chapter 3: Filer Utility (filer) Starting Filer -u -w -x -y -z -n WMI Disabled Disabled WREC Disabled Disabled WSUMMARY Disabled The diagnostic options have no direct effect on the Transient Journal (TJ). However, if manual repairs are made to the WAL log, the TJ records might also be affected. When Filer is started with any of the diagnostic options, an informational message will be appended to the streams log(s) specifying which vprocs were affected. Filer -w Error Log When Filer is started with the -w option, a separate log file is created for each vproc where Filer -w is invoked. By default, the log is created in /var/opt/teradata/tdtemp. The location and name of the file is displayed on the Filer screen. The file logs any warnings and errors encountered during Filer startup on that vproc. After Filer is started, users should check the log files for errors or warnings, and take the appropriate action to ensure that the file system structures are consistent. The log file is useful for diagnosing WAL log problems, and identifying tables that may need repair. Example: Starting Filer in -w diagnostic mode on one vproc The following example shows the Filer screen when Filer is invoked with the -w option. The output screen displays the location and names of the log files created. To start Filer -w diagnostic mode on vproc 1, and have Filer run in normal mode on other vprocs in the system, type the following into the Supervisor window of Database Window: start filer -w -r 1 Filer shows the following output in the Filer Application window of Database Window, and subsequent Filer commands are entered into this window: Filer will be started with the -w option on the following VPROCs VPROC(s) (0001) Filer will be started normally on all other VPROCs WARNING: The Filer option that you've entered allows you to run Filer in a special diagnostic mode. Running in this mode can cause serious damage to your data if you do not understand both the internals of the file system and the basic startup sequence. Only trained field support personnel should attempt to use these options. Do you wish to continue?? (Y/N) y File System Debug Utility Filer 60 Support Utilities, Release 16.00

61 Waiting for Filer Slave tasks to start Command has been sent to Slave tasks. Chapter 3: Filer Utility (filer) Filer Command Syntax vproc 1 (0001) response Filer -w started on VPROC 1. Startup information and errors will be logged in the file /var/opt/teradata/tdtemp/filer.w txt. Mon Apr 07, :19:18 4 of 4 amps have been selected Filer ==> Example: Starting Filer in -w diagnostic mode on three vprocs The following example starts Filer with the -w option on vprocs 2, 3 and 4, but starts Filer normally (without the -w option) on all other vprocs. start filer -w -r 2, 3, 4 This command can also be invoked using the following syntax: start filer -w -r 2:4 Example: Starting Filer in -y diagnostic mode on four vprocs The following example starts Filer with the -y option on vprocs 1, 4 through 7, and 9, but starts Filer normally on all other vprocs. start filer -y -r 1, 4:7, 9 Example: Starting Filer in diagnostic -v mode on one vproc The following example starts Filer with the -n option on vproc 3 only. Since the Filer -v option (the second - v) is used, the target vproc (vproc 3 as specified by the -v 3 option) must be listed in the arguments for the -r option. start -v 3 filer -v -n -r 3 Filer Command Syntax Filer presents a command-line environment that allows the entry of Filer commands. This section describes the general conventions involved in entering Filer commands. Entering Filer Commands The following is the general form for a command entry: Support Utilities, Release

62 Chapter 3: Filer Utility (filer) Filer Command Syntax Syntax Element cmd cmdoption parameterlist Description The Filer command. The options that are specific to the command that you type. Different options pertain to specific commands. To determine the allowed options, see the specific command. The parameters included with specific commands. Command Usage Rules The following command usage rules apply to Filer commands: A space is required between the following: The cmdoption and the parameterlist The cmd and the parameterlist if you do not specify cmdoption or if the cmdoption is at the end of the command You can combine multiple Filer commands on a single command line, as shown below: Filer is case insensitive, which means that you can enter any command, keyword, or option in uppercase, lowercase, or mixed-case letters. If you end a command line with a backward slash (\), the command line continues on the next line. If an error occurs during processing of a string of commands, processing stops because the latter commands usually depend on the correct processing of commands typed earlier, as shown below: ci /p ;delete /y In this case, if there is a problem accessing the CI, you might not want to execute the DELETE command. If an error occurs while parsing the command line, you can type a single question mark (?) as the next command. Filer displays a list of alternatives expected by the parser at that point of the syntax error. If you type another single question mark (?) after Filer displays the alternatives, Filer returns the HELP text for the command in question. The following example shows help information about the RADIX command: Filer ==>radix in dex radix in de<-syntax error->x Filer ==>? Valid input at the point of error is: ; end of command Filer ==>? RADIX [ ( IN[PUT] OUT[PUT] ) ] [ ( H[EX] D[EC] ) 62 Support Utilities, Release 16.00

63 Filer ==>? RADIX [ ( IN[PUT] OUT[PUT] ) ] [ ( H[EX] D[EC] ) ] Chapter 3: Filer Utility (filer) Filer Command Syntax Sets the Flags for how to treat Unqualified number. Either Hex (base 16) or Decimal (Base 10), respectively. See HELP NUMBER for a description of unqualified INPUT. The initial setting of these Flags are HEX. If neither INPUT nor OUTPUT is specified, the command applies to both Flags. If neither HEX nor DEC is specified, the current setting of the Flag is displayed. Filer ==>? No more information available. Use HELP /L Comments are allowed anywhere a blank is allowed. Enclose comments between braces ({}). If Filer does not find a closing brace (}) on the line, Filer interprets the rest of the line as a comment, as shown in the following example: Filer ==> db 0 2 1fa { this is the broken data block If you use a single question mark (?) in place of a legal syntactic element, Filer informs you as to what you can type at that point in the command, as shown below: Filer ==> output? Valid input at the? is :, INTO OVER TO ON ; end of command In this example, Filer accepts any of the following: A comma (,) INTO OVER TO ON A semicolon (;) Other examples of using a single question mark (?) in place of a legal syntactic element are shown below: output? output?{comment} output? {comment output {comment}? Unlike other commands, the PATCH command cannot be combined on the command line with other commands. For more information, see PATCH. Filer Modes When started, Filer displays the Filer command prompt in the Filer application window of Database Window: Filer ==> Support Utilities, Release

64 Chapter 3: Filer Utility (filer) Filer Command Syntax Filer enters different modes of operation, based on the last command entered. The modes are generally associated with the hierarchical level of the file system architecture at which Filer is currently operating. The Filer command prompt changes to reflect the current mode: Filer ==> mi /p Command has been sent to Slave tasks. Mon Apr 21, :06:26 4 of 4 vprocs responded with no messages or errors. MI (Range) ==> Filer commands are prompt-sensitive: Global and upper-level commands are valid at any prompt. Lower-level commands are valid only under certain prompts that are initiated by upper-level commands. For more information on Filer command types, see Filer Command Types. Filer retains data object parameters for the current level of operation until they are changed or until Filer changes to a different mode. For example, once a range of rows has been specified for Filer to act upon, subsequent commands will operate on those same rows, without requiring re-entry of the row range specification. For more information, see Stored Input Values. Specifying Data Objects This section provides detailed information on the following: Numeric Input Multitoken Parameters Cylinders and Data Blocks Rows and Row Ranges Specifying a Subtable Identifier (tid) Stored Input Values Numeric Input Numeric values can be entered in either decimal or hexadecimal format. The default numeric base for data input to and output depends on the radix settings for input and output: When the radix for input is decimal, numeric input is interpreted as decimal numbers. For example, input of 45 is interpreted as the value 45. When the radix for input in hexadecimal, numeric input is interpreted as hexadecimal numbers. For example, input of 45 is interpreted as the value 69. The initial radix setting is hexadecimal for input and output. For more information on setting the radix, see the documentation for the RADIX command. The following special numeric formatting conventions can be used to force entered numeric values to be interpreted as decimal or hexadecimal, regardless of the radix setting: 64 Support Utilities, Release 16.00

65 Chapter 3: Filer Utility (filer) Filer Command Syntax Decimal values can be signified by adding a 0i or i prefix, or by adding a decimal point suffix: 0i I45 Hexadecimal values can be signified by adding a 0x or x prefix, or by adding an h suffix. Leading zeros are optional: 0x2D X2D 2Dh 002DH The valid range of for numeric values is unsigned 16-bit values, 0 through (0x0 through 0xFFFF), except when patching and using /L or /W for 32-bit integers, in which case the valid range is 0 through 4,294,967,295 (0x0 through 0xFFFFFFFF). You must separate two numbers on the same line from each other by a space or a comma. Multitoken Parameters Multitoken parameters, such as subtable identifiers, which are typed as three values, are typed on a single line with one or more spaces separating the individual tokens. You can also enter multitoken parameters separated by hyphens. For example, either of the following as subtable identifiers is acceptable: You can separate parameters from each other and from the command by spaces or a comma. To specify a command option (cmdoption ), type it on the same line as the command (cmd). Cylinders and Data Blocks Each cylinder has a 64-bit identifier, represented as a cylid in Filer commands. Cylids must be entered as 16-character hexadecimal numbers regardless of the Filer input radix setting. Data blocks are specified by a cylid, and by their starting sector number and length, represented as sectornum and sectorcount in Filer commands. Type each of these parameters as a single numeric value. The format for entering sectornum and sectorcount values is subject to the input radix setting. For more information on the formatting of numeric input and output, see Numeric Input and RADIX. Rows and Row Ranges Every physical row is uniquely identified by a row ID. A row ID specification, also referred to as a ridspec or a rowspec, is a value that includes a partition number, rowhash or hash bucket number, and uniqueness value. It can occupy up to 16 bytes in a physical row: The partition number is stored as 2 bytes if the number of partitions defined for the table is up to 65535, or 8 bytes if the number is greater. If the table is not partitioned, no space is reserved in the row ID for a partition specification. A value of zero means the table is not partitioned. The specification of the remainder of the row ID depends on whether the table includes a primary index: PI tables have a 4-byte rowhash value and a 4-byte uniqueness value, which is used to impose uniqueness in cases of rowhash synonyms for different physical rows. Tables without a primary index (NoPI tables) and tables with a primary AMP index (PA tables) have a 16- or 20-bit hash bucket value and a 44-bit uniqueness value. For systems using 16-bit hash bucket values, the four bits between the hash bucket value and the uniqueness value are unused. Support Utilities, Release

66 Chapter 3: Filer Utility (filer) Filer Command Syntax Note: The remainder of the RowID is treated as two 4-byte values, and present them as hash and uniq fields: hash0 hash1 uniq0 uniq FFA9 CB For NoPI and PA tables displayed or manipulated, hash0 and hash1 actually represent a 20-bit hash bucket value and the 12 high-order bits of the uniqueness value, while uniq0 and uniq1 display the 32 low-order bits of the uniqueness value. For more information on NoPI and PA tables, see SQL Data Definition Language - Syntax and Examples. A range of rows, also called a ridrange, is specified by the ridspec of the starting and ending rows: For purposes of input, the required and optional components of a ridspec depend on whether the table containing the physical row is partitioned. Table Type partitioned Ridspec Syntax nonpartitioned Syntax Element row_partition Description The partition number of the physical row. The partition number is a single numeric value. It is the internal partition number used by the file system, which is the value shown in Filer command output screens under the column heading part. The value must be between 0 and (0 and 7FFFFFFFFFFFFFFF in hexadecimal notation). For tables with multiple levels of partitioning, each physical partition results from the combination of all partitioning expressions in the PARTITION BY clause for a specific set of values in the partitioning columns. This gives a single, unique combined partition number, which is mapped to an internal partition number. Although the combined partition number of a partition can 66 Support Utilities, Release 16.00

67 Chapter 3: Filer Utility (filer) Filer Command Syntax Syntax Element row_hash row_uniq Description change if partition ranges are dropped or added for the partition level, the internal partition number does not change. For rows in nonpartitioned tables, the partition number must be zero, if specified, or can be omitted from the ridspec. The row hash value of the physical row. The row hash determines the AMP to which the physical row is distributed. The row hash consists of 2 numeric values, usually shown in output screens under column headings hash0 and hash1. For start_rid_spec : row_hash defaults to 0x0000 0x0000. For end_rid_spec : row_hash defaults to the value of row_hash specified in start_rid_spec. If no value for row_hash was specified in start_rid_spec, row_hash in end_rid_spec defaults to 0xFFFF 0xFFFF. The system-generated uniqueness value which is used with the row hash value to uniquely identify a physical row. The uniqueness value consists of two numeric values, usually shown in output screens under column headings u0 and u1. For start_rid_spec : If row_uniq is omitted, and no end_rid_spec is specified, all the rows in row_hash are selected. If row_uniq is omitted, and an end_rid_spec is specified, row_uniq defaults 0x0000 0x0000. For end_rid_spec : row_uniq defaults to the value of row_uniq specified in start_rid_spec. If no value for row_uniq was specified in start_rid_spec, row_uniq in end_rid_spec defaults to 0xFFFF 0xFFFF. = The program will use the most recently saved values for row_partition, row_hash, or row_uniq. The most recently specified values for row_partition, row_hash, and row_uniq are stored. Separate values are stored for start_rid_spec and end_rid_spec. row_partition, row_hash, or row_uniq. An error is reported if a saved start_rid_spec or end_rid_spec value is used for a partitioned table, but the table referenced in the current command is not partitioned, and vice versa. Example: Row range notation in SHOWDEFAULTS command output Filer ==> showdefaults The current object selected is Nothing The current defaults are: The present value for the = is: Dec (Hex) Cylinder Id = : 0000FFFFFFFFFFFF Sector Number 972 (03CC) Sector Count 16 (0010) Support Utilities, Release

68 Chapter 3: Filer Utility (filer) Filer Command Syntax Memory Address is undefined. Length is undefined. Patch Offset is undefined. TID ( A 0400) Start Rowid ( ) End Rowid (FFFF ) Start WLSN is undefined End WLSN is undefined Where Clause is undefined Filer is running as a startable utility Filer is running with Locking -- Enabled Filer is running with Script Mode -- Disabled File System has Internal Tracing -- Enabled Example: Row IDs in DBROW command output DB ==> dbrow /m ROW length rowid flags presences part hash0 hash1 uniq0 uniq CD C6B 71AB C D47 ADC D57 B91E C 00D F CDAE D F BD Specifying a Subtable Identifier (tid) Commands operate on subtables, rather than on tables. A table is a collection of subtables, and a subtable is a collection of physical rows. Each subtable has physical rows that store one particular type of data. For example, primary and fallback data subtables have data rows, primary and fallback index subtables have index rows, and the physical rows of table header subtables contain table metadata. Each subtable is identified by a subtable identifier ( tid ). The tid has two parts. The first part identifies the table to which the subtable belongs. All subtables that are part of the same table have the same table identifier as part of their tid. The second part identifies one of the subtables that comprise the table, and indicate the subtable type. A tid is defined as follows: table_unique and type_and_index together uniquely identify a subtable. Syntax Element table_unique Description Unique identifier of the table to be acted upon. The table can be uniquely identified in either of two ways: 68 Support Utilities, Release 16.00

69 Chapter 3: Filer Utility (filer) Filer Command Syntax Syntax Element Description 1. Specify the name of the database to which the table belongs and the table name, separated by a period. The names individually or together must be delimited. The following formats are valid: " database_name. table_name" " database_name "." table_name " ' database_name. table_name' ' database_name '.' table_name' 2. Specify the unique numeric identifier of the table, which consists of two numeric values separated by a space. The table identifier consists of the first two numeric values returned by the TABLEID command. These two numbers are common to all subtables that comprise the table. Note: The format of the input numbers depends on the current radix setting, which is displayed by the RADIX command. = Specifies to use the most recently saved value for table_unique. type_and_index Uniquely identifies the subtable within the group of subtables that comprise a table. The type_and_index identifies both the type of subtable, and the individual subtable itself. The syntax for type_and _index is as follows: Syntax Element type Description The type of subtable. type can be one of the following: Type Description * All the subtables of this table H The table header subtable. Because table header subtables have only one row and no secondary indexes or work subtables, any index and variant specifications that follow an H type are ignored. Support Utilities, Release

70 Chapter 3: Filer Utility (filer) Filer Command Syntax Syntax Element Description Type P F number F * Description The primary data subtable. The fallback data subtable specified by number. The default is 1. All of the fallback subtables. index The index subtable to examine. If an index subtable value is not provided, it defaults to the data subtable. Index 1 is the first secondary index. index 2 is the second secondary index, and so forth. index can be one of the following: number * D A secondary index that can be used instead of specifying X number. if you enter number by itself, it must be a multipleof 4, and it is interpreted in the following ways: 0 is the same as entering X0 4 is the same as entering X1 8 is the same as entering X2 12 is the same as entering X3 and so forth. All the indexes of the table. The primary data index (same as X0 or 0). X number X* The secondary index specified by number. The default is 1. All the secondary indexes, starting at 0. variant The possible subtables. variant can be one of the following: * All of the possible variant subtables. The default used if a variant value is not specified. The alue during a Sort or Table Modify operation. 70 Support Utilities, Release 16.00

71 Chapter 3: Filer Utility (filer) Filer Command Syntax Syntax Element Description 3 Unused. Unused. number A single number that represents internally the type of subtable (header, primary data, or fallback data), the index to use wehn ordering the physical rows (primary data index or one of hte secondary indexes), and the variant. The table header has a type_and_index number of zero. Examples of these numbers are listed below. The following table gives examples that describe the type_and _index fields. Subtable Description Number Type Type/Index Type/Index/Variant Table header 0 H Primary data subtable First secondary index 1024 (0x0400) P P/D 1028 (0x0404) P/X1 First fallback table 2048 (0x0800) F1 F1/D Second secondary index of the third fallback table All primary subtables All primary secondary indexes All fallback subtables All subtables of this table Sort table of the first secondary index Both tables during a sort of the first secondary index 4104 (0x1008) F3/X2 P/* P/X* F* * 1029 (0x0405) P/X1/1 P/X1/ * For example, assume that table T4 is a table in database MYDB and has a table_number of Also assume that input is accepted in hexadecimal format. Some valid specifications of a tid for primary subtables of table T4 are as follows: Support Utilities, Release

72 Chapter 3: Filer Utility (filer) Filer Command Syntax "MYDB.T4" 400 "MYDB.T4" "MYDB"."T4" P u 'MYDB'.'T4' 400 h P Stored Input Values Filer stores the most recent values entered for the following parameters: cylid endcylid length memaddr offset row_hash row_partition row_uniq sectorcount sectornum startcylid tablenumber tid You can enter an equal sign (=) for any of these parameters to use the currently stored value. These context values are position-dependent. (For example, values stored for a start_rid_spec can be used only as part of a start_rid_spec argument in a later command and not as part of an end_rid_spec). The stored values are updated by the commands as well. For example, if you select a data block, the stored value for the tid is updated to the subtable identifier of the data block, and the RowIDs are updated to the first and last physical row in the data block. Note: To use = to represent a stored value for a current command, you must have entered an input value for that parameter in a previous command, or issued a previous command which saved a value for that parameter. Use the SHOWDEFAULTS command to see the currently stored values. For more information, see SHOWDEFAULTS. Specifying Display Ranges (ordinalrange) Filer presents a command-line environment that allows the entry of Filer commands. This section describes the general conventions involved in entering Filer commands. The ordinalrange parameter is used to specify a range of objects relative to the current position to be displayed. ordinalrange can be used on the following commands: DISPLAY TABLE WAL 72 Support Utilities, Release 16.00

73 Chapter 3: Filer Utility (filer) Filer Command Syntax When used on the DISPLAY command TABLE command WAL command ordinalrange restricts or expands the initial context, producing a different resulting context. DISPLAY shows the objects in the new context. Filer saves the context from a previous command, and the ordinalrange is interpreted with respect to this initial context. The initial context for the DISPLAY command can be a physical row, subtable, CI, CID, DB, DBD, MI, WAL log record, WCI, WCID, WDB, WDBD, WMI, or a range thereof (except for the MI and WMI which are singular contexts). restricts the range of physical rows to display, but does not influence the resulting context. An implicit initial context consisting of the physical rows of the table, and the ordinalrange is interpreted with respect to this context. The initial context for the TABLE command is always this same row range context. No matter what is displayed, the resulting context is always a subtable rather than a row range. restricts the WAL log records that are displayed. The initial context is from the first WAL log record to the last, and the ordinalrange is interpreted with respect to this context. The resulting context is a WAL log record or WAL log record range; therefore, the resulting context matches the records selected. Note: The Filer online help refers to ordinalrange as rangeopt, but the usage is the same. ordinalrange input is as follows: startand stop are defined as follows: Support Utilities, Release

74 Chapter 3: Filer Utility (filer) Filer Command Syntax Syntax Element BEG END Description BEG always refers to the first object in the context. If the context is singular, BEG and END refer to the same object. END always refers to the last object in the context. If the context is singular, BEG and END refer to the same object. +/- number The meaning of +/- number without BEG/END depends on whether it is used on a DISPLAY, TABLE, or WAL command: BEG/END +/- number DISPLAY command + number is equivalent to END + number, and number is equivalent to BEG - number, as described below: A + number is a displacement from the last object in the context, expanding the context beyond the last object by the specified number of objects. A - number is a displacement from the first object in the context, expanding the context before the first object by the specified number of objects. TABLE command + number is equivalent to BEG + number - 1, and number is equivalent to ENDnumber, as described below: A +number is an ordinal row number, where 1 is the first physical row. A - number is a backwards displacement from the last physical row, so that -1 is the second to the last physical row. The exception is if the TABLE command uses a wildcard for the type part of the tid. In this case, the resultant context is special (nothing), and the display is a multiple subtable display, where each subtable is constrained by the ordinal range. For each subtable, the BEG/END +/- number rules apply, except that the number construct is not allowed. WAL command + number is equivalent to BEG + number - 1, and number is equivalent to ENDnumber, as described below: A + number is an ordinal record number, where 1 is the first WAL log record. A - number is a backwards displacement from the last record, so that -1 is the second to the last WAL log record. For examples, see DISPLAY, TABLE, and WAL. Allows specifying an object relative to the first or last object of a context. When used on a DISPLAY command, BEG/END +/- number can be used to expand or contract a context. When used on a TABLE or WAL command, BEG/END +/- number can only restrict the initial context for display purposes. If the current position encompasses a range of objects, then start or the period (.) means from the first object in the range. $ If the current position encompasses a range of objects, then stop or the dollar sign ($) means from the last object in the range. 74 Support Utilities, Release 16.00

75 Chapter 3: Filer Utility (filer) Filer Command Syntax Syntax Element */ALL Description If you specify an asterisk (*) or ALL as the end field, ALL is translated into an end field. After an operation involving ordinalrange, the context, including the currently selected object, is adjusted to be these new objects just acted upon by this command. Specifying a WHERE Clause The WHERE clause can be supplied to the DISPLAY, WAL, and WREC commands to restrict selection based on row/record type or content. The constraint is in addition to the ridrange, ordinalrange and wlsnrange constraints. For more information about these constraints, see the description of the DISPLAY, WAL, or WREC command. Most of the constraints in the WHERE clause apply to WAL log records only, and not to physical rows in normal tables. Syntax Syntax Element kind Description Types of WAL log records to display. The following are the valid values. ALLOCSECT BACKOUTSECT BEFORECHECKSUM CHECKPOINT DBD DIAGNOSTIC DUMMYSNAPMI FREECYL FREELINEAGE FSP HDRMIGINFO HDRONLY Support Utilities, Release

76 Chapter 3: Filer Utility (filer) Filer Command Syntax Syntax Element Description NEWDB NONTJ ROW ROWCOPY ROWEXISTCOMPDB ROWNEWSECT SRD TJRECORD UNFSE action flags ntastate tjkind Indicates the action to be redone in certain NONTJ records. The following are the valid values. INSERT UPDATE DELETE The following are the derived flags values that are valid in the WHERE clause. NTA NTABEGIN NTAEND The following are the valid values. OPEN CLOSED Types of Transient Journal (TJ) records to display. The following are the valid values. NILL TIP HSTSTRT BEG END CLRDMP FALL DELTBL BEFROW DELROW NEWROW DELIDX NEWIDX MODTBL DBLOK PJTRNBEG PJROW 76 Support Utilities, Release 16.00

77 Chapter 3: Filer Utility (filer) Filer Command Syntax Syntax Element Description DELPJ DELALL INSETB TIPLOK TIPAPJ TBLCACTION SESSEP 2PCVOTE 2PCABRT 2PCSTAT IDT REQABT NEWTMPTBL BEGSAVEPT ENDSAVEPT SAVEPTABT NEWLOB DELETELOB DELPART INSPART QTNEWROW QTDELROW FATMODTBL REPACT MARKER NUSIDEL NUSIINS BEFTHROW BEFTHROWFLOAD CRESNAPSHOT REMSNAPSHOT RELCHN NESTEDDONE NEWOLJ TJTABLE tid TJROW rowid tid of the table referenced inside a TJ record. Some TJ records contain the subtable identifier (tid) and row identifier for the applicable physical row (the row to which the TJ record will be applied during rollback). This is displayed as a 4-byte table number and 2-byte typeandindex. However, as a TJTABLE argument, this is supplied in the normal Filer tid format, using two numbers for the halves of the table number, and a third number for the typeandindex. For more information on tid formatting, see Subtable Identifiers (tids). rowid of the row referenced inside a TJ record. This is the RowID of the database row that the TJ record applies to. For Support Utilities, Release

78 Chapter 3: Filer Utility (filer) Filer Command Syntax Syntax Element TJTRANS procnum transuniq trxstate TJHOST number TJSESSION number TJREQUEST number TJRCVSESS number Description example, this identifies the physical row whose update caused the TJ record to be created. TJ record transaction identification. procnum and transuniq are both numbers. For the TJ records, display only the TJ records for given transaction states. The following are the valid values. OPEN ABORTED CLOSED TJ record host number. TJ record session number. For the TJ records, display only the TJ records for these requests. For the TJ records, display only the TJ records for these recovery sessions. Usage Notes You can enter Keyword arguments in any order. Separate Keyword arguments (some of which contain lists) with a space. Within a list, separate items with a comma. KIND=TJRECORD implies a WAL type of TJ. The kind value NONTJ does not represent an actual WAL log record kind, but is provided as a convenience. When specifying the KIND keyword, a mix of NONTJ and TJRECORD kinds is allowed. In qualifying records for display, if the record is a NONTJ kind, the NONTJ selectors (kind, action, flags, and ntastate) are used. If the record is a TJRECORD kind, then the TJ selectors (tjkind, tid, etc.) are used. The NTASTATE, TRXSTATE, and TJRCVSESS arguments require special processing because these values are not recorded in the log records. The log is read backwards to determine transaction states or to find recovery session boundaries. It is then read forwards to display selected log records, which can be further filtered by other arguments. The WHERE clause provides selection from the WAL log in a single, integrated display stream, displaying both NONTJ records and TJ records in time sequence when used with Filer commands such as WAL and WREC. Example: Using the WHERE clause to restrict Filer actions The following example selects TJ BEG and END records for host 7, and checkpoint records. wal /m where kind=checkpoint ; tjkind=beg,end ; tjhost=7 78 Support Utilities, Release 16.00

79 Example: Using the WHERE clause to delete a TJ END record Chapter 3: Filer Utility (filer) Repairing Cylinders The following example finds a TJ END record and deletes it. The output is not shown, but the desired TJ END record with WLSN 43e72 is assumed to be found amongst the TJ END records returned. wal /L where tjkind=end wrec /L 43e72 delete y // display TJ END records // set context to record found // request delete of selected record // confirm delete request Example: Using the WHERE clause to find open transactions on a specified host The following example finds all TJ records for open transactions from host 7. wal /L where tjhost=7 trxstate=open Example: Using the WHERE clause to find WAL and TJ insert records The following example displays WAL insert records and TJ insert records. wal /l where action=insert ; tjkind=newrow Repairing Cylinders Using Filer to make repairs to certain areas of the file system requires knowledge of data structures related to the Teradata File System. Teradata Database allocates cylinders for data storage of two basic types: Normal cylinders are used for most file system storage, including PERM, SPOOL, JOURNAL, and WAL. Normal cylinders can be of two fixed sizes: Small cylinders each have 3872 sectors. Each sector is 512 bytes. Therefore a small cylinder is nearly 2 MB. On systems that have been initialized to use large cylinders, small cylinders are still used to store WAL and DEPOT data. Large cylinders are the size of six small cylinders, nearly 12 MB. On systems that have been initialized to use large cylinders, large cylinders store all data types except WAL and DEPOT. Note: For 4 KB storage devices, Teradata Database treats 4 KB sectors as eight contiguous 512 byte sectors. Durable cylinders are used to store certain special file system objects, such as the master index (MI) and file trace buffer (FTR). These are called durable structures because, after they are created in the file system, they persist. Unlike normal cylinders, durable cylinders are never freed back into general storage. Teradata Database employs a directory structure known as the cylinder map. The map stores a list of all cylinders (small, large, and durable) on the system. Although it is unlikely, if one or more entries in the map becomes corrupted, Teradata Database cannot properly identify the corresponding cylinder types or whether they are large or small cylinders. For example, a durable cylinder can appear to the file system Support Utilities, Release

80 Chapter 3: Filer Utility (filer) Filer Commands incorrectly as a set of normal cylinders with invalid data, or a set of contiguous small cylinders can appear a single large non-durable cylinder. Such situations result in file system errors. Use the following approaches to repairing such problems, depending on the nature of the problem: If a durable cylinders is misidentified as a set of normal small cylinders, the simplest method of repair is to reallocate the durable structure, using the REALLOCDUR command. The normal cylinder sized portions of storage that had been part of the durable cylinder before reallocation can be returned to the pool of general cylinder storage using the QUARLIST /R, BADLIST /R, ASSIGN, or INIT command. Alternatively, an attempt can be made to manually repair the first CI of the durable structure that was corrupted, such that it will again be recognized as a durable cylinder. Once the CI has been repaired, use the REBUILDMAP command to force the Teradata Database File System to again recognize the durable cylinder. When a set of normal small cylinders is misidentified as a large (non-durable) cylinder, either use the ASSIGN command with the BREAKUP option, or use the PATCH command to appropriate size the first CI of the cylinder set. In either case, quit and restart Filer after the repairs. This restarts the file system with the newly repaired CIs, which allows verification of the repairs. For more information on Filer commands related to cylinder repair, see ASSIGN, BADLIST, INIT, PATCH, QUARLIST, REALLOCDUR, and REBUILDMAP. Filer Commands Filer Command Types There are three general types of Filer commands: Global commands can be run from any Filer prompt. They display information (help, Filer defaults, current date and time), determine how Filer runs (redirect output and input, set numeric format for output and input, enable and disable Filer features, set the priority of the Filer process), and exit the Filer program. Upper-level commands can be run from any Filer prompt. They direct Filer to display a specific object in the file system, for example a cylinder index. The Filer prompt changes to reflect the object type, for example CI for cylinder index. While the prompt shows a particular object type, Filer accepts commands specific to objects at that level. Most upper-level commands require one or more parameters that specify the particular object on which to act. For example, each cylinder index is specified by a unique cylinder ID (cylid). Filer remembers the specifications and continues to act on the same object until either new parameters are entered that identify a different object at the same level, or a command is entered that switches Filer to a different level in the file system hierarchy. Lower-level commands manipulate the currently selected file system object. These commands behave slightly differently, depending on the currently selected object type. While some of these commands are harmless, such as DISPLAY, DROP, NEXT and PREVIOUS, others manipulate file system data that is critical to the operation of the system. Notice: You should be completely familiar with file system concepts and architecture before using the lower-level commands. Improper use of these commands can seriously damage your system. 80 Support Utilities, Release 16.00

81 Hex Dumps Format Chapter 3: Filer Utility (filer) Filer Commands Some Filer commands display disk content as a hex dump of the byte values recorded on the disk. Hex dumps are displayed in hexadecimal format, regardless of the radix output setting. For example: 0000: A 0B 0C 0D 0E 0F *...* 0010: A 2B 2C 2D 2E 2F *!"$%& ()*+,-./ * 0020: A 1B 1C 1D 1E 1F *...* The hexadecimal number in the first column represents the offset of the rest of the row from the beginning of the file segment. In the example, the 0020 in the first column of the third row means that the character that follows, having a hex value of 10, is 0020 bytes (32 bytes in decimal notation) from the beginning of the segment. The last column in the row, delimited by stars at each end, shows the ASCII representation of the hexadecimal numbers displayed in the row. List of Commands Filer presents a command-line environment that allows the entry of the following Filer commands, which are discussed in more detail in the sections that follow. Command ACCESS ALTCI ALTWCI ASSIGN BADLIST BLK CHECKSUM CI CID CLEAREIOCACHE DATE/TIME DB Description Accesses a specific copy of a CI or DB segment located on DISK, DRAM, or is the LATEST. Makes the non-active CI copy the active copy, and the formerly active copy the non-active copy. Makes the non-active WCI copy the active copy, and the formerly active copy the non-active copy. Assigns a repaired cylinder back to a specified AMP and storage type. Normally used after a cylinder that is in the bad list or quarantine list has been repaired. Displays list of bad cylinders, scans them for preliminary diagnosis, initializes and releases unrepairable cylinders back to storage. Selects the segment that starts at the specified cylinder and sector number, determines the type of block, and displays it accordingly. Validates, displays, and optionally repairs the checksum for a DB, WDB, CI, or WCI segment. Selects and displays the Cylinder Index (CI) segment for the specified cylinder. Selects and displays the Cylinder Index Descriptor (CID) for the specified cylinder. Clears the EIO cache. The EIO cache is an additional, fast, readonly I/O cache that exists between the PDE FSG cache and the disk storage subsystem on some platforms. Displays the current system day, date, and time. Selects and displays the Data Block (DB) segment that starts at the specified cylinder and sector number. Support Utilities, Release

82 Chapter 3: Filer Utility (filer) Filer Commands Command DBD DBROW DELETE DEPOT DISABLE DISCARD DISCARD DROP DUMPDISK ENABLE ERRORS FIB FIND FREECACHE HELP IDENT INFO INIT INPUT MEMBLK Description Displays the Data Block Descriptors (DBDs) of the current cylinder. Displays the physical rows of a DB. Deletes the currently selected object or range of objects, including all descendants and antecedents. Displays cylinder summary information for cylinders in the Depot, or block summary information for blocks in a particular Depot cylinder. Sets a specific flag in the file system to FALSE, disabling certain features of Filer. Most such flags are for internal use only. Discards a rebuilt CI in memory without affecting the original CI. Shows a formatted display of the currently selected object or range of objects. If you specify an ordinalrange, then DISPLAY shows a related object or range. Drops the access on the current object and returns to the default prompt. Selects the disk segment that starts at the specified cylinder and sector number, determines the type of block, and displays it accordingly. Sets a specific flag in the file system to TRUE, enabling certain features of Filer. Most such flags are for internal use only. Redirects diagnostic messages to a file that you specify or to the default file, stderr. Displays the File Information Block (FIB) and information about cylinder usage. Given the table and optional ridrange, displays the cylinders containing the range. Displays the status of the free cylinder cache. Provides context sensitive help for Filer commands. Identifies the objects in the specified sectors, and displays the properties of the objects. Displays details of a rebuilt CI that is still in memory, not yet written to disk. Initializes the current cylinder or range of cylinders to free. Directs Filer to read commands from a specified file rather than from the default input file, stdin. Accesses the file system segment at the specified memory address, determines the type of the segment, and returns a formatted display of the segment. 82 Support Utilities, Release 16.00

83 Chapter 3: Filer Utility (filer) Filer Commands Command MEMCTX MEMDB MEMDUMP MEMROW MEMSORCTX MEMWCTX MEMWDB MEMWREC MI MODIFY REDO NEXT OUTPUT PACK PATCH PREV PRIORITY QUARLIST QUIT RADIX REALLOCDUR REBUILDCI REBUILDMAP Description Returns a formatted display of a file system context structure (ctxcontext_t) beginning at the specified memory address. Returns a formatted display of a DB beginning at the specified memory address. Dumps memory in the hexadecimal format, starting at the specified location for a specified length of bytes. Formats a row beginning at the specified memory address. Returns a formatted display of a context structure (sorcontext_t) beginning at the specified memory address. Returns a formatted display of a context structure (wtjcontext_t) beginning at the specified memory address. Returns a formatted display of a WDB beginning at the specified memory address. Formats a WAL log record beginning at the specified memory address. Selects and displays the Master Index (MI). Replays the specified WAL log records, or defines whether or not to replay the WAL log records during the next file system startup. Displays the next row, WAL log record, DB, WDB, DBD, WDBD, CI, WCI, CID, WCID or subtable. Redirects Filer output to a file you specify or to the default file, stdout. Packs the segment to eliminate fragmentation. Modifies file system segments and other objects in memory. Displays a previous row, WAL log record, DB, WDB, DBD, WDBD, CI, WCI, CID, WCID or subtable. Sets the priority class of the Filer process. Displays list of quarantined cylinders, scans them for preliminary diagnosis, initializes and releases unrepairable cylinders back to storage. Ends a Filer session. Sets the default radix (decimal or hexadecimal) used as the numeric base for Filer data input and output. Reallocates durable cylinders. Rebuilds the cylinder index for a specified cylinder. Rebuilds the map of allocated cylinders. Support Utilities, Release

84 Chapter 3: Filer Utility (filer) Filer Commands Command RENAME REPAIRSECTS RESETFREECYL ROW ROW SCOPE (or VPROC) Description Renames the currently selected segment to the specified location on the disk. Deletes specified overlapping data blocks from rebuilt CIs in memory. Resets the file system internal free cylinder count back to the correct value. Displays physical rows from one or more subtables. For a single subtable, ROW sets the context to those rows. Verifies the integrity of the file system. Selects a set of vprocs that Filer will communicate with until Filer quits or a different set of vprocs is selected. SEGZERO (or SEG0) Selects and displays Segment Zero (0). SET FREECYL SHOWDEFAULTS SNAPMI SRD STAMP STORAGEINFO SYNCSCAN TABLE TABLEID TRACEPRINT TRASH VPROC WABC WAL WCI Sets the file system internal Free Cylinder count to the specified number. Displays the default settings and the saved context. Selects and displays the snapshot MI. Returns a formatted display of all the Subtable Descriptors (SRDs) on the current CI that have any rows from the specified subtable tid. If tid is not specified, then all SRDs are displayed. Takes five hexadecimal numbers and converts them to a file system timestamp. Displays per-amp system control information for storage. Shows all currently running scans on an AMP that are eligible for the synchronized full file scan feature. Displays rows from the subtable specified by tid. Displays the table number of a specified table. Dumps the internal trace buffer of the file system. The cylid limits the display to entries to that cylinder. Removes the FSG name of the copy of the currently selected segment in memory, and retains access to it. Selects a set of vprocs that Filer will communicate with until Filer quits or a different set of vprocs is selected. Displays the WAL Append Buffer Control (WABC) structure. Displays Write Ahead Logging (WAL) log records in an ordinal record number range, such as the 10th through the 20th WAL log records. Selects and displays the WAL Cylinder Index (WCI). 84 Support Utilities, Release 16.00

85 Chapter 3: Filer Utility (filer) Filer Commands Command WCID WDB WDBD WDBREC WFIND WFLUSH WMI WREC WRITE WSUMMARY Description Selects and displays the WAL Cylinder Index Descriptor (WCID) associated with the requested WAL log cylinder. Selects and displays the WAL Data Block (WDB) segment that starts at the specified cylinder and sector number. Displays the WAL Data Block Descriptors (WDBDs) of a WAL log cylinder. Displays WAL log records for a WDB. Displays the WAL log cylinders containing WAL log records in a given range, or if no range is specified, then all of the WAL log cylinders. Manually force some or all of the WAL log records in memory to disk. Returns a formatted display of the WAL Master Index (WMI). Displays the WAL log records within a given range and within the restrictions of the whereclause. Forces the currently selected segment to disk. Retrieves table IDs from WAL records within a given WAL log sequence number (WLSN) range. Support Utilities, Release

86 Chapter 3: Filer Utility (filer) ACCESS ACCESS Purpose The ACCESS command drops access to the current CI or DB segment, and accesses a copy of the same segment in the specified location. If a location is not given, ACCESS identifies the copy of the segment that is currently accessed. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /X, /H, /V) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /X /H /V place Positions to the segment. Nothing is displayed unless there is a problem. Displays the segment in a short format. This is the default. Displays the segment in a medium-length format. Displays the segment in a long format. Displays the segment in an extended format. Displays the segment in hexadecimal. Verifies that the segment is logically correct, but does not display it. The location of the segment you want to access. The valid values are DISK, DRAM, or LATEST. If you omit place, the ACCESS command identifies which copy of the segment is currently accessed. 86 Support Utilities, Release 16.00

87 Usage Notes Chapter 3: Filer Utility (filer) ACCESS The ACCESS command must be invoked from a CI or DB context (as indicated by the CI ==> or DB ==> prompt). Support Utilities, Release

88 Chapter 3: Filer Utility (filer) ALTCI ALTCI Purpose There are two copies of each CI, only one of which is active at any time. The active copy reflects the current state of the cylinder. The ALTCI command makes the non-active copy the active copy, and the formerly active copy the non-active one. Note: This command can be used only when Teradata Database is not running on the affected AMP; the AMP cannot be ONLINE. Syntax cylid The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. Usage Notes ALTCI can be used only when the scope is set to one vproc. For more information on the SCOPE command, see SCOPE (or VPROC). Picking the alternate copy of a CI might result in overlapping tableids/rowids across cylinders and mismatches between the WCI and the blocks on the associated cylinder. After performing this command, use scandisk to determine whether additional repairs are required. Example: Filer ALTCI command Filer ==> scope vproc 0 1 of 4 amps have been selected Filer ==> altci A *** WARNING *** Picking the alternate copy of a [W]CI might result in overlapping tableids/rowids across cylinders and mismatches between the [W]CI and the blocks on the associated cylinder. After performing this command, use scandisk to determine whether additional repairs are required. *** WARNING *** Do you want to switch the effective copy to the alternate copy? (Y/N) > y vproc 0 (0000) response Active CI Location : PRIMARY Blk Code : CI Version Number : 16 WLSN : ( FD453) Stamp : Mon Jul 25, :02:54.06 (1E F6 2D 4E 06) Non-Active CI Support Utilities, Release 16.00

89 Location : SECONDARY Blk Code : CI Version Number : 2 WLSN : ( FD294) Stamp : Mon Jul 25, :02:29.53 (05 F6 2D 4E 35) New Active CI Location : SECONDARY Blk Code : CI Version Number : 2 WLSN : ( FD294) Stamp : Mon Jul 25, :02:29.53 (05 F6 2D 4E 35) Successfully alternated the CI. Chapter 3: Filer Utility (filer) ALTCI Support Utilities, Release

90 Chapter 3: Filer Utility (filer) ALTWCI ALTWCI Purpose There are two copies of each WCI, only one of which is active at any time. The active copy reflects the current state of the WAL cylinder. The ALTWCI command makes the non-active copy the active copy, and the formerly active copy the non-active copy. Note: This command can be used only when Teradata Database is not running on the affected AMP. Syntax cylid The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. Usage Notes ALTWCI can be used only when the scope is set to one vproc. For more information on the SCOPE command, see SCOPE (or VPROC). Notice: Picking the alternate copy of a WCI might result in overlapping record numbers across cylinders and mismatches between the WCI and the blocks on the associated cylinder. After performing this command, use scandisk to determine whether additional repairs are required. Example: Filer ALTWCI command Filer ==> scope vproc 0 Command has been sent to Slave tasks. Tue Jul 03, :47:39 1 of 8 amps have been selected Filer ==> altwci 0 2A Active WCI Location: PRIMARY Blk Code: WCI Version Number: 19 WLSN : ( E32) Stamp : Wed Jul 11, :28:36.51 (64 2F ) Non-Active WCI Location: SECONDARY Blk Code: WCI Version Number: 3 90 Support Utilities, Release 16.00

91 WLSN : ( AF68) Stamp : Tue Jul 03, :51:10.01 (5E 62 8A 46 01) *** WARNING *** Picking the alternate copy of a [W]CI might result in overlapping tableids/rowids across cylinders and mismatches between the [W]CI and the blocks on the associated cylinder. After performing this command, use scandisk to determine whether additional repairs are required. *** WARNING *** Chapter 3: Filer Utility (filer) ALTWCI Do you want to switch the active copy to the alternate copy? (Y/N) > y y Active WCI Location: SECONDARY Blk Code: WCI Version Number: 3 WLSN : ( AF68) Stamp : Tue Jul 03, :51:10.07 (5E 62 8A 46 07) Successfully alternated the WCI. Support Utilities, Release

92 Chapter 3: Filer Utility (filer) ASSIGN ASSIGN Purpose The ASSIGN command reassigns a cylinder back to a specified AMP and storage type. ASSIGN is normally used after a cylinder that has been marked bad or quarantined has been repaired. Some types of errors can cause the file system to identify several contiguous small cylinders as a single large cylinder. In these situations, the ASSIGN command can be used to break the apparent large cylinder back into its component small cylinders, and reassign the small cylinders back to their proper AMPs and storage types. Note: This command can be used only when Teradata Database is not running on the affected AMP. Notice: This command should be used only to reassign repaired cylinders back to their original AMP vprocs. Improper use of this command can result in data corruption. Syntax cylid The 64-bit identifier of a small or large cylinder, entered as a 16-character hexadecimal number. 92 Support Utilities, Release 16.00

93 TARGETVPROC=vproc_num_or_list or TAR=vproc_num_or_list Chapter 3: Filer Utility (filer) ASSIGN A list of the vproc numbers that identify the AMPs to which the repaired cylinders will be assigned. For large cylinders, from which several small cylinders will be created, TARGETVPROC can be either: a single vproc number, to which all small cylinders will be assigned If the BREAKUP option is specified, all resulting small cylinders are assigned to the specified vproc. If the BREAKUP option is not specified, the large cylinder itself will be assigned to the specified vproc. a list of vproc numbers, one for each of the newly created small cylinders. The numbers must be separated by commas, and should enumerate in order the AMP numbers to which the resulting small cylinders will be assigned. If more than one vproc is specified, the BREAKUP option is implied. For large cylinders, if TARGETVPROC is not specified, the resulting small cylinders are assigned to the AMP vproc to which the source large cylinder belonged. TYPEOFCYL=typeofcyl The type of storage to which the cylinders are assigned. Valid types are: PERMCYL WALCYL JRNLCYL GLOBALTEMPCYL PSPOOLCYL For large cylinders, if TYPEOFCYL is not specified, the resultant small cylinders are assigned the same type as the large cylinder they comprised. They can be assigned different cylinder types by subsequently running the ASSIGN command on each individual small cylinder. BREAKUP or BREAK Splits the specified large cylinder into smaller cylinders, each with a new cylinder ID. Note: The BREAKUP option should not be used for small cylinders. If TARGETVPROC, TYPEOFCYL, or both are specified, the new cylinders will be explicitly assigned to the specified vproc and storage type. Usage Notes The ASSIGN command is related to the /R option of BADLIST and QUARLIST. ASSIGN re-associates a repaired cylinder back to a specific AMP, while the /R option releases a repaired cylinder back into the general storage pool. For more information on the proper use of this command to repair durable file system structures, see Repairing Cylinders Support Utilities, Release

94 Chapter 3: Filer Utility (filer) ASSIGN Example: Assigning a cylinder to a specific AMP and storage type The following example assigns a cylinder to a specific AMP and storage type. Filer ==> ssign typeofcyl=jrnlcyl targetvproc=vproc 3 Fri May 07, :00:31: Assign command will assign Cylinder belonging to vproc 2 with target cyltype of JOURNAL. Do you wish to continue? (Y/N) y The following is the extent reserved on amp 3: C Please quit filer and restart the database for the changes to take effect. Assign command completed successully Example: Using the Filer ASSIGN command to break large cylinders into component small cylinders Certain file system errors can cause several contiguous small cylinders to be misidentified as a single large cylinder. The ASSIGN command can break such large cylinders back into their component small cylinders, and assign the small cylinders back to their original AMP vprocs. The following example demonstrates using the ASSIGN command to break a large cylinder into its six component small cylinders, and assigns the small cylinders back to their original AMP vprocs. In this case, the first cylinder is assigned to vproc 2, the next to vproc 4, and so forth. Filer ==> assign F targetvproc=2,4,3,1,2,1 break Fri Jan 08, :03:32: Assign command will split Cylinder F on vproc 2 and with the spe cified type. Do you wish to continue? (Y/N) y vproc 2 (0002) response The following are the reserved 6 small cylinders: Please run rebuildci command on each small cylinder. Then quit filer and restart the database for the changes to take effect. Assign command completed successully 94 Support Utilities, Release 16.00

95 BACKDOWNCHECK Chapter 3: Filer Utility (filer) BACKDOWNCHECK Purpose The BACKDOWNCHECK command shows whether the current file system can be backed down to a previous version. If back down is not supported, the reasons are displayed. Note: This command should be used only under the direction of Teradata Support Center personnel. Syntax COMMIT /Y Performs the steps necessary to prepare the Teradata File System for backdown to the major (xx.0) release that corresponds to the current minor (xx.10) release. Bypasses the confirmation prompt normally displayed with the COMMIT option. Support Utilities, Release

96 Chapter 3: Filer Utility (filer) BADLIST BADLIST Purpose The BADLIST command displays a per-amp list of cylinders whose contents cannot be determined. Cylinders are designated as bad automatically when a data corruption in the cylinder index prevents the system from determining the contents of the cylinder. A list of bad cylinders is maintained on each AMP. AMPs that own bad cylinders are kept down, and the file system is not started on these AMPs during Teradata Database system startup. To bring the AMPs back up, use Filer to repair or reinitialize the bad cylinders. The BADLIST command can scan these cylinders for possible diagnoses. After the bad cylinders have been fixed, use the ASSIGN command to reassign the cylinders to an AMP and storage type, or use the /R option of BADLIST to remove the cylinder from the list, making the cylinder available for use from the general pool of storage cylinders. For more information on the ASSIGN command, see ASSIGN. Syntax cylid /P /S /R /X cylid The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. Prints the list of bad cylinders. This is the default. Scans a bad cylinder and attempts to diagnose problem. If no cylid is specified, scans all bad cylinders. Initializes a bad cylinder and releases it back into the general storage pool. If no cylid is specified, initializes and releases all bad cylinders. Initializes the cylinder specified by cylid, but does not release it back to storage. Used for durable cylinders, which should never be released to general storage, such as those used for master indexes and startup information. Usage Notes BADLIST can be used only when the Teradata Database is down. The /R option is related to the ASSIGN command. The /R option releases a repaired cylinder back into the general storage pool, while ASSIGN re-associates a repaired cylinder back to a specific AMP. 96 Support Utilities, Release 16.00

97 Chapter 3: Filer Utility (filer) BADLIST For more information on the proper use of this command to repair large file system structures, see Repairing Cylinders Support Utilities, Release

98 Chapter 3: Filer Utility (filer) BLK BLK Purpose The BLK (Block) command recognizes file system blocks and displays CI, WCI, DB, and WDB blocks based on the block code in the block header. For other types of blocks, the command indicates the block type. This command also displays DB or WDB blocks in Depot cylinders and displays snapshot MI chunks. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /X, /V) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /H /X /V /C Positions to the block, but does not display it. The display options: Short, Medium, Long, Hex, and Extended. The display format depends on the block type. For descriptions of how the display options are applied to the different block types, see the associated commands:ci, DB, WCI, and WDB. Validates the block according to the block type, but does not display the block. For compressed data blocks, displays the uncompressed DB header information, including extended header information pertaining to data compression. If the scoped DB is not compressed, the output is similar to the display when the /S option is used. 98 Support Utilities, Release 16.00

99 /Z cylid sectornum sectorcount = Usage Notes For compressed data blocks, displays: the uncompressed DB header information, including extended header information pertaining to data compression. a hex format display of the compressed data stored in the DB. If the /Z option is not used, the hex display shows the uncompressed data. If the scoped DB is not compressed, the output is similar to the output when the /x option is used. The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. The sector number of the entries to be displayed. The number of sectors to examine, starting from sectornum, but not extending beyond the end of the cylinder. This can be used for either normal or Depot sectors. All blocks within this range are displayed. Filer will use the most recently saved values for cylid, sectornum, or sectorcount. This option cannot be used unless input values have already been defined for cylid, sectornum, or sectorcount. For more information, see Stored Input Values. Chapter 3: Filer Utility (filer) BLK If the requested sector is in a segment in memory, BLK displays the segment from memory. If the segment is not in memory, BLK attempts to locate and display the sector from the file system B-Tree structures. If the requested sector is currently not in use in the file system, the requested sectors are read directly from disk and displayed in hexadecimal format. The DB command is similar to BLK, except it displays only Data Block segments. For more information, see DB. Support Utilities, Release

100 Chapter 3: Filer Utility (filer) CHECKSUM CHECKSUM Purpose The CHECKSUM command validates and displays the checksum for a DB, WDB, CI or WCI segment and optionally stores a recalculated checksum. If you specify the FIX option, the CHECKSUM command attempts to repair any checksum mismatches if detected. Syntax FIX DB CI WDB WCI cylid sectornum = Overwrites the old checksum with a newly calculated checksum if there was an error in the old checksum. The segment that contains the data rows saved on disk. The compound data structure that is saved on disk to describe the current state of a cylinder. The segment that contains the WAL log records and the TJ rows. The structure that identifies the WDBs that are contained on the cylinder. The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. The sector of the entries displayed. Filer will use the most recently saved values for cylid or sectornum. This option cannot be used unless input values have already been defined for cylid or sectornum. For more information, see Stored Input Values. 100 Support Utilities, Release 16.00

101 CI Chapter 3: Filer Utility (filer) CI Purpose The CI (Cylinder Index) command accesses the latest copy of the CI for the given cylinder and formats it in accordance with the display options. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /X, /H, /V) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /X /H /V Positions to this CI. If this cylinder is not a file system cylinder, an error occurs. Nothing is displayed unless a problem occurs. Shows a formatted display of the following: header and trailer. first and last DBDs. first and last SRDs. Shows a formatted display of the following: header and trailer. all of the SRDs, with each one followed by the first and last DBDs for that SRD. Shows a formatted display of the following: header and trailer. all of the SRDs, with each one followed by all of the DBDs for that SRD. the Free Sector (FSE) and the Unfree Sector (UNFSE) lists, if they exist. Shows the same display as /L, but also include the timestamp and other details from the DB for each DBD. Shows a hexadecimal dump of the requested CI. Verifies that the CI is logically correct, but does not display it. Support Utilities, Release

102 Chapter 3: Filer Utility (filer) CI FORCE FIX cylid = Usage Notes Formats as CI even if the block code indicates otherwise. Without FORCE, the command fails on a mismatched block. Filer will attempt automatic repair of the CI. You must supply the /V option with the FIX option, and the Teradata Database must be down. The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. Filer will use the most recently saved value for cylid. This option cannot be used unless input value has already been defined for cylid. For more information, see Stored Input Values. If the segment acquired is not a CI, then an error occurs unless you specify FORCE, in which case Filer will attempt to format the block as a CI. The CI command output displays additional columns for the CI checksum (checksum), checksum algorithm (ag), and checksum sample count (ct) fields in the CI header if the CI currently has an associated checksum. The checksum and checksum algorithm are in both the CID and the CI, providing both internal and external checksums. The count is only in the CI and is used for either an internal or an external check. In addition, the checksum, checksum algorithm, and checksum sample count fields contained in the DBD structure are displayed for each DBD listed. The addition of these three checksum fields to the DBD structure expands the width of the DBD display from 88 to 105 characters. The value of the checksum sample count field is a byte in the range This number corresponds to the number of 32-bit words to sample per disk sector in the file system block. The CI is stored on the first n sectors of each cylinder and describes only that cylinder. The CI has a redundant copy immediately following it on the disk. The CI can reside in one of two places: Disk DRAM The default access is the latest copy of the CI. The prompt for this selection is as follows: CI ==> The following commands are allowed at the CI prompt: Lower-level commands DELETE, DISPLAY, NEXT, PACK, PATCH, PREV, TRASH, AND WRITE. Although the current object selected can be the latest CI, the redundant CI, an SRD, or a DBD, the prompt reflects the current object. NEXT and PREV always access the latest copy of the next or previous CI. To get to the next or previous CI, when at an SRD or DBD object, type one of the following: CI =; NEXT CI =; PREV The display options for any of these commands are the same as that of the currently selected object, either the CI or the SRD, and makes no difference on a DBD. 102 Support Utilities, Release 16.00

103 Chapter 3: Filer Utility (filer) CI Upper-level commands that switch Filer to another object or level within the file system. Global commands, which include DATE, DISABLE, ENABLE, ERRORS, HELP, INPUT, OUTPUT, QUIT, RADIX, and SHOWDEFAULTS. Support Utilities, Release

104 Chapter 3: Filer Utility (filer) CID CID Purpose The Master Index is the highest level of the file system structure and contains a sorted list of all the Cylinder Index Descriptors (CIDs), which represent the cylinders in use. The CID command displays the CID associated with the requested cylinder. Syntax cylid = The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. Filer will use the most recently saved value for cylid. This option cannot be used unless an input value has already been defined for cylid. For more information, see Stored Input Values. Usage Notes The checksum and the checksum algorithm columns (checksum and ag) appear in both the CID and the CI displays if there are associated checksums, providing both internal and external checksums. The checksum sample count column (ct) appears only in the output of the CI command, and is used for either an internal or an external check. The prompt for this selector follows: MI:CID ==> In addition to global and upper-level commands, the following lower-level commands are allowed at the MI:CID prompt: DISPLAY DELETE NEXT PATCH PREV WRITE 104 Support Utilities, Release 16.00

105 CLEAREIOCACHE Chapter 3: Filer Utility (filer) CLEAREIOCACHE Purpose Clears the EIO cache. The EIO cache is an additional, fast, read-only I/O cache that exists between the PDE FSG cache and the disk storage subsystem on some platforms. Syntax Usage Notes EIO cache is used only for data from permanent tables. The EIO cache is cleared on a node-wide basis. The amount of cache cleared depends on the AMPs in the current Filer scope: If scope is set to one AMP, the EIO cache is cleared on the node containing that AMP. If the scope is set to all AMPs, the EIO cache is cleared on all nodes in the system. This cache does not exist on all platforms. If it does not exist, this command does nothing. Support Utilities, Release

106 Chapter 3: Filer Utility (filer) DATE/TIME DATE/TIME Purpose The DATE and TIME commands are synonyms. Both show the same display of current systemday, date, and time. Syntax Usage Notes The DATE or TIME command shows a timestamp in the following format: DDD mmm dd, yyyy HH:MM:SS Format DDD mmm dd, yyyy HH:MM:SS Specifies the abbreviated name of the day of the week. the abbreviated name of the calendar month, numeric day, and four-digit year. the current system time in hour, minutes, and seconds. Examples: DATE and TIME command outputs > date Thu Jul 02, :17:05 > time Thu Jul 02, :17: Support Utilities, Release 16.00

107 DB Chapter 3: Filer Utility (filer) DB Purpose The DB (Data Block) command selects and displays the DB segment that starts at the specified cylinder and sector number. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /X, /V) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /H /X /V Positions to the DB. If one does not exist, an error occurs. Nothing is displayed unless a problem exists. A formatted display of the block, header, block trailer, and first and last row header. A formatted display of the block, header, block trailer, first and last row header, and the intervening row headers. A formatted display of the block, header, block trailer, first and last row header, the intervening row headers, and the row content in hex following each row header. A formatted display of the DB header followed by the entire DB in hex. Same as /L. Verifies that the DB is logically correct but does not display it. Support Utilities, Release

108 Chapter 3: Filer Utility (filer) DB /C /Z FORCE cylid sectorcount = Usage Notes For compressed data blocks, displays the uncompressed DB header information, including extended header information pertaining to data compression. If the scoped DB is not compressed, the output is similar to the display when the /S option is used. For compressed data blocks, displays: the uncompressed DB header information, including extended header information pertaining to data compression. a hex format display of the compressed data stored in the DB. If the /Z option is not used, the hex display shows the uncompressed data. If the scoped DB is not compressed, the output is similar to the output when the /x option is used. Formats as DB even if the block code indicates otherwise. Without FORCE, the command fails on a mismatched block. The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. The sector count. Filer will use the most recently saved values for cylid, sectornum, or sectorcount. This option cannot be used unless input values have already been defined for cylid, sectornum, or sectorcount. For more information, see Stored Input Values. Since the DB command only displays parameters from the data block itself, this command does not display any checksum field values for the DB. Checksum field values for the DB are stored in the DBD in the CI. To display the checksum fields for a DB, either the DBD should be displayed using the DBD command or use the CHECKSUM command. A DB contains one or more rows, all of which are members of the same subtable. Any single row is contained fully within a single DB. In addition, every DB must be fully contained within a cylinder. If count is omitted, Filer reads in the first sector of the block and obtains the sector count from the DB header. If given, Filer attempts to display only the information contained in those sectors. If the segment accessed is not a DB, Filer displays a message indicating an assumed type, unless you specify FORCE, in which case Filer tries to format the block as a DB. The prompt for the DB selector is shown below: DB ==> In addition to global and upper-level commands, the following lower-level commands are allowed at the DB prompt: 108 Support Utilities, Release 16.00

109 ACCESS DBROW DISPLAY DELETE NEXT PACK PATCH PREV RENAME TRASH WRITE Chapter 3: Filer Utility (filer) DB The BLK command is similar to DB, except it can display other types of segments in addition to Data Blocks. For more information, see BLK. Support Utilities, Release

110 Chapter 3: Filer Utility (filer) DBD DBD Purpose The DBD (Data Block Descriptor) command displays the DBDs of the current cylinder. Syntax /H tid ridspec sectornum = Usage Notes Shows a formatted display of all the DBDs in the current CI followed by a display of the DBDs in hexadecimal. Shows only the DBDs from the subtable specified by tid for the current CI. For more information on tid formatting, see Subtable Identifiers (tids). Shows only the DBDs within the row(s) specified by ridspec within the subtable (tid) for the current CI. For information on ridspec formatting, see Rows and Row Ranges. The online help lists this option as rowspec. Shows only the DBDs covering this sector of the cylinder. Specifies that Filer should use the most recently saved value for sectornum. This option cannot be used unless an input value has already been defined for sectornum. For more information, see Stored Input Values. The DBD command must be invoked from a CI context (as indicated by the CI ==> prompt). Without arguments, DBD shows all DBDs in the current CI. 110 Support Utilities, Release 16.00

111 DBROW Chapter 3: Filer Utility (filer) DBROW Purpose The DBROW command displays the rows of a Data Block. Syntax Note: The online help lists the row options (/S, /M, /L, /H) as /rowopt. where: /S or /C /M or /I /L or /R /H Displays the row count without the block header. Displays the row headers. Displays the row headers, each followed by the formatted row. Displays the row headers, each followed by the row in hexadecimal. Usage Notes The DBROW command must be invoked from a DB context (as indicated by the DB> prompt). Support Utilities, Release

112 Chapter 3: Filer Utility (filer) DELETE DELETE Purpose The DELETE command deletes the currently selected object or range of objects, including all descendants and antecedents, meaning the delete cascades both up and down through the index structure. The DELETE command can damage the File System B*Tree such that data can be removed or lost permanently. If used incorrectly, changes might be irreversible. Syntax /Y Prevents Filer from prompting for confirmation before executing the command. Usage Notes The selected object or range is from the last operation and is saved in the context. Executing SHOWD displays the selected object or range. For more information, see SHOWDEFAULTS. The delete context can be a row, table, DB, DBD, CI, CID, WAL log record, WDB, WDBD, WCI, WCID, or DEPOT, and ranges thereof. If a WAL object is selected, DELETE cascades throughout the WAL log index structure. DELETE can apply to one, several or all AMPs, depending on the current scope. The DELETE confirmation scope can span AMPs but is condensed into a single range. For some ranges, DELETE is allowed only when the Teradata Database is down. If you delete a single DEPOT cylinder, the scope can be on one AMP or all AMPs. If you delete the entire DEPOT, the scope must be on all AMPs. This command is allowed only when the Teradata Database is offline. If there is a whereclause in the context, then only those rows or WAL log records that satisfy the whereclause are deleted. For more information on the whereclause, see Specifying a WHERE Clause. Only tables and rows can be deleted if running from remote DBW console with multiple vprocs selected. If the selected object is a CI, then the DELETE command effectively performs the same action as the INIT command. For more information, see INIT. For example, deleting a CI would remove the CID from the MI and all DBs with DBDs in the CI. Deleting Records from the WAL Log When DELETE is used to delete records in the WAL log that have been selected by using the WAL or WREC command qualified with a WHERE clause, you can use the INQUIRE and ABORT commands to check the status or halt the DELETE operation. The INQUIRE command has the following syntax: 112 Support Utilities, Release 16.00

113 Chapter 3: Filer Utility (filer) DELETE INQUIRE displays the status of the fastest and slowest vprocs, and minimum WAL log sequence number (WLSN) of all WLSNs that are currently being processed across all vprocs: > inquire Slowest vproc 2 is 30% done Fastest vproc 0 is 58% done DELETE is currently checking WLSN 0x B7D84A DELETE is about 43% done The ABORT command has the following syntax: After it is initiated, ABORT stops the DELETE process and reports the number of records deleted so far and the range of WAL log that has not been processed. When the abort is successful, the following appears: > abort Abort request has been sent Type ABORT to stop the command before completion Type INQUIRE to check on progress of command vproc 0 (0000) response (93772) record(s) deleted. WAL records from 0x D331E to 0x are not processed 1 of 1 vprocs responded with no messages or errors. WAL Record(s) DELETE Aborted Filer ==> Support Utilities, Release

114 Chapter 3: Filer Utility (filer) DEPOT DEPOT Purpose The DEPOT command displays cylinder summary information for cylinders in the Depot, or block summary information for blocks in a particular Depot cylinder. Syntax cylid = The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. Filer will use the most recently saved value for cylid. This option cannot be used unless an input value has already been defined for cylid. For more information, see Stored Input Values. Usage Notes The Teradata Database File System stages in-place writes of permanent and system tables through a disk area called the Depot, which is a collection of cylinders. In-place modifications are first written to the Depot and only then written to their original Home Disk Address (HDA). During system startup, the Depot is examined and any blocks that are found are read in, validated, and rewritten to their original HDA. This ensures that any blocks that might have been partially written to their HDA, when a system outage occurred, can be rewritten as complete blocks. Handling writes this way greatly reduces the possibility of a partial-write scenario destroying pre-existing data. Any blocks partially written to the Depot will not pass validation, and thus will not overwrite the previous version of the block that still exists at the HDA. Because a write to the Depot must be complete before the block is written to the HDA, an interruption can only corrupt one of the copies of the block. A failed write to the Depot can be recovered by applying the WAL log records to the copy of the block at the HDA. A failed write to the HDA can be recovered by copying the block in the Depot over the block at the HDA. The Depot can contain both DBs and WDBs. The contents of the Depot may be different each time you view it since data is constantly moving in and out of the Depot. 114 Support Utilities, Release 16.00

115 DISABLE Chapter 3: Filer Utility (filer) DISABLE Purpose The DISABLE command sets a specified flag in the file system to FALSE. This disables certain features of Filer. Syntax flag? Specifies the flag that will be set to false. The following flags are available: LOCKING When locking is disabled, Filer does not observe the file system locking protocols, which normally prevent system hangs and deadlocks. This flag should only be disabled when Teradata Database is in the Debug-Stop state. SCRIPT, SCRIPT MODE, and SCRIPTMODE When script mode is disabled, the Filer ABORT and INQUIRE commands are enabled. Script mode is disabled by default. Script mode should be enabled before running scripts that call Filer. The following flags are for internal use only, and should not be disabled or enabled: CHECK and CHECKTS PANIC STOP SUSPEND TPARESET TRACE Displays a complete list of available flags. Support Utilities, Release

116 Chapter 3: Filer Utility (filer) DISCARD DISCARD Purpose The DISCARD command discards the rebuilt CI or WCI that is currently in memory. Syntax Usage Notes This command can be used only from the REBUILDCI command prompt. 116 Support Utilities, Release 16.00

117 DISPLAY Chapter 3: Filer Utility (filer) DISPLAY Purpose The DISPLAY command shows a formatted display of the currently selected object or range of objects. If you specify an ordinalrange, then DISPLAY shows a related object or range. Syntax Note: You can type /dispopt and /rowopt at the beginning or end of the command. /dispopt /rowopt ordinalrange whereclause Usage Notes Specifies the format of the display output for the selected object. The display options are dependent on the object selected and may differ from those in the saved context. /dispopt can have the following values: P, S, M, L, X, H, Z Not all display options work with all object types. For descriptions of how the display options are applied to different object types, see the command for the selected object, such as CI, WCI, DB, WREC, etc. Specifies the format of the display output for the selected object. The row options are dependent on the object selected and may differ from those in the saved context. /rowopt can have the following values: P, S, C, M, I, L, R, H Not all row options work with all object types. For descriptions of how the row options are applied to different object types, see the command for the selected object, such as DBROW, ROW, WDBREC, etc. Used to establish a new context based upon the old context, and then DISPLAY shows the objects in the new context. The online help lists this option as rangeopt. For a WAL log record range context, a WHERE clause can be supplied to replace the current WHERE clause saved in the context. For more information, see Specifying a WHERE Clause. The DISPLAY command is based on the current context, which may be a row, subtable, CI, CID, DB, DBD, MI, WAL log record, WCI, WCID, WDB, WDBD, or WMI. Except for the MI and WMI contexts, DISPLAY can also be based on a range selection. Support Utilities, Release

118 Chapter 3: Filer Utility (filer) DISPLAY The current context is established by the previous BLK, CI, CID, DB, DBD, DISPLAY, MI, ROW, TABLE, WAL, WCI, WCID, WDB, WDBD, WMI, or WREC command. These commands save a single object or range of objects as the context. This saved context can be displayed using the SHOWDEFAULTS command. For more information, see SHOWDEFAULTS. If a prior command was a WAL command, the resulting context is a WAL log record or a range of WAL log records. For the WAL all command, the context begins with the first record in the WAL log and ends with the last record in the WAL log. This differs from the TABLE command where the resulting context is a subtable rather than a row range. You can use the ordinalrange argument of the DISPLAY command to expand the current context to a larger range of objects of the same type. Similarly, you can use ordinalrange to shrink a range to a smaller range or to a single object. The standalone integers (without the BEG/END prefix) in an ordinalrange always refer to objects outside the current context, so that positive values refer to objects past the end of the current context, and negative values refer to objects before the beginning of the current context, thus extending the context. For more information, see Specifying Display Ranges (ordinalrange). Note: This differs from the interpretation of an ordinalrange for the TABLE or WAL command, where positive integers without the BEG/END prefix are ordinal row/record numbers, so that 1 is the first row or record. Negative integers without the BEG/END prefix are backwards displacements from the last row or record. For more information, see TABLE and WAL. The DISPLAY command with an ordinalrange functions similarly for a WAL log record, WDB, WDBD, WCI, or WCID context, as it does for a Row, DB, DBD, CI, or CID context. Both singular and range contexts are allowed. If a prior command was a TABLE command which used a wildcard for the type part of the tid, the resultant context is special (nothing), and this context is not usable with the ordinalrange argument. For more information, see TABLE. The whereclause for the WAL and WREC commands is applied in conjunction with the ordinalrange or wlsnrange. The whereclause is recorded in the context and remains active for subsequent DISPLAY, NEXT, and PREV commands. As the display context is expanded or contracted, DISPLAY continues to use the whereclause in the context. The whereclause for a DISPLAY command must be appropriate to the saved context, or the DISPLAY command is rejected. For example, a whereclause is not allowed with a WCI context. For more information about the whereclause, see Specifying a WHERE Clause. 118 Support Utilities, Release 16.00

119 DROP Chapter 3: Filer Utility (filer) DROP Purpose The DROP command drops its access on the current object and returns to the default prompt. Syntax Usage Notes Filer always retains access to FIB and MI, even if the default prompt is showing. Support Utilities, Release

120 Chapter 3: Filer Utility (filer) DUMPDISK DUMPDISK Purpose The DUMPDISK command positions Filer at a specified disk block, bypassing the FSG cache, and displays the contents according to the selected display options. Syntax /P /S /M /L /H /V /C /Z cylid Positions to the specified disk segment, but does not display it. Shows a short formatted display of the disk segment. Shows a medium-length formatted display of the disk segment. Shows a long formatted display of the disk segment. Shows a hexadecimal display of the disk segment. Verifies the disk segment and displays any errors encountered. For compressed data blocks, displays the uncompressed DB header information, including extended header information pertaining to data compression. If the scoped DB is not compressed, the output is similar to the display when the /S option is used. For compressed data blocks, displays: the uncompressed DB header information, including extended header information pertaining to data compression. a hex format display of the compressed data stored in the DB. If the /Z option is not used, the hex display shows the uncompressed data. If the scoped DB is not compressed, the output is similar to the output when the /x option is used. The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. 120 Support Utilities, Release 16.00

121 sectornum sectorcount = The sector number. The number of sectors to display, starting from sectornum. If sectorcount is omitted, DUMPDISK will display one sector. Filer will use the most recently saved values for cylid, sectornum, or sectorcount. This option cannot be used unless input values have already been defined for cylid, sectornum, or sectorcount. For more information, see Stored Input Values. DIRECTSTORAGE Usage Notes Forces block to be read directly from disk subsystem, rather than from the EIO cache. Chapter 3: Filer Utility (filer) DUMPDISK The EIO cache is an additional, fast, read-only I/O cache that exists between the PDE FSG cache and the disk storage subsystem on some platforms. If the type is invalid, the block is displayed in the hexadecimal format. This command differs from the BLK command because FSG cache is always bypassed. Support Utilities, Release

122 Chapter 3: Filer Utility (filer) ENABLE ENABLE Purpose The ENABLE command sets a specified flag in the file system to TRUE. This enables certain features of Filer. Syntax flag? Specifies the flag that will be set to true. The following flags are available: LOCKING When locking is enabled, Filer observes the file system locking protocols, which prevent system hangs and deadlocks. Locking is enabled by default when Filer is started without the -l or -i command-line options. SCRIPT, SCRIPT MODE, and SCRIPTMODE Script mode is disabled by default. When script mode is enabled, the Filer ABORT and INQUIRE commands are disabled to allow a script to run without interruption. Script mode should be enabled before running scripts that call Filer. The following flags are for internal use only, and should not be enabled or disabled: PANIC STOP SUSPEND TPARESET TRACE Displays a complete list of available flags. 122 Support Utilities, Release 16.00

123 ERRORS Chapter 3: Filer Utility (filer) ERRORS Purpose The ERRORS command redirects diagnostic messages to a file that you specify or to the default file, stderr. Syntax TO INTO OVER file STDERR ME Usage Notes Specifies that Filer is to write diagnostic messages to a new file or to STDERR. If the file specified exists, Filer returns an error. Specifies that Filer is to append diagnostic messages to a specified file or to STDERR. If the file already exists, Filer appends the error messages to the end of the file If the file does not exist, Filer creates it automatically. Specifies that Filer is to overwrite an existing file or STDERR with current diagnostic messages. If the file already exists, Filer writes over the file If the file does not exist, Filer creates the file. The name of the destination file for diagnostic messages. The default file to which Filer writes diagnostic messages. A synonym for STDERR. When you start Filer, it writes diagnostic messages to STDERR by default. You can use the ERRORS command to redirect the diagnostic messages in the following ways: To write to a new file only To append an existing file To overwrite an existing file To display on your console If you include the file parameter in the ERRORS command, the file you specify becomes the destination for diagnostic messages redirected from STDERR. If you type the ERRORS command without any options, Filer shows the name of the current diagnostic messages file STDERR on your system console. Support Utilities, Release

124 Chapter 3: Filer Utility (filer) FIB FIB Purpose The FIB (File Information Block) command displays information about the file system and cylinder usage, including WAL log cylinder and Depot cylinder usage. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /H Positions to the FIB. Nothing is displayed unless a problem exists. Displays the fixed area of the FIB. Same as /S. Same as /M but also includes the Cylinder Allocation History Array and FTR information. Displays the contents of the FIB in hexadecimal. Usage Notes The prompt for the FIB selector is as follows: FIB ==> In addition to global and upper-level commands, the following lower-level commands are allowed at the FIB prompt: DISPLAY (without an ordinalrange) DROP PATCH WRITE 124 Support Utilities, Release 16.00

125 FIND Chapter 3: Filer Utility (filer) FIND Purpose The FIND command, given the subtable and optional RowID range, displays cylinders containing the range with detail, depending on the options. Syntax Note: The online help lists the display options (/S, /M, /L) as /dispopt. You can type the options at the beginning or end of the command. /S /M /L tid ridrange Displays the first CID, SRD, and DBD for range. Displays the first CID, SRD, and DBD for range, as well as all CIDs and the last SRD and DBD for range. Displays every CID, SRD, and DBD for range. the subtable of interest. For more information on tid formatting, see Subtable Identifiers (tids). the portion of the subtable of interest. For more information on ridrange formatting, see Rows and Row Ranges. The online help lists this option as rowspec [TO rowspec]. Usage Notes The range default is the entire subtable. FIND sets the context to null. If Filer is invoked from the Coroner mode of the debugger, gdb, the FIND command works in a limited manner, displaying whatever is found in the dump. For information on invoking Filer from the Coroner, see Running Filer from the System Debugger and Coroner. Support Utilities, Release

126 Chapter 3: Filer Utility (filer) FREECACHE FREECACHE Purpose The FREECACHE command displays the status of the free cylinder cache. Syntax Usage Notes By default, the cache contains entries for up to 100 cylinders. The maximum number of entries is determined by the Free Cylinder Cache Size setting in DBS Control. For more information on Free Cylinder Cache Size, see the DBS Control chapter of Utilities. 126 Support Utilities, Release 16.00

127 HELP Chapter 3: Filer Utility (filer) HELP Purpose The HELP command provides context sensitive help for Filer commands. The help displayed is based on the mode that Filer is running in as indicated by the Filer prompt. Syntax /L ALL keyword? Displays the command syntax and a description of the command. Without /L, HELP displays only the command syntax. Displays the help for all Filer commands, regardless of the current Filer context. Displays the help for the Filer command or parameter specified by keyword. If you do not specify either a keyword or ALL, HELP displays help only for the commands that are valid in the current context. Displays the list of all the keywords for which help is available. Support Utilities, Release

128 Chapter 3: Filer Utility (filer) IDENT IDENT Purpose The IDENT command identifies the objects in the specified sectors, and displays the properties of the objects. Syntax cylid sectornum sectorcount = Usage Notes The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. The sector number. The number of sectors to examine, starting from sectornum, but not extending beyond the end of the cylinder. Filer will use the most recently saved values for cylid, sectornum, or sectorcount. This option cannot be used unless input values have already been defined for cylid, sectornum, or sectorcount. For more information, see Stored Input Values. An object is identified even if the sector is not the first sector in that object. A description of all DBs or WDBs in the specified area is provided for both normal and Depot cylinders. Other objects or regions are identified but not described in detail. IDENT accepts any cylid and sector number. Therefore, in addition to the DB sectors, CI, FREE, MI, FIB, SEG0, etc. are also accepted and the owner identified. IDENT also displays the WAL log objects containing WAL log sectors. For example, for a sector in a data block, IDENT displays the SRD, DBD, database name, and table name. For a sector in the cylinder index, IDENT indicates that the sector is in the CI. Wherever the sector is, IDENT indicates the object, such as SEG0, FIB, MI, CI, DB, the cylinder s free sector list, WCI, WDB, or WAL log cylinder free list. The WMI is omitted because the WMI never is written to disk. If the storage has not yet been profiled, information on Grade and Temperature is not displayed. 128 Support Utilities, Release 16.00

129 INFO Chapter 3: Filer Utility (filer) INFO Purpose The INFO command displays information about the rebuilt CI or WCI that is currently in memory. Note: This command can be used only when Teradata Database is not running on the affected AMP. Syntax Usage Notes This command can be used only from the REBUILDCI command prompt. Support Utilities, Release

130 Chapter 3: Filer Utility (filer) INIT INIT Purpose Notice: The INIT command can damage the File System B*Tree such that data can be removed or lost permanently. If used incorrectly, any changes might be irreversible. The INIT command initializes the current cylinder or range of cylinders to free by deleting the contained DBs, releasing the cylinder back into the general storage pool, and removing the cylinder from the MI. Syntax /Y Prevents Filer from prompting for confirmation before executing the command. Usage Notes This command is not valid if the prompt reflects a range selection. The INIT command is allowed only at the CI prompt. INIT is the same as DELETE with a cylinder context. The command also initializes WAL log cylinders. For information on how to initialize or remove other objects, see DELETE. INIT displays a confirmation prompt before initializing the cylinder. 130 Support Utilities, Release 16.00

131 INPUT Chapter 3: Filer Utility (filer) INPUT Purpose The INPUT command directs Filer to read commands from a specified file rather than from the default input file, stdin. Syntax file The name of a file you specify as the source of command input to Filer. Usage Notes When you first start Filer, it accepts input from the stdin file by default. Using the INPUT command, you can redirect input from any file you specify. Note: The input file must be located on the node that runs the control AMP. To determine which node is running the control AMP, use the STATUS command of the Vproc Manager utility. The control AMP is designated under the Logical Configuration with an asterisk. When Filer reaches the end of the file you specify, Filer again accepts input from you. You can nest the INPUT command inside command files to a maximum of nine files deep. Example: Reading Filer commands from a file The following command example shows the format for redirecting input from a file in a specific directory: Filer ==> input from /home/user1/commands Support Utilities, Release

132 Chapter 3: Filer Utility (filer) MEMBLK MEMBLK Purpose The MEMBLK command accesses the file system segment at the specified memory address, determines the segment type, and returns a formatted display of the segment. Note: The MEMBLK command is available only when Filer is invoked from the System Debugger or Coroner. For information on invoking Filer from the System Debugger or Coroner see Running Filer from the System Debugger and Coroner. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /X) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /H /X /C Positions to the segment. Nothing is displayed unless there is a problem. The display options: Short, Medium, Long, Hex, and Extended. The display format depends on the segment type. For descriptions of how the display options are applied to the different segment types, see the associated commands:ci, DB, WCI, and WDB. For compressed data blocks, displays the uncompressed DB header information, including extended header information pertaining to data compression. If the scoped DB is not compressed, the output is similar to the display when the /S option is used. 132 Support Utilities, Release 16.00

133 /Z memaddr = Usage Notes For compressed data blocks, displays: the uncompressed DB header information, including extended header information pertaining to data compression. a hex format display of the compressed data stored in the DB. If the /Z option is not used, the hex display shows the uncompressed data. If the scoped DB is not compressed, the output is similar to the output when the /x option is used. The memory address specifying the start of the segment. Specifies that Filer should use the most recently saved value for memaddr. This option cannot be used unless an input value has already been defined for memaddr. For more information, see Stored Input Values. The MEMBLK command is similar to the BLK command. See BLK. Chapter 3: Filer Utility (filer) MEMBLK MEMBLK might access obsolete copies of segments, so the displays might not be current. In addition, sometimes segments are inaccessible from the System Debugger when run against a dump. Segments that are inaccessible within the debugger are also inaccessible within Filer. Support Utilities, Release

134 Chapter 3: Filer Utility (filer) MEMCTX MEMCTX Purpose The MEMCTX command returns a formatted display of a file system context structure (ctxcontext_t) beginning at the specified memory address. Note: The MEMCTX command is available only when Filer is invoked from the System Debugger or Coroner. For information on invoking Filer from the System Debugger or Coroner see Running Filer from the System Debugger and Coroner. Syntax memaddr = Usage Notes The memory address at which to begin the file system context structure display. Specifies that Filer should use the most recently saved value for memaddr. This option cannot be used unless an input value has already been defined for memaddr. For more information, see Stored Input Values. The MEMCTX command displays the internal file system context structure, ctxcontext_t. To display the saved Filer context information, use the SHOWD command. For more information about the SHOWD command, see SHOWDEFAULTS. 134 Support Utilities, Release 16.00

135 MEMDB Chapter 3: Filer Utility (filer) MEMDB Purpose The MEMDB command returns a formatted display of a data block (DB) beginning at the specified memory address. Note: The MEMDB command is available only when Filer is invoked from the System Debugger or Coroner. For information on invoking Filer from the System Debugger or Coroner, see Running Filer from the System Debugger and Coroner. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /X) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /H /X /C /Z FORCE memaddr = Positions to the DB. Nothing is displayed unless there is a problem. The options used to specify the format of the DB display. For detailed information about these display options, see DB. Attempts to format and display the segment as a DB even if the segment is not a DB. The memory address at which to begin the DB display. Specifies that Filer should use the most recently saved value for memaddr. This option cannot be used unless an input value has already been defined for memaddr. For more information, see Stored Input Values. Support Utilities, Release

136 Chapter 3: Filer Utility (filer) MEMDB Usage Notes The MEMDB command is similar to the DB command. See DB. MEMDB reports an error if the segment is not a DB. However, if you specify FORCE, Filer tries to display the memory block as a data block. MEMDB might access obsolete copies of segments, so the displays might not be current. In addition, sometimes segments are inaccessible from the System Debugger when run against a dump. Segments that are inaccessible within the debugger are also inaccessible within Filer. 136 Support Utilities, Release 16.00

137 MEMDUMP Chapter 3: Filer Utility (filer) MEMDUMP Purpose The MEMDUMP command displays a hexadecimal memory dump defined by a starting address and specified number of bytes. Note: The MEMDUMP command is available only when Filer is invoked from the System Debugger or Coroner. For information on invoking Filer from the System Debugger or Coroner, see Running Filer from the System Debugger and Coroner. Syntax memaddr length = Usage Notes The memory address at which to begin the memory dump. The amount of memory (in bytes) to be displayed. 256 is the default. Specifies that Filer should use the most recently saved values for memaddr or length. This option cannot be used unless the input values have already been defined for memaddr or length. For more information, see Stored Input Values. If length is omitted, 256 bytes are dumped. Support Utilities, Release

138 Chapter 3: Filer Utility (filer) MEMROW MEMROW Purpose The MEMROW command displays a row beginning at the specified memory address. Note: The MEMROW command is available only when Filer is invoked from the System Debugger or Coroner. For information on invoking Filer from the System Debugger or Coroner, see Running Filer from the System Debugger and Coroner. Syntax Note: The online help lists the row options (/P, /S, /M, /L, /H) as /rowopt. You can type the options at the beginning or end of the command. /P /S or /C /M or /I /L or /R /H memaddr = Positions to the physical row. Nothing is displayed unless there is a problem. Physical rows store one of several types of data, such as regular table row data or column partition data. Counts the number of physical rows selected. This is the default. Displays only the row header information (length, RowID, and flags). Formats the row header and the fields in the physical row. Formats the row header, and displays the entire physical row in hexadecimal, including the row header. The memory address at which to begin the row display. Specifies that Filer should use the most recently saved value for memaddr. This option cannot be used unless an input value has already been defined for memaddr. For more information, see Stored Input Values. 138 Support Utilities, Release 16.00

139 MEMSORCTX Chapter 3: Filer Utility (filer) MEMSORCTX Purpose The MEMSORCTX command returns a formatted display of a context structure (sorcontext_t) beginning at the specified memory address. Note: The MEMSORCTX command is available only when Filer is invoked from the System Debugger or Coroner. For information on invoking Filer from the System Debugger or Coroner, see Running Filer from the System Debugger and Coroner. Syntax memaddr = The memory address at which to begin the context structure display. Specifies that Filer should use the most recently saved value for memaddr. This option cannot be used unless an input value has already been defined for memaddr. For more information, see Stored Input Values. Support Utilities, Release

140 Chapter 3: Filer Utility (filer) MEMWCTX MEMWCTX Purpose The MEMWCTX command returns a formatted display of a context structure (wtjcontext_t) beginning at the specified memory address. Note: The MEMWCTX command is available only when Filer is invoked from the System Debugger or Coroner. For information on invoking Filer from the System Debugger or Coroner see Running Filer from the System Debugger and Coroner. Syntax memaddr = The memory address at which to begin the context structure display. Specifies that Filer should use the most recently saved value for memaddr. This option cannot be used unless an input value has already been defined for memaddr. For more information, see Stored Input Values. 140 Support Utilities, Release 16.00

141 MEMWDB Chapter 3: Filer Utility (filer) MEMWDB Purpose The MEMWDB command returns a formatted display of a WAL data block (WDB) beginning at the specified memory address. It is similar to the WDB command. See WDB. Note: The MEMWDB command is available only when Filer is invoked from the System Debugger or Coroner. For information on invoking Filer from the System Debugger or Coroner, see Running Filer from the System Debugger and Coroner. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /X) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /H /X FORCE memaddr = The options used to specify the format of the WDB display. For detailed information about these display options, see WDB. Attempts to format and display the segment as a WDB even if the block code does not indicate a WDB. The memory address at which to begin the WDB display. Specifies that Filer should use the most recently saved value for memaddr. This option cannot be used unless an input value has already been defined for memaddr. For more information, see Stored Input Values. Support Utilities, Release

142 Chapter 3: Filer Utility (filer) MEMWREC MEMWREC Purpose The MEMWREC command formats a WAL log record beginning at the specified memory address. Note: The MEMWREC command is available only when Filer is invoked from the System Debugger or Coroner. For information on invoking Filer from the System Debugger or Coroner, see Running Filer from the System Debugger and Coroner. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /X) as /nextopt. You can type the options at the beginning or end of the command. /P /S /M /L /H /X memaddr = Positions to the WAL log record. Nothing is displayed unless there is a problem. The options used to specify the format of the WAL log record. For detailed information about these display options, see WREC. The memory address specifying the start of the WAL log record. Specifies that Filer should use the most recently saved value for memaddr. This option cannot be used unless an input value has already been defined for memaddr. For more information, see Stored Input Values. 142 Support Utilities, Release 16.00

143 MI Chapter 3: Filer Utility (filer) MI Purpose The Master Index is the highest level of the B-Tree structure of the file system and contains a sorted list of all the CI descriptors. The MI command displays the following: The Master Index with the requested display option The checksum contained in the CID for each one that is displayed with the command Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /V, /X) as /dispopt. /P /S /M /L /H /V /X Positions to the MI. Nothing is displayed unless a problem exists. Shows the fixed area and a formatted display of the first and last CID of the MI segment. Same as /L. Shows the fixed area and a formatted display of all the CIDs in the MI segment. Displays a hex dump of the MI segment. Verifies that the MI segment has all CIDs in the correct order, that all sectors are accounted for, and matches the setting of the FIB. Same as /L. Usage Notes The prompt for the MI selector is shown below: MI ==> Support Utilities, Release

144 Chapter 3: Filer Utility (filer) MI In addition to global and upper-level commands, the following lower-level commands are allowed at the MI prompt: DISPLAY (without an ordinalrange) DROP PATCH WRITE (if the database is not running on the AMP in the current scope) No checksum exists for the MI itself, however, if checksums are enabled, CIDs with checksums will show additional checksum columns in the MI output screen. 144 Support Utilities, Release 16.00

145 MODIFY REDO Chapter 3: Filer Utility (filer) MODIFY REDO Purpose The MODIFY REDO command replays or performs Redo processing on specified WAL log records. The command also determines whether to perform WAL replay during the next file system startup. Improper use of this command can lead to irreparable data loss. Syntax /Y REPLAY LINEAGE wlsn REPLAY WREC Prevents Filer from prompting for confirmation before executing the command. Replays the records in the WAL log with the lineage value specified by wlsn. Filer will not perform any WAL log analysis before replaying the records. Therefore, be absolutely sure that the specified records must be replayed before using this option. An attempt to replay an obsolete lineage may result in data loss. The lineage number that identifies the Redo records to be replayed by the REPLAY LINEAGE option. Replays the WAL log record or range of records specified in the wlsnrange. Only modifications from complete NTAs are committed. If a range of records is specified, Filer will attempt to replay closed NTAs and will skip or discard modifications related to open NTAs in the portion of the log specified. Support Utilities, Release

146 Chapter 3: Filer Utility (filer) MODIFY REDO Note: wlsnrange REPLAY WREC cannot be used if Filer is scoped to multiple AMPs. Filer will not perform any WAL log analysis before replaying the records. Therefore, be absolutely sure that the specified records must be replayed before using this option. An attempt to replay one or more obsolete records may result in data loss. Selects one or a range of WAL log records based on the WAL log sequence number (WLSN) in the record header. To define wlsnrange, use this syntax: ACTIVATE Replays the WAL log records during the next file system startup. Filer scans the WAL log for the most recent checkpoint and restores that checkpoint as the basis for the next WAL replay. DEACTIVATE Usage Notes Disables the replay of WAL log records during the next file system startup. This option is useful in cases where so many repairs have been made to the file system that attempting to replay WAL log records during the next file system startup will likely fail. The MODIFY REDO command is useful for repairing systems that encounter problems when attempting to replay WAL log records. MODIFY REDO can be run when Filer is started explicitly on a down (FATAL/OFFLINE status) AMP or when the entire Teradata Database system is down. Lineage refers to a chain of modifications that occurs on a particular block while the block is in memory. A lineage chain begins when the first modification using WAL is made to an unmodified block in memory. The chain dies when the block is written to disk with the accumulated changes. To track a particular lineage, a lineage number is assigned to the block. The MODIFY REDO REPLAY LINEAGE command allows you to process Redo records belonging to a particular lineage. When a transaction requires one or more changes to multiple blocks, the modifications are tied together via an NTA (Nested Top Action). An NTA joins a series of WAL log records into a single atomic unit. All records participating in a particular NTA must be written completely to the WAL log to be complete. The MODIFY REDO REPLAY WREC command will not process incomplete NTAs. Example: Replaying WAL log records from a specified WLSN range The following example replays the WAL log records in the WLSN range from 00491D2 to D4. Since the /Y option is not specified, Filer prompts for confirmation before executing the command. WREC ==> modify redo replay wrec 00491D2 to D4 You have chosen to attempt the replay of all records between D2 and D4 inclusive. 146 Support Utilities, Release 16.00

147 WARNING: This command performs the replay of the specified WAL records WITHOUT performing the WAL Analysis. Erroneous use of this command could potentially lead to irrepairable data loss. Please be sure this is the desired action before proceeding. Are you sure you want to continue? (Y/N) y Command has been sent to Slave tasks. vproc 0 (0000) response SKIPPED : Filer skipped WAL record with wlsn D2 Record already applied. SKIPPED : Filer skipped WAL record with wlsn D3 Record already applied. SKIPPED : Filer skipped WAL record with wlsn D4 Record already applied. Filer attempted to replay a total of 3 record(s). Chapter 3: Filer Utility (filer) MODIFY REDO Example: Replaying WAL log records of a specified lineage The following example replays all of the WAL log records with a lineage value of Filer ==> modify /y redo replay lin You have chosen to attempt the replay of all records with a linwlsn of Command has been sent to Slave tasks. vproc 0 (0000) response REPLAYED: Filer replayed WAL record with wlsn REPLAYED: Filer replayed WAL record with wlsn REPLAYED: Filer replayed WAL record with wlsn REPLAYED: Filer replayed WAL record with wlsn REPLAYED: Filer replayed WAL record with wlsn REPLAYED: Filer replayed WAL record with wlsn BF Filer attempted to replay a total of 6 record(s). Example: Activating replay of WAL log records during the next file system startup The following example activates the replay of WAL log records in the next file system startup. Filer ==> modify redo activate next redo You have chosen to enable the WAL replay on the next File System startup. Filer will search from the end of the WAL log for the latest checkpoint and utilize that checkpoint for the next startup. Are you sure you want to continue? (Y/N) y Command has been sent to Slave tasks. vproc 0 (0000) response The checkwlsn in segment zero has been updated to E5. The chkptredowlsn in segment zero has been updated to D8. The maxdeletablewlsn in segment zero has been updated to D8. Support Utilities, Release

148 Chapter 3: Filer Utility (filer) MODIFY REDO Example: Deactivating replay of WAL log records during the next file system startup The following example disables the replay of WAL log records during the next file system startup. The command has been scoped to four AMPs. Filer ==> modify redo deactivate next redo You have chosen to disable the WAL replay on the next File System startup. Are you sure you want to continue? (Y/N) > y y vproc 0 (0000) response The WAL replay has been deactivated for the next File System startup. To undo this command, you can issue the following: MODIFY REDO ACTIVATE NEXT vproc 1 (0001) response The WAL replay has been deactivated for the next File System startup. To undo this command, you can issue the following: MODIFY REDO ACTIVATE NEXT vproc 2 (0002) response The WAL replay has been deactivated for the next File System startup. To undo this command, you can issue the following: MODIFY REDO ACTIVATE NEXT vproc 3 (0003) response The WAL replay has been deactivated for the next File System startup. To undo this command, you can issue the following: MODIFY REDO ACTIVATE NEXT 148 Support Utilities, Release 16.00

149 NEXT Chapter 3: Filer Utility (filer) NEXT Purpose The NEXT command displays the next row, WAL log record, DB, WDB, DBD, WDBD, CI, WCI, CID, WCID or subtable. Syntax Note: You can type /dispopt or /rowopt at the beginning or end of the command. /dispopt /rowopt number Can be one of the following values: /S displays a short formatted display of the object. This is the default. /M displays a medium-length formatted display of the object. /L displays a long formatted display of the object. /X displays an extended display of the object. /H displays a hexadecimal display of the object. /Z displays for compressed values: the uncompressed DB header information, including extended header information pertaining to data compression. a hex format display of the compressed data stored in the DB. If the /Z option is not used, the hex display shows the uncompressed data. If the scoped DB is not compressed, the output is similar to the output when the /X option is used. Can be one of the following values: /P positions to the physical rows or table requested. At least one physical row must be in the selected range. Nothing is displayed unless there is a problem. Physical rows store one of several types of data, such as regular table row data or column partition data. /C or /S counts the number of physical rows selected. This is the default for all subtables except the table header subtable. /I or /M displays only the row header information (length, RowID, and flags). /L or /R display the formatted row header and the fields in the physical row. This option is the default for the table header subtable. However, this option will not be the default formatted field display because the table header subtable is a special row format and requires special formatting. /H displays the entire physical row in hexadecimal, including the row header. The number of objects to display, starting with the first object following the end of the current context. The default is 1. Support Utilities, Release

150 Chapter 3: Filer Utility (filer) NEXT Usage Notes If number is specified, NEXT displays subsequent objects of the same type until the requested number is reached. NEXT sets the context to the objects displayed. If a whereclause is saved in the context, NEXT will apply the conditions of the whereclause. For more information about the whereclause, see DISPLAY and Specifying a WHERE Clause. NEXT displays objects which follow the end of the current context as shown in the following table. From row context table context (whether a subset of rows was displayed or ALL) DB context DBD context CI context CID context WAL log record context WDB context WDBD context WCI context WCID context NEXT displays next rows within the table. next subtables. next DBs. next DBDs. next CIs. next CIDs. next WAL log records. next WDBs. next WDBDs. next WCIs. next WCIDs. 150 Support Utilities, Release 16.00

151 OUTPUT Chapter 3: Filer Utility (filer) OUTPUT Purpose The OUTPUT command redirects Filer output to a file you specify or to the default file, stdout. Syntax TO INTO OVER file STDOUT ME Usage Notes Specifies that Filer is to redirect output to a new file or to STDOUT. If the file exists, Filer returns an error. Specifies that Filer is to append output to an existing file specified by file or to STDOUT. If the file exists, Filer appends the output to the end of the file. If the file does not exist, Filer creates the file. Specifies that Filer is to overwrite an existing file or STDOUT with new Filer output. If the file exists, Filer writes over the file. If the file does not exist, Filer creates the file. The name of a file you specify as the destination of Filer output. The default file to which Filer writes output. A synonym for STDOUT. When you first start Filer, diagnostic messages are written to STDOUT by default. You can use the OUTPUT command to redirect Filer output in any of the following ways: To write to a new file only To append an existing file To overwrite an existing file To display on your console Note: The output file is located on the node that runs the control AMP. To determine which node is running the control AMP, use the STATUS command of the Vproc Manager utility. The control AMP is designated under the Logical Configuration with an asterisk. Support Utilities, Release

152 Chapter 3: Filer Utility (filer) OUTPUT When Filer redirects output to a file, all input and diagnostic messages are echoed to the output file as well as to their usual destinations. If you include the file parameter in the OUTPUT command, Filer uses that parameter as the destination for output redirected from STDOUT. If you type the OUTPUT command without any options, Filer displays the name of the current output file STDOUT to your system console. 152 Support Utilities, Release 16.00

153 PACK Chapter 3: Filer Utility (filer) PACK Purpose The PACK command packs the segment to eliminate fragmentation. Syntax /Y Prevents Filer from prompting for confirmation before executing the command. Usage Notes Currently, only a CI and DB can be packed. Support Utilities, Release

154 Chapter 3: Filer Utility (filer) PATCH PATCH Purpose The PATCH command does the following: Modifies file system segments while the segment is in memory, whether the segment is a CI, a DB, the FIB, the MI, a row, or a WAL log segment or object. The modified segments can later be forced to disk with the WRITE command. The PATCH command can also be used with interior file system objects, such as a CID, SRD, DBD, WCID, and WDBD. Automatically updates the checksum of the segment being modified when the modified segment is written. Syntax Note: The online help lists the patch options (/B, /C, /W, /S, /D, /L) as /patchopt. The syntax diagram only shows the options at the beginning of the command, but you can type the options at the beginning or end of the command. where the following tells how to address the segment being patched: /B or /C /W or /S /D or /L offset Usage Notes Input is one byte at a time. You can type input as a number (0x0 to 0xFF) or a character string delimited by apostrophes or double quotation marks (a character is a string of length 1). Input is one numeric Half Word (Short Integer, two bytes) at a time (0x0 to 0xFFFF). This is the default. Input is one numeric Word (Long Integer, four bytes) at a time (0x0 to 0xFFFFFFFF). the first location to be patched is specified as offset bytes from the beginning of the current selected object. In most cases, the current selected object is reflected by the prompt, but in some cases, this might not be true. For example, even though the prompt shows Row==>, the last selected object can be one row or a range of rows. With a range of rows, the offsetis displaced from the beginning of the first row in the range. The PATCH command syntax is the same, regardless of the current input prompt, except when you are in Patch Mode already, in which case offset is optional. If omitted, offset defaults to the current location. 154 Support Utilities, Release 16.00

155 If you type an invalid command, an error message is displayed, and the command is ignored. Chapter 3: Filer Utility (filer) PATCH If you use PATCH from a CI or DB prompt, the snapshot MI is invalidated, forcing the system to read CIs to rebuild the MI at the next file system startup. When Filer is started on an AMP with the database down, patching at the MI prompt will result in the patched MI being written to the snapshot MI location on disk. This patched MI copy will be used to start the next instance of the file system. Notice: Do not attempt to patch the snapshot MI. Instead, use the SNAPMI INVALIDATE command to force the MI to be rebuilt from the CIs at the next file system startup. When you type the PATCH command, it must be the only command on the command line because the command switches from general command syntax to a specific command syntax. Therefore, both the following command entries will generate an error: CI /P 200; PATCH /W 1567H PATCH /W 1567H;WRITE After you issue the PATCH command, the prompt changes to reflect that Filer has entered Patch Mode, and shows the object and address being patched. The current value of the cell open for patching is displayed in both HEX and DECIMAL. Prompting continues until you issue the DONE or CANCEL command, at which time the prompt is restored to what it was before patching started. The following table describes what happens when exiting Patch Mode. IF you exit Patch Mode using DONE CANCEL THEN changes are saved to the segment and other tasks will see the changes immediately. changes are not saved and the segment will be left exactly the same as when you entered Patch Mode. Because the checksum is stored externally, patching a single file system segment causes an additional segment to be updated with the new checksum of the segment if checksums are enabled on the block. PATCH Commands In Patch Mode, the following commands are allowed: patch_replacement BACK CANCEL DONE PATCH REPEAT SKIP Global Filer Commands patch_replacement Replaces the contents of the current cell with the specified input. Support Utilities, Release

156 Chapter 3: Filer Utility (filer) PATCH Syntax Element number string SKIP or SK Description For a description of numeric entries, see Numeric Input. A list of characters delimited by apostrophes or double quotation marks (only valid in /Bor /C mode). A closing delimiter is required. Two consecutive delimiter characters translates to a single character of the same type. Skips the contents of the current cell and moves to the next cell. BACK Causes PATCH not to change the value of the current cell, and instead move back a specified number of cells. Syntax Element number Description The number of cells to move back. CANCEL Exits from Patch Mode without saving any changes that were made. Leaves the segment exactly as it was when Patch Mode was started. DONE Exits from Patch Mode, saving the changes that were made to the segment. Because the segment is now patched, modifications are visible to the outside world. 156 Support Utilities, Release 16.00

157 Chapter 3: Filer Utility (filer) PATCH Syntax Element QUIT, Q, STOP, ST, END, EXIT /NOCHECKSUM Description Synonyms for DONE. If checksums are enabled on the segment being modified or written, the checksum is not updated and written to the external segment that contains the checksum. The /NOCHECKSUM option is not available with any of the DONE synonyms. PATCH The PATCH command itself, using the normal PATCH syntax, can be used at the Patch Mode prompt to change the current patch options. The offsetis counted from the start of the current selected entity. If offsetis omitted, the current location is not changed. REPEAT Repeats the patch_replacement a specified number of times, or until the end of segment is reached. Syntax Element ALL or * number Description Repeat the patch_replacement operation for every cell until end of the current segment. Specifies how many times to repeat the patch_replacement operation. SKIP Moves to the next cell without making any changes in the current cell. Global Filer Commands All global Filer commands (except QUIT) behave exactly as they do in regular Filer Command Mode. For additional information, see Filer Command Types. Example: Using the Filer PATCH command DB ==> patch /b 0 Now entering PatchMode Patch: DB : 0000 : 0000 ( 0) ==> 2 Patch: DB : 0001 : 0008 ( 8) ==> 4 Patch: DB : 0002 : 0006 ( 6) ==> sk Patch: DB : 0003 : 00f0 ( 240) ==> done Exiting Patch Mode DB ==>> Support Utilities, Release

158 Chapter 3: Filer Utility (filer) PREV PREV Purpose The PREV command displays a previous row, WAL log record, DB, WDB, DBD, WDBD, CI, WCI, CID, WCID or subtable. Syntax Note: You can type /dispopt or /rowopt at the beginning or end of the command. /dispopt /rowopt number Can be one of the following values: /S displays a short formatted display of the object. This is the default. /M displays a medium-length formatted display of the object. /L displays a long formatted display of the object. /H displays a hexadecimal display of the object. /X displays an extended display of the object. /Z displays for compressed values: the uncompressed DB header information, including extended header information pertaining to data compression. a hex format display of the compressed data stored in the DB. If the /Z option is not used, the hex display shows the uncompressed data. If the scoped DB is not compressed, the output is similar to the output when the /X option is used. Can be one of the following values: /P positions to the physical rows or table requested. At least one physical row must be in the selected range. Nothing is displayed unless there is a problem. Physical rows store one of several types of data, such as regular table row data or column partition data. /C or /S counts the number of physical rows selected. This is the default for all subtables except the table header subtable. /I or /M displays only the row header information (length, RowID, and flags). /L or /R display the formatted row header and the fields in the physical row. This option is the default for the table header subtable. However, this option will not be the default formatted field display because the table header subtable is a special row format and requires special formatting. /H displays the entire physical row in hexadecimal, including the row header. The number of objects to display, starting with the first object following the end of the current context. The default is Support Utilities, Release 16.00

159 Usage Notes Chapter 3: Filer Utility (filer) PREV If number is specified, PREV displays previous objects of the same type until the requested number is reached. PREV sets the context to the objects displayed. If a whereclause is saved in the context, PREV will apply the conditions of the whereclause. For more information about the whereclause, see DISPLAY and Specifying a WHERE Clause. PREV displays the objects previous to the beginning of the current context as shown in the following table. From row context table context (whether a subset of rows was displayed or ALL) DB context DBD context CI context CID context WAL log record context WDB context WDBD context WCI context WCID context PREV displays previous rows within the table. previous subtables. previous DBs. previous DBDs. previous CIs. previous CIDs. previous WAL log records. previous WDBs. previous WDBDs. previous WCIs. previous WCIDs. Support Utilities, Release

160 Chapter 3: Filer Utility (filer) PRIORITY PRIORITY Purpose The PRIORITY command allows you to set the priority of the Filer process. Syntax priorityclass Usage Notes the priority of the Filer process. Valid values for priorityclass are: LOW, L, or 0 MEDIUM, M, or 1 (This is the default.) HIGH, H, or 2 RUSH, R, or 3 The values are not case-sensitive. The PRIORITY command is most commonly used with the SCANDISK command. 160 Support Utilities, Release 16.00

161 QUARLIST Chapter 3: Filer Utility (filer) QUARLIST Purpose The QUARLIST command displays a list of cylinders whose owning AMPs cannot be determined. Cylinders are designated for quarantine automatically when a data corruption in the cylinder index prevents the system from determining the AMP to which the cylinder belongs. A list of quarantined cylinders is maintained on a single AMP in each storage subpool, usually the AMP with the lowest vproc number. AMPs that can potentially own these cylinders are kept down, and the file system is not started on these AMPs during Teradata Database system startup. To bring the AMPs back up, use Filer to repair or reinitialize the quarantined cylinders. The QUARLIST command can scan these cylinders for possible diagnoses. After the quarantined cylinders have been fixed, use the ASSIGN command to reassign the cylinders to their appropriate AMPs, or use the /R option of QUARLIST to remove the cylinder from the list, making the cylinder available for use from the general pool of storage cylinders. For more information on the ASSIGN command, see ASSIGN. Syntax cylid /P /S /R /X cylid The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. Prints the list of quarantined cylinders. Scans a quarantined cylinder and attempts to diagnose problem. If no cylid is specified, scans all quarantined cylinders. Initializes a quarantined cylinder and releases it back into the general storage pool. If no cylid is specified, initializes and releases all quarantined cylinders. Initializes the cylinder specified by cylid, but does not release it back to storage. Used for durable cylinders, which should never be released to general storage, such as those used for master indexes and startup information. Usage Notes After repairing a cylinder on the quarantine list, use the ASSIGN command to re-associate the cylinder with its AMP. For more information, see ASSIGN. Support Utilities, Release

162 Chapter 3: Filer Utility (filer) QUARLIST The /R option is related to the ASSIGN command. The /R option releases a repaired cylinder back into the general storage pool, while ASSIGN re-associates a repaired cylinder back to a specific AMP. For more information on the proper use of this command to repair large file system structures, see Repairing Cylinders 162 Support Utilities, Release 16.00

163 QUIT Chapter 3: Filer Utility (filer) QUIT Purpose The QUIT command ends a Filer session. Syntax Usage Notes STOP, END, and EXIT are synonyms for the QUIT command. Support Utilities, Release

164 Chapter 3: Filer Utility (filer) RADIX RADIX Purpose The RADIX command displays or sets the default radix used as the numeric base for data input to and output from Filer. as either hexadecimal or decimal. Syntax INPUT OUTPUT HEX DEC Usage Notes Numeric input to Filer defaults to the radix you select, either hexadecimal or decimal. Numeric output from Filer defaults to the radix you select, either hexadecimal or decimal. Numeric input to or output from Filer defaults to a radix of hexadecimal. Numeric input to or output from Filer defaults to a radix of decimal. When you start Filer, the default radix for both input and output from Filer is hexadecimal. If you omit both the INPUT and OUTPUT options from the RADIX command, the radix (HEX or DEC) that you select applies to both numeric input and output. You can select either INPUT or OUTPUT, but not both when changing the settings of Filer data. If you omit both the HEX and DEC options, Filer displays the current RADIX setting. 164 Support Utilities, Release 16.00

165 REALLOCDUR Chapter 3: Filer Utility (filer) REALLOCDUR Purpose The REALLOCDUR command reallocates cylinders for a specified durable cylinder type. Durable cylinders are special cylinders used by the file system itself, such as cylinders used to store the master index (MI) and the file system trace buffer (FTR). Durable cylinders are not available to be allocated for general storage. Use REALLOCDUR when durable cylinders have been lost, or cannot otherwise be repaired. This command allocates new durable cylinders for these structures. If cylinders which used to be part of the durable cylinder remain on the bad or quarantine lists, they should be freed by using the BADLIST /R or QUARLIST /R command, as appropriate. For more information on these commands, see BADLIST, and QUARLIST. Note: This command can be used only when Teradata Database is not running on the affected AMP. Syntax cyltype Specifies the type of durable cylinders to be allocated. cyltype can be one of the following: MICYL Specifies that durable cylinders for the master index be reallocated. FTRCYL Specifies that durable cylinders for the trace buffers be reallocated. Usage Notes Use this command when there are multiple large durable cylinders in the quarantine list. For more information, see QUARLIST. For more information on the proper use of this command to repair large file system structures, see Repairing Cylinders. Support Utilities, Release

166 Chapter 3: Filer Utility (filer) REBUILDCI REBUILDCI Purpose The REBUILDCI command reconstructs a cylinder index from the sectors in the data area of the cylinder. This command works for both regular cylinder indexes (CIs) and WAL cylinder indexes (WCIs). REBUILDCI can be used when one or both copies of a CI or WCI have been corrupted. Note: This command can be used only when Teradata Database is not running on the affected AMP. Freed sectors that contain obsolete data from previously removed tables and rows can be restored to the index by the REBUILDCI command. Be aware of this possibility, and inspect rebuilt cylinders before writing them to disk. Syntax Usage Notes The REBUILDCI command can be used only when Filer scope is set to a single vproc. The index for the specified cylinder is rebuilt in memory, but is not committed to disk until the WRITE command is issued. A rebuilt index in memory must be written to disk or discarded before changing the scope or quitting Filer. Running this command switches Filer to the REBUILDCI mode, indicated by the REBUILDCI command prompt. In this mode, the INFO, REPAIRSECTS, and DISCARD commands can be used to inspect and modify the rebuilt index before it is written to disk. Example: Using the Filer REBUILDCI command Filer ==> rebuildci C REBUILDCI is not allowed if more than one amp is selected Filer ==> scope 1 1 of 2 amps have been selected Filer ==> rebuildci C *** Note: Rebuilding a cylinder s index could restore cylinder data that was previously deleted or replaced. *** Rebuild of CI C has been completed Found 4 PERM DB(s) representing 971 sector(s) 166 Support Utilities, Release 16.00

167 Chapter 3: Filer Utility (filer) REBUILDCI REBUILDCI ==> quit There is a rebuilt CI in memory. You must write or discard the rebuilt [W]CI before changing the scope or quitting from Filer. REBUILDCI ==> write WARNING!!: Writing the rebuilt CI will overwrite the existing CI on disk. Are you sure you want to WRITE this?? (Y/N) y Rebuilt [W]CI has been written to disk FILER ==> quit Filer has exited Support Utilities, Release

168 Chapter 3: Filer Utility (filer) REBUILDMAP REBUILDMAP Purpose The REBUILDMAP command rebuilds the cylinder allocation map on a storage device after the block code or cylinder ID control fields of a cylinder index have been manually changed or patched. Such changes will not be effective until the allocation map has been rebuilt. Rebuilding the map requires a Teradata Database restart, which is performed automatically by REBUILDMAP. Note: REBUILDMAP can be used only when Teradata Database is not running, and Filer scope is set to a single AMP vproc. For more information on scope, see SCOPE (or VPROC). Notice: Rebuilding the allocation map on a device destroys the existing allocation map. For more information on the proper use of this command to repair large file system structures, see Repairing Cylinders. Syntax cylid Specifies the original cylinder ID of the modified cylinder. cylid is the 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. 168 Support Utilities, Release 16.00

169 RENAME Chapter 3: Filer Utility (filer) RENAME Purpose The RENAME command renames the currently selected segment to the specified location on the disk. Syntax /Y cylid sectornum = Usage Notes Prevents Filer from prompting for confirmation before executing the command. The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. The sector number. Filer will use the most recently saved values for cylid or sectornum. This option cannot be used unless input values have already been defined for cylid or sectornum. For more information, see Stored Input Values. RENAME makes a copy of the object in a new location. The original object remains in the original location. The corresponding DBD or CID must be updated to reflect the new location of the object. This might involve deleting the DBD or CID which is pointing to the original copy of the object and creating a new DBD or CID to point to the object in its new location. Support Utilities, Release

170 Chapter 3: Filer Utility (filer) REPAIRSECTS REPAIRSECTS Purpose The REPAIRSECTS command helps to resolve overlapping DBs on rebuilt CIs, and overlapping WDBs on rebuilt WCIs. Syntax Usage Notes This command can be used only from the REBUILDCI command prompt. Example: Using the Filer REPAIRSECTS command Filer ==> info Command has been sent to Slave tasks. vproc 0 (0000) response REBUILDCI has rebuilt CI C on this vproc. The rebuilt CI has not been written to disk. The DBs listed below have overlapping rowids: Start Sectors TableID FirstrowID Lastrowkey 02DE * F6 D26A *0000 9BFB F9E *0000 9BFA *0000 9FFD BFFC The DBs listed below have overlapping sectors: Start Sectors TableID FirstrowID Lastrowkey *00FC * BF7 3EA FF3 96E0 *0140 * FF3 FB5F F2 F87B *0185 * F F6 77EE Rebuilt [W]CI has overlapping [W]DBs. Run REPAIRSECTS command to resolve overlapping [W]DBs REBUILDCI ==> repairsects The rebuilt CI has overlapping DBs. To resolve this issue, remove some of the overlapping DBs. First resolve any DBs that have overlapping sectors. Then resolve DBs that have overlapping rowids (if any still exist). Enter the Start sector of the overlapping DB to remove, or Enter 0 to quit Support Utilities, Release 16.00

171 Command has been sent to Slave tasks. Chapter 3: Filer Utility (filer) REPAIRSECTS vproc 0 (0000) response The specified [W]DBs have been deleted. The DBs listed below have overlapping rowids: Start Sectors TableID FirstrowID Lastrowkey 02DE * F6 D26A *0000 9BFB F9E *0000 9BFA *0000 9FFD BFFC The rebuilt CI still has overlapping DBs. Run REPAIRSECTS again to try to resolve the problem. REBUILDCI ==> repairsects The rebuilt CI has overlapping DBs. To resolve this issue, remove some of the overlapping DBs. First resolve any DBs that have overlapping sectors. Then resolve DBs that have overlapping rowids (if any still exist). Enter the Start sector of the overlapping DB to remove, or Enter 0 to quit 0323 Command has been sent to Slave tasks. vproc 0 (0000) response The specified [W]DBs have been deleted. REBUILDCI ==> Support Utilities, Release

172 Chapter 3: Filer Utility (filer) RESETFREECYL RESETFREECYL Purpose The RESETFREECYL command resets the file system internal free cylinder count back to the correct value. Syntax Usage Notes This command complements the SET FREECYL command. 172 Support Utilities, Release 16.00

173 ROW Chapter 3: Filer Utility (filer) ROW Purpose The ROW command displays rows from one or more subtables. For a single subtable, ROW sets the context to those rows. Syntax Note: The online help lists the row options (/P, /S, /M, /L, /H) as /rowopt. You can type the options at the beginning or end of the command. /P /S or /C /M or /I /L or /R /H tid ridrange Positions to the physical rows or subtable requested. At least one physical row must be in the selected range. Nothing is displayed unless there is a problem. Physical rows store one of several types of data, such as regular table row data or column partition data. Counts the number of physical rows selected. This is the default for all subtables except the table header subtable. Displays only the row header information (length, RowID, and flags). Formats the row header and fields in the physical row. This option is the default for the table header subtable; however, this option is not the default formatted field display because the table header subtable has a special row format and requires special formatting. Format the row header and display the entire physical row in hexadecimal, including the row header. Specifies the subtable to display, which must include a typeandindex value. For detailed information, see Specifying a Subtable Identifier (tid). A range of rows selected based upon row identification in the row header rather than the ordinal position within the subtable. For information on ridrange formatting, see Rows and Row Ranges. The online help lists this option as rowspec [TO rowspec]. Support Utilities, Release

174 Chapter 3: Filer Utility (filer) ROW Usage Notes The prompt for the ROW selector is shown below: ROW ==> The ROW command operates on subtables. In the Teradata Database, a subtable is a collection of physical rows, and a table is a collection of subtables. The physical rows of a particular subtable may store regular data rows, index rows, table header rows, column partition values, or other types of data. Therefore, you can use the ROW command to display rows from the primary data subtable, the table header subtable, a secondary index subtable, etc. For example, to obtain a formatted display of the table header, use ROW /L xy H, 0 1 where xand y make up the tablenumber which identifies the table, and H is the typeandindex value which identifies the table header subtable. For more information on identifying tables and subtables, see Specifying a Subtable Identifier (tid). In addition to global and upper-level commands, the following lower-level commands are allowed at the ROW prompt: DELETE DISPLAY (without ordinalrange) DROP NEXT PATCH PREV WRITE 174 Support Utilities, Release 16.00

175 SCANDISK Chapter 3: Filer Utility (filer) SCANDISK Purpose The SCANDISK command validates the file system and reports any errors found, including discrepancies in the following items: Key file system data structures, such as master index (MI), current (memory) copy of the snapshot MI, cylinder indexes (CIs), data blocks (DBs), and free CIs. Additionally, SCANDISK validates the WAL Log structures: WAL master index (WMI), WAL cylinder indexes (WCIs), and WAL data blocks (WDBs). The RowIDs, including the partition numbers, for key physical rows should match the values stored for those rows in the cylinder index and master index. "Physical row" means the low-level row structure that is stored by the file system. A physical row can store one of several types of data, such as table rows, the data from one or more column partitions, a table header (metadata about the table), index data structures, and other types of data. Every physical row stores only one kind of data. The WAL Logical Sequence Number (WLSN) for key physical TJ and REDO records should match the values stored for those records in the WAL Cylinder Index and WAL Master Index. Within a subtable, the internal partition number of a physical row should be greater than or equal to the internal partition number in the preceding row, if any. Within a subtable, either all the physical rows are partitioned (the physical rows include an internal partition number) or all the physical rows are nonpartitioned (the physical rows do not include an internal partition number). Within a partition, the RowID of a physical row should be greater than the RowID in the preceding row, if any. The calculated checksum values of key file system data structures (Cylinder Index, Data Blocks and their WAL counterparts) should match the recorded checksum values (if checksums are enabled). SCANDISK does not validate the following items: The internal partition number is not validated for consistency with the result of the partitioning expression applied to the partitioning columns in a logical row, or for consistency with currently defined column partition numbers. The hash value in the RowID of the physical row header is not validated with the result of hashing the index columns for a primary index or primary AMP index. Support Utilities, Release

176 Chapter 3: Filer Utility (filer) SCANDISK Syntax /Y /S /M /L CI DB FREECIS Prevents Filer from displaying a confirmation message. Scans the MI and WMI. Scans the MI, CIs, WMI, and WCIs. Scans the MI, CIs, DBs, WMI, WCIs, and WDBs. Scans the MI and CIs. The free CIs are also scanned unless a tid is provided. Scans the MI, CIs, and DBs. This is the default for the normal file system, which can be overridden by the CI, MI, or FREECIS options. The free CIs also are scanned unless a tid is provided. Scans CIs of cylinders that the allocation map indicates are free. Reports any cylinder for which the CI does not also indicate that the cylinder is free. 176 Support Utilities, Release 16.00

177 FIX Chapter 3: Filer Utility (filer) SCANDISK Used with FREECIS only. SCANDISK attempts to resolve discrepancies by updating allocation map to match state of cylinder as indicated by the CI. Remaining discrepancies, as from corrupted cylinder data, are reported. MI WAL WCI WDB WMI Note: Run SCANDISK FREECIS without the FIX option before running it with the FIX option. This allows for inspection of the initial output prior to attempting to fix any problems. Note: The SKIPDECOMPRESS option cannot be used together with the FIX option. Scans the MI only. Scans the WAL log. Scans the WMI and WCIs. Scans the WMI, WCIs, and WDBs. This is the default for the WAL log, which can be overridden by the WCI or WMI options. Scans the WMI only. SKIPDECOMPRESS SKIPDECOMP WITHSPOOL tid ridrange Prevents SCANDISK from uncompressing block-level compressed data blocks. This skips some validation checks, but can speed SCANDISK when there are many compressed data blocks in the system. Note: The SKIPDECOMPRESS option cannot be used together with the FIX option. Causes scandisk to check all spool data in addition to checking non-spool data. If WITHSPOOL is not specified, only the first few spool cylinders are checked. If the scope of the command is limited to the MI, the entire MI is always scanned including non-spool and spool data. Specifies the subtable to scan. For more information on tid formatting, see Specifying a Subtable Identifier (tid). Restricts the scan of a subtable to the specified RowID range. For more information on ridrange formatting, see Rows and Row Ranges. The online help lists this option as rowspec [TO rowspec]. Support Utilities, Release

178 Chapter 3: Filer Utility (filer) SCANDISK TABLE ridspec inquire_opt Scans the specified subtable range. SCANDISK starts the scan at the subtable location specified by tid1 and ridspec1, and ends the scan at the subtable location specified by tid2 and ridspec2. Note: The starting subtable location,tid1 ridspec1, must be less than the ending subtable location, tid2 ridspec2. If the starting subtable does not exist (that is, the table was deleted), SCANDISK starts from the next tid which is greater than the starting tid. Similarly, if the ending subtable does not exist, SCANDISK ends the scan at the tid which is just less than the specified ending tid. If the ending subtable is not specified, SCANDISK scans from the starting subtable until the end of all the subtables, followed by the free CIs and the WAL log. For more information on tid formatting, see Specifying a Subtable Identifier (tid). For more information on ridspecformatting, see Rows and Row Ranges. The RowID specification used with the TABLE option to specify the subtable locations to start and stop the scan. For more information on RowIDs and ridspec formatting, see Rows and Row Ranges. The online help lists this option as rowspec. Displays the lowest tid and rowid being scanned among the AMPS involved. This option also reports SCANDISK progress as a percentage of total time to completion. SCANDISK continues reporting new errors it encounters after the inquiry. The syntax for the INQUIRE option is as follows: NONE specifies that only one INQUIRE request is sent for the SCANDISK job. number is an integer that defines the time interval to send an INQUIRE request to display SCANDISK progress. If timeopt is not specified, number defaults to SECONDS. timeopt specifies the time unit that number represents. It should be one of the following: SECONDS, SECOND, SECON, SECO, SECS, SEC, S MINUTES, MINUTE, MINUT, MINU, MINS, MIN, M HOURS, HOUR, HOU, HO, HRS, HR, H DAYS, DAY, DA, D For example, scandisk inquire 5 m will start a SCANDISK job which reports SCANDISK progress every five minutes. NOCR Note: The maximum time interval allowed is seven days. Specifies to use regular data block preloads instead of cylinder reads. This is the default when SCANDISK is started on online AMPs when the database is running. 178 Support Utilities, Release 16.00

179 CR MAXERRCNT n MAXERR n Usage Notes Specifies to use cylinder reads instead of regular data block preloads. This is the default when SCANDISK is run with the database offline or run on an AMP that is offline. Specifies the maximum number of errors SCANDISK should show per subtable per vproc. SCANDISK stops when it finds the specified number of errors. Use this setting to reduce the SCANDISK running time when there are an excessive number of errors reported. You can run SCANDISK while the system is online and the Teradata Database is available for normal operations. You can also use Filer to run SCANDISK when the database is offline, or on an AMP that is offline to help diagnose and fix file system issues. For more information see Starting Filer. Teradata recommends you run SCANDISK in the following situations: To validate data integrity before or after a system upgrade or expansion. If you suspect data corruption. As a routine data integrity check (perhaps weekly). Note: A slight performance impact might occur while SCANDISK is running. When Filer is started in vproc-specific mode (using the -v startup option), SCANDISK displays errors as they are detected, even if the INQUIRE command is not issued during the SCANDISK run. You can rearrange the order of the options following the SCANDISK command. Chapter 3: Filer Utility (filer) SCANDISK If you do not type any options, SCANDISK defaults to DB and all subtables on the vproc. The default scope is to scan both the normal file system and the WAL log, each from the lowest (DB, WDB) level through the highest (MI, WMI). The free CIs are also scanned. The SCANDISK command can be limited by the SCOPE command to scan, for example, just one table, just the WAL log, or just certain AMPs. For more information, see SCOPE (or VPROC). If you specify either tid or ridrange or both, all of the Teradata Database and CIs containing the requested table and/or rows are verified. If you use the TABLE option with the CI option, SCANDISK will check all of the cylinders that contain the specified subtable range. Note: The first and last of the cylinders checked may contain subtables outside of the specified range because multiple subtables can reside on a given cylinder. The lowest level file system structure that SCANDISK verifies is the data block. If you specify RowIDs, SCANDISK will start and end the scan at the data blocks which contain the specified rows. The entire data block is checked; therefore, rows outside of the specified row range may also be checked since a DB contains multiple rows. By default, SCANDISK uses regular data block preloads instead of cylinder reads. The CR option allows you to run SCANDISK using cylinder reads to preload data into cylinder slots which may improve SCANDISK performance. However, if other work also requires the use of cylinder slots, the competition for slots could Support Utilities, Release

180 Chapter 3: Filer Utility (filer) SCANDISK slow down SCANDISK and the other work. In addition, the performance gain is dependent on the amount of data loaded, the distribution of the data, and the average block I/O size. The NOCR option lets you turn off cylinder slot usage by SCANDISK, which could result in slower SCANDISK performance, but which will allow other work requiring cylinder slots to progress unimpeded. SCANDISK reports only what it finds when scanning is completed. The output from SCANDISK may include table row identifiers (RowIDs). A RowID consists of a 16-byte value that includes the following information: The first 8-bytes is the internal partition number of the row. (For a nonpartitioned table, the internal partition number is zero, and the internal partition number is not actually stored in the row itself.) The next 8-bytes include a hash bucket value and uniqueness value for the row. For more information on internal table row formats, see Database Design. Note: Teradata recommends that you do not run SCANDISK FREECIS FIX directly. First, run SCANDISK FREECIS and store and inspect the output. Then, run SCANDISK FIX. The Teradata Database must be down in order to run SCANDISK FREECIS FIX. If Filer aborts, or the system resets during a SCANDISK FREECIS operation, the recently freed cylinders are marked as unvalid because Filer will not have completed validating them. The AMPs that own these cylinders will be kept down when Teradata Database restarts. To bring the AMPs back up, start Filer again, set scope to the affected AMPs and reissue the SCANDISK FREECIS command. SCANDISK start, finish, abort, and system reset times are logged to the Linux /var/log/messages file and also to the DBS.SW_Event_Log table. Aborting SCANDISK Because SCANDISK DB verifies that every byte in the file system is accounted for, this process can be very time consuming. Therefore, you have the option of stopping the process by typing the following command: ABORT can take up to 30 seconds to process. After it is initiated, ABORT stops the SCANDISK process and reports the current status. When the abort is successful, the following appears: Are you sure you want to scandisk? (Y/N) y Tue Feb 28, :16:50 : Scandisk has been started on all AMP Vprocs in the SCOPE. Type ABORT to stop them before completion Type INQUIRE to check on progress of command ABORT Abort request has been sent Vproc 0 response Cylinder F4C2 Sector 16 (0010) length 1 (0001) DB ref count doesn t match DBD row Count Tue Feb 28, :16:50 : The scandisk found problems 180 Support Utilities, Release 16.00

181 Vproc 1 response Tue Feb 18, :16:50 : The scandisk found nothing wrong Filer ==> Chapter 3: Filer Utility (filer) SCANDISK Checking SCANDISK Status Because SCANDISK can take a long time to run, you might want to do a status check after starting a scan. You can do this using the INQUIRE command below: INQUIRE displays the lowest tid and rowid being scanned among the AMPS involved, and reports SCANDISK progress as a percentage of total time to completion. It also displays a list of errors that occurred since the last INQUIRE command. This.INQUIRE command is similar to the INQUIRE startup option for SCANDISK, but this command runs only once when you invoke it. Unlike the INQUIRE startup option, the INQUIRE command cannot be set to automatically repeat at specified intervals. The SCANFREE section of INQUIRE output shows the status of scanning free CIs. Starting and Restarting SCANDISK You can start SCANDISK directly from Filer, from Ferret, or from a supplied Perl script. The process of restarting SCANDISK if it was stopped or aborted during the scan differs depending on how it was started. (You cannot restart SCANDISK if it was started from Ferret.) When started from Filer, SCANDISK checks the file system structures from the beginning to the end of the SCANDISK job based on the options selected. If SCANDISK is stopped or aborted in the middle of the scan, you normally will have to rerun the job from the beginning, unless you do the following: 1. Before aborting the SCANDISK job, send an INQUIRE request. For details, see Checking SCANDISK Status. 2. When you see the output from the INQUIRE command in the application window, manually capture the tid and rowid displayed, then abort the SCANDISK job. You can restart the aborted SCANDISK job in one of the following ways: Start SCANDISK with the TABLE option, and pass the captured tid and rowid to the TABLE option to indicate the subtable where the scan should restart. Start SCANDISK using the restartscandisk.pl script with the -t option, and pass the captured tid and rowid to the -t option. The restartscandisk.pl Script restartscandisk.pl is a Perl script supplied by Teradata to aid in using SCANDISK. When SCANDISK is started using the script, it is much easier to start a SCANDISK job, stop it, then restart it from the place at which it was stopped. Teradata recommends that you start SCANDISK using the script if you expect to restart it later. The script is located in /usr/tdbms/bin. To run SCANDISK from the script, execute the restartscandisk.pl Perl script at the command line or at the Teradata Command Prompt. The following requirements apply when running the script. Support Utilities, Release

182 Chapter 3: Filer Utility (filer) SCANDISK restartscandisk.pl must be executed in the Teradata environment. Therefore, to run the script on a Linux system, make sure you have executed tdatcmd first to set up the Teradata environment. For more information, see Starting the Utilities. restartscandisk.pl must be run on the Control Node (the node running with the lowest node ID number). To determine the Control Node, type cnscim -host at the command line. restartscandisk.pl configures Filer to use decimal numeric values as input and output. Therefore, you must enter decimal values when specifying numeric input to the script. For more information, see RADIX. The syntax for restartscandisk.pl is shown below. Syntax Element scandisk command Description The SCANDISK command and options you want to run. Note: You must enter decimal values when specifying numeric input for SCANDISK options, such as tid or ridspec. input filename The file to be used as input to the script. This file should contain the output from the previously executed SCANDISK job. Note: If you specify SCANDISK with the TABLE option, restartscandisk.pl will use the TABLE option values and ignore this input file. In this case, SCANDISK can only be manually restarted. tid [ridspec] Specifies the subtable and optionally, the row where SCANDISK should restart the scan job. For more information on tid formatting, see Specifying a Subtable Identifier (tid). For more information on RowIDs and ridspecformatting, see Rows and Row Ranges. The online help lists the ridspec option as rowspec. Note: You must enter decimal values for tid and ridspec. If you specify SCANDISK with the TABLE option, restartscandisk.pl will use the TABLE option values and ignore these tidand ridspec values. In this case, SCANDISK can only be manually restarted. output filename The file which will store the SCANDISK output. 182 Support Utilities, Release 16.00

183 Chapter 3: Filer Utility (filer) SCANDISK Syntax Element Description ENA Note: You must specify an absolute path to this file. If you use a relative path, the SCANDISK output from the application log may be redirected to a different location than the output from restartscandisk.pl. If you do not specify an output file or input file, the script will use the standard output file (stdout). If you specify SCANDISK with the TABLE option, restartscandisk.pl will use the TABLE option values and ignore this output file. In this case, SCANDISK can only be manually restarted. Specifies to start Filer with locking enabled, which means that Filer will observe the file system locking protocols used to avoid hang or deadlock situations. This is the default. DIS Specifies to start Filer with locking disabled. This option starts Filer with the -l locking option, which means that Filer will not observe the file system locking protocols used to avoid hang or deadlock situations. For information about the -l locking option, see Standard Filer Startup Syntax. time interval The time in seconds specifying the interval at which the INQUIRE request is sent to display SCANDISK progress. The default is 600 seconds. Note: The minimum time interval allowed is 600 seconds. If you enter a time interval that is less than 600 seconds, restartscandisk.pl uses the default value of 600 seconds. AMP number Specifies the vproc number to be used with the Filer SCOPE command. The SCANDISK command will be sent only to the specified vproc. For more information, see SCOPE (or VPROC). Usage Notes for the restartscandisk script Note: If the file system data structures were fixed to repair a problem, and the location (tid and rowid) where the previous SCANDISK halted is beyond the location where the file system fix occurred, Teradata recommends that you restart the SCANDISK job from the beginning instead of restarting the scan at the location where it last halted. Entering options to the restartscandisk script There are many ways to enter options to the restartscandisk script. The following describes how you can specify a SCANDISK command and options for the script to run. IF you specify... a SCANDISK command an input file, but no SCANDISK command THEN... the script starts Filer to run the SCANDISK command and any options specified. the script starts Filer to run the SCANDISK command and options used in the previous scan job as specified in the input file. Support Utilities, Release

184 Chapter 3: Filer Utility (filer) SCANDISK IF you specify... an input file, and a SCANDISK command no input file and no SCANDISK command THEN... the script will check if the submitted SCANDISK command is the same as the command previously executed. If the submitted command and options are different from the previous command, the script will display a warning and ask for confirmation to run the new command. the script starts Filer to run SCANDISK with no options. The following describes how you can specify the location where SCANDISK should restart a previous scan job. IF you specify... an input file the -t option with a tid and optional ridspec the TABLE option with the SCANDISK command no input file, no -t option, and no TABLE option THEN... SCANDISK restarts from the subtable specified by the tidand rowid from the last INQUIRE command in the input file. SCANDISK restarts from the subtable specified by tid and ridspec. the script ignores any input file, -t option, or output file you specify, and runs SCANDISK with the values of the TABLE option. To restart SCANDISK later, you must manually capture the last tid and rowid from the application window. SCANDISK restarts the job from the beginning. Note: You can specify either the -i option or the -t option, but not both options together. The following describes how you can specify where the SCANDISK output will be saved or sent. IF you specify... an output file an input file, but no output file no input file and no output file THEN... the output of SCANDISK is saved to this file. the script reuses the input file to save the output of the SCANDISK job. The new output is appended to the end of the file. the output of SCANDISK is sent to the standard output file (stdout). The output from the restartscandisk script includes output from the INQUIRE command, which contains the lowest tid and rowid scanned among the AMPS involved. This INQUIRE output can be used to identify the location where a scan ended. When this output is saved to a file, you can use the file as input to the restartscandisk script, by using the -i option. The script restarts SCANDISK at the location specified by the tid and rowid from the last INQUIRE command in the file. If the database restarts while executing this script, the output file may not capture the complete application window output. Therefore, the output file may not contain a restart point for SCANDISK, or the restart point may be incorrect. If the exact restart point is not important, you can still use this output file as input to restart SCANDISK. However, if the exact restart point is important, then it is better to manually capture the last tid and rowidfrom the application window, and use these values to restart SCANDISK. You can pass the tid and 184 Support Utilities, Release 16.00

185 rowid to the restartscandisk script using the -t option, which restarts SCANDISK at the specified location. If you do not use the -o or -i options to save the SCANDISK output to a file, then you must manually capture the tid and rowid from the last INQUIRE command in the application window if you want to restart SCANDISK later by using the -t option. It is recommended that you use the -o or -i option to ease restartability. Note: If you manually capture the SCANDISK output in the application window to a separate log file, this file cannot be used as an input file to the restartscandisk script because the application window log does not contain the output required by the restartscandisk script. Aborting the restartscandisk script You can use one of the methods listed below to interrupt the restartscandisk script and abort the running SCANDISK job. For specific key sequences used to generate the signals specified, see the documentation for signal generation for your Operating System platform. Send a SIGINT signal (CTRL+C) Send a SIGQUIT signal (CTRL+\, on Linux systems) Enter the ABORT command in the application window where Filer is running In the first 2 cases, when the restartscandisk script receives the signal to abort, it displays the current percentage of SCANDISK completion and requests confirmation to abort the scan job. Upon confirmation, the script gracefully aborts the running SCANDISK. You can restart the aborted SCANDISK job later by using the -i or -t option. If you enter the ABORT command in the application window where Filer is running, the script will gracefully abort the running SCANDISK job without asking for confirmation. You can restart the aborted SCANDISK job later by using the -i or -t option. If you abort SCANDISK when it is scanning free CIs or the WAL log, and you restart the scan job, SCANDISK will restart the scan from the beginning of the particular stage where it was aborted (either scanning free CIs or the WAL log). restartscandisk errors and warnings The restartscandisk script displays various error messages when it encounters problems. It also displays warnings and informational messages to assist the user. The following table lists possible error messages returned by the restartscandisk script, with explanations and suggested remedies for each. Error Message Explanation Remedy Chapter 3: Filer Utility (filer) SCANDISK The AMP number specified with the -s option is invalid. The AMP number specified with the -s option is invalid. The AMP number is greater than the total number of AMPs in the system. Cannot get the application window number. The user specified an invalid AMP number with the -s option. The user specified an AMP number that is greater than the number of AMPs in the system. The Teradata Database restarted just after the start filer command was submitted. Specify a valid AMP number with the -s option. Specify an AMP number that is less than the number of AMPs in the system. Check the PDE state, and submit the script when the PDE is running. Support Utilities, Release

186 Chapter 3: Filer Utility (filer) SCANDISK Error Message Explanation Remedy Cannot get the total number of AMPs in the system to validate the specified AMP number. cnscim -s failed. ENA and DIS are options to the restartscandisk script and cannot be used as output file names. Please specify another file name for the output file. Error encountered when submitting the SCANDISK command to Filer. Possible SCANDISK syntax error. Failed to open filename. Filer cannot be started because all of the interactive partitions are busy. Input file is invalid. Please specify a correct input file to restart SCANDISK. Output file is invalid. Please check the output file name. PDE must be running to start the restartscandisk script. The script tried to verify that the AMP number specified with the -s option is not greater than the number of AMPs in the system; however, it failed to get the total number of AMPs present in the system. The Teradata Database was restarted or stopped after the script submitted the start filer command. The user specified ENA or DIS as the output file, which is not allowed since ENA and DIS are options to the script. The script submitted the SCANDISK command to Filer, and it received an error from SCANDISK. The submitted SCANDISK command may be invalid. The script failed to open a needed file. The file may be a temporary file used by the script, the specified input or output file, or cnstool. All of the application windows are currently used for other purposes, and the script cannot find an available application window where Filer can be started. The user specified an invalid input file which does not contain SCANDISK output. The user specified an invalid output file. The PDE is not running, and the script cannot be started if the PDE is down. Verify that the DBS is running and that you have the correct permissions to run the vprocmanager command. If the database was restarted, wait until the PDE is running, then resubmit the script. If the PDE is not running, start the PDE, then resubmit the script. Specify a file name other than ENA or DIS for the output file. Make sure the SCANDISK command you submit to the script is valid. If you are not specifying a SCANDISK command to the script, and the error persists, report the problem to the Teradata Support Center. Verify that the required file has the correct permissions. Run the script at a later time when an application window is free. Specify an input file which contains output from a previous SCANDISK job. If multiple output files exist, submit the latest output file as input for restarting SCANDISK. Check the path of the output file for correctness. Do not specify a directory name as the output file name. Start the PDE, and submit the script when the PDE is running. 186 Support Utilities, Release 16.00

187 Chapter 3: Filer Utility (filer) SCANDISK Error Message Explanation Remedy The restartscandisk script cannot be run on nodes other than the Control Node. SCANDISK command submitted with incomplete options. Please resubmit the SCANDISK command with all the required options. The specified input file does not exist. Syntax Error: Invalid syntax or options were entered for the restartscandisk script. Syntax Error: Please specify a time interval with the -tm option. Syntax Error: Please specify an AMP number with the -s option. Syntax Error: Please specify an input file with the -i option. Syntax Error: Please specify an output file with the -o option. Syntax Error: Please specify table options tid[ridspec] with the -t option. Syntax Error: -t option cannot be used together with the -i option. Syntax Error: The ENA and DIS options cannot be specified together. The time interval specified is invalid. Please enter a number which represents the number of seconds as the time interval. The minimum time interval allowed is 600. The user did not run the restartscandisk script on the Control Node. The script can be run only on the Control Node. The user submitted a SCANDISK command with incomplete options, and Filer prompted for additional input. The user specified an input file that does not exist. The user specified invalid options to the script. The user specified the -tm option, but did not specify a time interval. The user specified the -s option, but did not specify an AMP number. The user specified the -i option, but did not specify an input file. The user specified the -o option, but did not specify an output file. The user specified the -t option, but did not specify a tid or ridspec. The user specified both -i and -t options to the script. The user specified both the ENA and the DIS options. The user specified the -tm option, but the time interval specified is not a number. Run the script on the Control Node, the node running with the lowest Node ID number. To determine the Control Node, run cnscim -host from the command line. Make sure the SCANDISK command you submit to the script includes all the required options. Verify the path and file name of the input file, and make sure that the file exists. Check the syntax and options entered for the script, and verify that they are correct. Specify a time interval in seconds with the -tm option. The time interval must be 600 seconds or more. Specify an AMP number with the -s option. The AMP number will be used with the Filer SCOPE command to limit the SCANDISK scope. Specify an input file with the -i option. Specify an output file with the -o option. Specify a tid and optional ridspec with the -t option. Specify either the -i option or the -t option. You cannot specify both options together. Specify either the ENA or the DIS option. You cannot specify both options together. Specify a number which represents the number of seconds as the time interval. The time interval must be 600 seconds or more. Support Utilities, Release

188 Chapter 3: Filer Utility (filer) SCANDISK Error Message Explanation Remedy The vprocmanager command failed. The script executed the vprocmanager command to get the number of AMPs in the system, and the command failed. Verify that the DBS is running and that you have the correct permissions to run the vprocmanager command. Message The following table lists warnings and informational messages that may be displayed by the restartscandisk script. Already received abort request, please wait... Based on the input file, SCANDISK will restart scanning at the FREECIS stage. Based on the input file, SCANDISK will restart scanning at the WAL log stage. Did not find any status to display. Filer is not in Read mode, please wait... Filer is starting, please wait... Initializing... The input file does not contain output from an INQUIRE command for the last SCANDISK job. Do you want to check previous SCANDISK jobs for status? The input file does not contain output from an INQUIRE command. SCANDISK will start the scan from the beginning. Processing the abort request, please wait Description The script already received the previous interrupt signal and is waiting until Filer is in Read mode before it can process the abort request. Based on information from the specified input file, the restart point will be FREECIS. That is, SCANDISK will start scanning the free cylinders. This implies that all of the subtables in the system were scanned already. Based on information from the specified input file, the restart point will be the WAL log. That is, SCANDISK will start scanning the WAL log buffers. This implies that all of the subtables and free cylinders in the system were scanned already. If the script is interrupted, the status of the SCANDISK job is normally displayed before aborting. In this case, the script has not submitted any INQUIRE commands to Filer before the interrupt. Therefore, no status information is available. The script will ask the user for confirmation before aborting. The script received an interrupt signal when Filer was not in Read mode and could not accept the ABORT command. The script will wait until Filer is in Read mode, then it will ask the user for confirmation to abort after displaying the latest SCANDISK status. The script has started Filer and is waiting for Filer to reach Read mode. The script is submitting initial commands to Filer and has not submitted the SCANDISK command yet. The specified input file does not contain INQUIRE command output from the latest execution of SCANDISK. The script asks the user for confirmation to check previous SCANDISK jobs to get a restart point. If the input file has output from more than one SCANDISK job, then the latest SCANDISK job containing INQUIRE command output will be used to determine the restart point. The specified input file does not contain any INQUIRE command output. Therefore, the script cannot determine a restart point. The script will ask the user for confirmation to start the scan from the beginning. The script received an interrupt signal and is processing the abort request. 188 Support Utilities, Release 16.00

189 Message Restart point found in the input file. Restart subtable is tid and RowID is rowid. SCANDISK will restart from this location. The restartscandisk script received an interrupt signal during the Initializing stage. SCANDISK completed the scan job in the previous run. Starting this scan job from the beginning. The specified INQUIRE time interval is less than 600 seconds. Setting the time interval to the default value of 600 seconds. Unable to get the current status. User specified an input file, but no output file. Using the file filename as both input and output file. User specified an input file, but no SCANDISK command. Using previous SCANDISK command: scandisk_cmd WARNING: No output file was specified. To restart this SCANDISK job later, capture the last tid and rowid from the application window. WARNING: Previous SCANDISK command and current SCANDISK command are different. WARNING: Specifying the INQUIRE option with the SCANDISK command may result in more INQUIRE outputs than expected. Use the -tm option to display INQUIRE output instead. Description The script found the restart point based on the last INQUIRE command output in the input file. SCANDISK will restart at the specified tid and rowid. The script received an interrupt signal before it could send the SCANDISK command to Filer. Chapter 3: Filer Utility (filer) SCANDISK The input file indicates that the previous SCANDISK job was completed. Therefore, the script will start a new SCANDISK job from the beginning. The user specified the -tm option with a time interval that is less than the minimum time interval allowed, which is 600 seconds. The script will set the time interval to the default value of 600 seconds. If the script is interrupted, the status of the SCANDISK job is normally displayed before aborting. However, since the user did not specify the -o or -i option, the script cannot get the SCANDISK status. It is recommended that the -o or -i option be used so that SCANDISK output will be saved to a file, which allows the script to display status information before aborting. The user specified an input file, but did not specify an output file. The script will reuse the input file as the output file. The user specified an input file, but did not specify a SCANDISK command. The script will submit the SCANDISK command used to run the previous scan job as specified in the input file. The user did not specify an output file. In order to restart this SCANDISK job later, the user must manually capture the tid and rowid from the latest INQUIRE output in the application window. It is recommended that the user specify an output file to ease restartability. The user is restarting a previous SCANDISK job and has specified both an input file and a SCANDISK command. The submitted SCANDISK command is different from the command used to run the previous scan job. In this case, the user will be asked for confirmation to run the submitted command. The user specified the INQUIRE option with the SCANDISK command. Since the restartscandisk script already sends periodic INQUIRE commands by default, this will result in more INQUIRE commands sent than expected. The user should use the -tm option to display INQUIRE output instead of using the INQUIRE option with the SCANDISK command. Support Utilities, Release

190 Chapter 3: Filer Utility (filer) SCANDISK Message WARNING: You have specified the TABLE option. The -i, -t, and -o options will be ignored. To restart this SCANDISK job later, capture the last tid and rowid from the application window. Description The user specified the TABLE option with the SCANDISK command. Therefore, SCANDISK will run with the values of the TABLE option and ignore the -i, -t, and -o options. The SCANDISK output will not be saved to a file. The user must manually capture the last tid and rowid from the application window to restart SCANDISK later. SCANDISK Examples Example: Starting, stopping, and restarting a SCANDISK job The following example illustrates a typical way to start, stop and restart a SCANDISK job. 1. Start the restartscandisk.pl script with the -o option. perl restartscandisk.pl -o /home/test/ofile.out Because no SCANDISK command or input file was specified, the script runs the default SCANDISK command from the beginning. That is, it runs SCANDISK with no options on all subtables, free CIs, and the WAL log from the lowest (DB, WDB) level to the highest (MI, WMI). The output of the scan is saved to the ofile.out file. 2. Stop the running SCANDISK job by entering the ABORT command or sending a signal to abort. For details, see Aborting the restartscandisk script. 3. Restart the SCANDISK job at the point in which it was interrupted by issuing the following command. perl restartscandisk.pl -i /home/test/ofile.out The output file from step one contains the location where the SCANDISK job was stopped; therefore, it is used as the input file to restart SCANDISK at that point. Since an input file was specified, but no SCANDISK command, the script starts the previous SCANDISK command from step one. Since no output file was specified, the script will reuse the input file and append the results of the restarted scan to /home/test/ofile.out. Example: Starting, stopping, and restarting a halted SCANDISK job The following example illustrates one way to start, stop, and restart a halted SCANDISK job from the location specified by a tid and ridspec. 1. Start the restartscandisk script. perl restartscandisk.pl SCANDISK CI Because no input file or tid[ridspec] were specified, the script runs the SCANDISK CI command from the beginning. The output is not saved to a file, but is sent to the standard output file (stdout). 2. Stop the running SCANDISK job by entering the ABORT command or sending a signal to abort. For details, see Aborting the restartscandisk script. 190 Support Utilities, Release 16.00

191 Chapter 3: Filer Utility (filer) SCANDISK 3. Because the SCANDISK output was not saved to a file, you must manually capture the tid and rowid displayed by the last INQUIRE command in the application window if you want to restart this SCANDISK job. 4. If the tidand rowid displayed by the last INQUIRE command was , you can restart SCANDISK from that location by issuing the following: perl restartscandisk.pl -t Note: Because no SCANDISK command or input file was specified, the script runs the default SCANDISK command (SCANDISK with no options) instead of the previous SCANDISK CI command. Example: Running SCANDISK with no options SCANDISK with no options scans all subtables from the DB level examining DBs, CIs, the MI, and free CIs. It also scans the WAL log at the WDB level, scanning WDBs, WCIs, and the WMI. In the following example, SCANDISK is started from Filer with no options. In order to restart this scan job later, you must submit an INQUIRE request to SCANDISK before aborting, and manually capture the tid and rowid from the last INQUIRE output in the application window. To restart SCANDISK, submit the captured tid and rowid to the SCANDISK command as part of the TABLE option, or restart SCANDISK using the restartscandisk script with the -t option. Filer ==> scandisk Cylinder F4C2 Sector 16 (0010) length 1 (0001) DB ref count doesn t match DBD row Count The scandisk found problems Example: SCANDISK output showing interrupted write pattern in CI The following is an example of output that SCANDISK generates when it finds an LSI interrupted write pattern in a CI. In this example, SCANDISK is started from Filer. Filer ==> scandisk ci Mon May 06, :12:20 :Scandisk has been started on all AMP Vprocs in the SCOPE. vproc 0 (0000) response Mon May 06, :55:21 : Cylinder F4C2 Mon May 06, :55:21 : LSI interrupted write pattern found in CI MAY 05 05:02:35 LUN 1111, Start Block , Blocks 0400 SRD num table id firstdbd dbdcount offset u0 u1 tai FFFF E Mon May 06, :55:21 : Invalid DBD sector length of (3A35) found Mon May 06, :55:21 : Invalid DBD sector length of (6920) found Mon May 06, :55:21 : Invalid DBD sector length of (6574) found Mon May 06, :55:21 : Invalid DBD sector length of (3030) found Mon May 06, :55:21 : First rowid out of order dbds 18 (0012) and 19 (0013) Support Utilities, Release

192 Chapter 3: Filer Utility (filer) SCANDISK Example: SCANDISK output showing interrupted write pattern in DB The following is an example of output that SCANDISK generates when it finds an LSI interrupted write pattern in a DB. In this example, SCANDISK is started from Filer. Filer ==> scandisk db Tue Feb 28, :16:50 :Scandisk has been started on all AMP Vprocs in the SCOPE. vproc 0 (0000) response Mon May 06, :12:20 1 of 1 vprocs responded with no messages or errors. Type 'ABORT' to stop the command before completion Type 'INQUIRE' to check on progress of command Reading vproc 0 (0000) response Mon May 06, :11:11 : Cylinder F4C2 Mon May 06, :11:11 : LSI interrupted write pattern found in DB MAY 05 05:02:35 LUN 1111, Start Block , Blocks 0400 Mon May 06, :11:11 : rows -1 (FFFFFFFF) and 0 (0000) are out of order Example: Running SCANDISK on a data subtable In the following example, SCANDISK scans the employee.ttt data subtable at the DB level. SCANDISK employee.ttt p Example: Running SCANDISK on a range of rows in a data subtable In the following example, SCANDISK scans a range of rows in a primary data subtable. SCANDISK employee.ttt p 09E8 C4AB to C Example: Running SCANDISK with the INQUIRE option In the following example, the INQUIRE command is automatically repeated at one minute intervals. Therefore, you get a display of SCANDISK progress every minute. Filer ==> > scandisk inq 1 m > Mon Jul 16, :02:45 SCANDISK STATUS : Slowest vproc 2 is 44% done Fastest vproc 3 is 64% done The scandisk is about 55% done 192 Support Utilities, Release 16.00

193 SCOPE (or VPROC) Chapter 3: Filer Utility (filer) SCOPE (or VPROC) Purpose The SCOPE or VPROC command selects a set of vprocs that Filer will communicate with until Filer quits or a different set of vprocs is selected. Syntax ALL number Sets the scope to all vprocs. Numeric identifier of a single vproc. Usage Notes In this mode, all subsequent Filer commands are sent to each vproc selected, and each vproc responds with its answer. After this command is issued, most commands requiring confirmation cannot be used. The SCOPE and VPROC commands cannot be used if Filer is started in vproc-specific mode using the -v start up option. For more information, see Standard Filer Startup Syntax. The SCOPE and VPROC commands cannot be used if Filer is started from the Debugger. For information on invoking Filer from the System Debugger or Coroner, see Running Filer from the System Debugger and Coroner. Support Utilities, Release

194 Chapter 3: Filer Utility (filer) SEGZERO (or SEG0) SEGZERO (or SEG0) Purpose The SEGZERO command selects and displays Segment Zero (0). Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /V) as /dispopt. /P /H Positions to Segment 0, but does not display it. Shows a hexadecimal display of Segment 0. Usage Notes This segment contains the following: The information the file system needs to interface with the FSG kernel subsystem. The flags that indicate to the file system what happened the last time the vproc was shut down. The FSG segment IDs of the MI and FTR trace buffer. 194 Support Utilities, Release 16.00

195 SET FREECYL Chapter 3: Filer Utility (filer) SET FREECYL Purpose The SET FREECYL command sets the file system internal Free Cylinder count to the specified number. Syntax /Y number Prevents Filer from prompting for confirmation before executing the command. The number of free cylinders. An error occurs if less than the specified number of cylinders free already exist. Usage Notes Use the RESETFREECYL command to reset the Free Cylinder count back to the correct value. Support Utilities, Release

196 Chapter 3: Filer Utility (filer) SHOWDEFAULTS SHOWDEFAULTS Purpose The SHOWDEFAULTS command shows the default settings and the saved context. The command displays the current default radix for input and output, the current input, output, and error file names, and the SCOPE defined, including what the equal sign (=) means, as shown in the following example: The present value for the = is: Dec (HEX) In addition, SHOWDEFAULTS also displays WAL log context information. Syntax 196 Support Utilities, Release 16.00

197 SNAPMI Chapter 3: Filer Utility (filer) SNAPMI Purpose Selects and displays the snapshot MI. Displays information about where the MI is stored on disk. Provides an option to invalidate the snapshot MI, forcing the MI to be rebuilt from CIs at the next file system startup. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /V, /X) as /dispopt. /P /S /M /L /H /V Positions to the snapshot MI in memory. Nothing is displayed unless a problem exists. Shows the fixed area and a formatted display of the first and last CID of the snapshot MI segment. When used with the CYLINFO option, displays only the snapshot MI CIDs. Shows the fixed area and a formatted display of all the CIDs in the snapshot MI segment. When used with the CYLINFO option, displays snapshot MI CIDs and the start sector of each snapshot MI chunk. Periodically, the snapshot MI in memory is divided into chunks that are individually written to disk. Shows the fixed area and a formatted display of all the CIDs in the snapshot MI segment. When used with the CYLINFO option, displays all snapshot MI chunk headers. Displays a hex dump of the snapshot MI segment. Verifies that the snapshot MI segment has all CIDs in the correct order, that all sectors are accounted for, and matches the setting of the FIB. Support Utilities, Release

198 Chapter 3: Filer Utility (filer) SNAPMI /X CYLINFO INVALIDATE Usage Notes Same as /L. Displays the disk location of the snapshot MI. Invalidates the snapshot MI, forcing the system to rebuild the MI at the next file system startup by reading the CIs. The prompt for the snapshot MI selector is shown below: SNAPMI ==> In addition to global and upper-level commands, the following lower-level commands are allowed at the SNAPMI prompt: DISPLAY (without an ordinalrange) DROP PATCH When the snapshot MI is written to disk, it is divided into chunks, and each chunk is written individually. If checksums are enabled, each chunk includes a checksum. The checksum value for the chunk is stored within the chunk header data. When the snapshot MI is restored during system startup, these checksums are verified to ensure the integrity of the snapshot MI. CIDs with checksums will show additional checksum columns in the SNAPMI output screen. 198 Support Utilities, Release 16.00

199 SRD Chapter 3: Filer Utility (filer) SRD Purpose The SRD (Subtable Descriptor) command returns a formatted display of all the SRDs on the current CI that have any rows from the subtable specified by tid. If you omit tid, all SRDs are displayed. Syntax Note: The online help lists the display options (/P, /S, /M, /L) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L tid Positions to the SRD with rows from the subtable, tid. This option is valid only if you specify a tid. If this tid does not exist on this CI, the command will position to the SRD of the next higher tid. Nothing is displayed unless there is a problem. Displays the selected SRDs on the CI. Displays the first and last DBDs of each selected SRD, in addition to the information displayed by the /S option. Displays all the DBDs of each selected SRD and the free sector entries, in addition to the information displayed by the /S option. The subtable of interest. For more information on tid formatting, see Specifying a Subtable Identifier (tid). Usage Notes The SRD command must be invoked from a CI context (as indicated by the CI> prompt). The command formats the SRDs and DBDs associated with the specified tid. If you omit the optional tid, then all the SRDs on the CI are selected and displayed. For more information, see Specifying a Subtable Identifier (tid). Support Utilities, Release

200 Chapter 3: Filer Utility (filer) STAMP STAMP Purpose The STAMP command takes five hexadecimal numbers and converts them to a file system timestamp. Syntax number number must be less than 256, since each number represents one byte of the timestamp. Usage Notes The STAMP command can be used to see when a segment was created. It is used in debugging when viewing a crash dump. The hex values at the offset where the time stamp is located can be converted to a timestamp which can be compared to the current version of the segment. 200 Support Utilities, Release 16.00

201 STORAGEINFO Chapter 3: Filer Utility (filer) STORAGEINFO Purpose The STORAGEINFO command displays storage control information for each AMP in the current scope. Syntax /M Shows how the cylinders are distributed among the different types of storage and cylinder states. Support Utilities, Release

202 Chapter 3: Filer Utility (filer) SYNCSCAN SYNCSCAN Purpose The SYNCSCAN command shows all currently running scans on an AMP that are eligible for the synchronized full-file scan feature. Syntax Note: The online help lists the display options (/S, /M, /L) as /dispopt. You can type the options at the beginning or end of the command. /S /M /L tid Displays global information and a one-line summary for each eligible subtable. Same as /S, but also displays the process IDs of the tasks scanning each subtable. Displays global information and detailed information for each task scanning each subtable. The subtable of interest. For more information on tid formatting, see Specifying a Subtable Identifier (tid). 202 Support Utilities, Release 16.00

203 TABLE Chapter 3: Filer Utility (filer) TABLE Purpose The TABLE command displays rows from the subtable specified by the tid argument. Syntax Note: The online help lists the row options (/P, /S, /M, /L, /H) as /rowopt. You can type the options at the beginning or end of the command. /P /S or /C /M or /I /L or /R /H tid ordinalrange Positions to the physical rows or subtable requested. At least one physical row must be in the selected range. Nothing is displayed unless there is a problem. Physical rows store one of several types of data, such as regular table row data or column partition data. Counts the number of physical rows selected. This is the default for all subtables except the table header subtable. Displays only the row header information (length, RowID, and flags). Formats the row header and fields in the physical row. This option is the default for the table header subtable; however, this option is not the default formatted field display because the table header subtable has a special row format and requires special formatting. Formats the row header and display the entire physical row in hexadecimal, including the header. Specifies the subtable to display. The tid specification must include a typeandindex value. For detailed information, see Specifying a Subtable Identifier (tid). Selects the range of rows to display. No matter what is displayed, the resulting context is the whole subtable. For detailed information, see Specifying Display Ranges (ordinalrange). The online help lists this option as rangeopt. Support Utilities, Release

204 Chapter 3: Filer Utility (filer) TABLE Usage Notes The prompt for the TABLE selector is shown below: TABLE ==> In addition to global and upper-level commands, the following lower-level commands are allowed at the TABLE prompt: DELETE DISPLAY DROP NEXT PREV Use the option definitions of the ROW command with these lower-level commands. The current selected object can be a subtable or range of subtables. The prompt reflects the current selection. The TABLE command operates on subtables. In the Teradata Database, a subtable is a collection of physical rows, and a table is a collection of subtables. The physical rows of a particular subtable may store data rows, index rows, table header rows, column partition values, or other types of data. Therefore, you can use the TABLE command to display rows from the primary data subtable, the table header subtable, a secondary index subtable, etc. For example, to obtain a formatted display of the table header, use Table /L x y H 1 where x and y make up the tablenumber which identifies the table, and H is the typeandindex value which identifies the table header subtable. For more information on identifying tables and subtables, see Specifying a Subtable Identifier (tid). The initial context for the TABLE command consists of the physical rows of the subtable, and the ordinalrange argument is interpreted with respect to this context. Positive integers without the BEG/END prefix are ordinal row numbers, where 1 is the first row. Negative integers without the BEG/END prefix are backwards displacements from the last row, so that -1 is the second to last row. Note: This differs from the interpretation of an ordinalrange for the DISPLAY command, where integers without the BEG/END prefix always refer to objects outside the current context, so that positive values refer to objects past the end of the current context, and negative values refer to objects before the beginning of the current context. The resulting context from a TABLE command is a whole subtable, a subtable range, or null, but never a row range. The ordinalrange can restrict the display to a subset of the resulting context, even though this ordinalrange has no affect on the resulting context. If the TABLE command uses a wildcard for the type part of the tid, the resultant context is special (nothing), and this context is not usable with an ordinalrange argument. The display is a multiple subtable display, where each subtable is constrained by the ordinal range. For each subtable, the BEG/END +/- number rules apply, except that the - number construct is not allowed. For more information on ordinalrange, see Specifying Display Ranges (ordinalrange). 204 Support Utilities, Release 16.00

205 Example: Using the Filer TABLE command to display headers of five rows of a table The following command shows a medium display of the first five rows of table : Chapter 3: Filer Utility (filer) TABLE table /m to 5 The output would look like this: TableID: ROWS: B SECTS: D ROW length rowid flags presences part hash0 hash1 uniq0 uniq F C A ED A A 4C5B A F FDC A A 65DD A Example: Using the Filer TABLE command to display the header of the second-to-last row of a table The following command shows a medium display of the row header for the second-to-last row of the startdates table in the EMPLOYEE database: table "employees.startdates" p end-1 /m Support Utilities, Release

206 Chapter 3: Filer Utility (filer) TABLEID TABLEID Purpose The TABLEID command displays the table number of the specified table when given the database name and table name. Syntax databasename tablename The name of the database containing the table for which the table number will be displayed. The name of the table for which the table number will be displayed. Note: The database name and table name must be delimited by apostrophes or double quotation marks, as indicated in the syntax diagram. Usage Notes This command is not allowed if the Teradata Database is in the Debug-Stop state. A table is identified in the Data Dictionary by a table number (tvm.tvmid). Each table number is unique across the whole system, rather than local to a database. Therefore, a table number uniquely identifies a table in the system. The TABLEID command displays the table number of the table specified by databasename and tablename. The output of the TABLEID command is a numeric subtable identifier (tid), which consists of three numbers: The first two comprise the table number. This pair of numbers is used to uniquely identify a table in the system. The third is the typeandindex value, which specifies a kind of subtable, such as a table header, data subtable, or a particular index subtable. TABLEID always returns a typeandindex value of zero (0), which specifies the table header. For more information on how to interpret a tid, see Specifying a Subtable Identifier (tid). The TABLEID command also sets the saved context to the located tid. (Use = in place of tid in subsequent commands to reference the found tid.) The following rules apply when specifying databasename and tablename: A period (.) is required to separate the database name from the table name. You must use either apostrophes ( ' ) or double quotation marks ( " ) when typing a database name and table name. The results are the same. You can specify a fully qualified table name using any one of the methods suggested in the syntax diagram with the following exceptions: 206 Support Utilities, Release 16.00

207 The object name itself includes an apostrophe, in which case you must enclose the object name within double quotation marks. Valid examples include the following: tableid "xyz.mark's table" tableid "xyz"."mark's table" Invalid examples include the following: tableid 'xyz.mark's table' tableid "xyz".'mark's table' The object name has a period, in which case you must type the fully qualifying tablename in the form of database. tablename or 'database'.'tablename'. Valid examples include the following: tableid "xyz.0's"."mark's table " tableid 'xyz'.'table.0' Invalid examples include the following: tableid "xyz.0's.mark's table " tableid 'xyz.0's.mark's table ' Chapter 3: Filer Utility (filer) TABLEID Support Utilities, Release

208 Chapter 3: Filer Utility (filer) TIME TIME See DATE/TIME. 208 Support Utilities, Release 16.00

209 TRACEPRINT Chapter 3: Filer Utility (filer) TRACEPRINT Purpose The TRACEPRINT command dumps the internal trace buffer of the file system. Syntax DUMPCOPY or DC cylid LAST number = Displays file system traces from a local trace copy buffer. When a file system error is detected, the trace copy buffer stores a copy of the traces from when the error was detected. These traces are likely to contain information about the cause of the error. Although the regular system trace buffer also contains traces, early traces can be overwritten due to the volume of trace information written, especially for larger systems. In these cases, use the DUMPCOPY option to investigate the initial trace information that is retained in the trace copy buffer. This option is available only when Filer is invoked from the System Debugger or started in Coroner mode. The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. When a cylid is provided, the output is limited to traces associated with that cylid. Only the last numberentries are displayed. Filer will use the most recently saved values for cylid. This option cannot be used unless an input has already been defined for cylid.for more information, see Stored Input Values. Support Utilities, Release

210 Chapter 3: Filer Utility (filer) TRASH TRASH Purpose The TRASH command modifies the FSG segment to remove the associated disk address. Access to trashed segments is retained, but they will not be written to disk. Syntax \Y Prevents Filer from prompting for confirmation before executing the command. 210 Support Utilities, Release 16.00

211 VPROC Chapter 3: Filer Utility (filer) VPROC The VPROC command is identical to the SCOPE command. For more information, see SCOPE (or VPROC). Support Utilities, Release

212 Chapter 3: Filer Utility (filer) WABC WABC Purpose The WABC (WAL Append Buffer Control) command displays the WABC structure. Syntax Note: The online help lists the display options (/S, /M, /L) as /dispopt. /S /M /L Displays only the WABC structure. This is the default. Displays the Buffer Control Array, in addition to the information displayed by the /S option. Displays the Free Cylinder Array, in addition to the information displayed by the /S and /M options. 212 Support Utilities, Release 16.00

213 WAL Chapter 3: Filer Utility (filer) WAL Purpose The WAL (Write Ahead Logging) command returns a formatted display of WAL log records based on an ordinal record number range, such as the 10th through the 20th WAL log records. If no range is given, all of the records are displayed. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /X, /H, /V) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /X /H /V Positions to the WAL log records. Nothing is displayed unless there is a problem. Displays a count of the selected WAL log records. Displays the type, WLSN, and other generic WAL log record headers, in addition to the information displayed by the /S option. Displays the type-specific WAL log record headers for NONTJ and TJ records, in addition to the information displayed by the /M option. Displays the change data (before and after images) in hexadecimal, in addition to the information displayed by the /L option. Displays the header in hexadecimal, in addition to the information displayed by the /L option. Validates the WAL log records instead of displaying them. This option performs as much checking as possible without running SCANDISK, which requires starting the normal file system. See SCANDISK. Support Utilities, Release

214 Chapter 3: Filer Utility (filer) WAL ordinalrange whereclause Usage Notes Specifies the range of WAL log records to be displayed by ordinal position. If no range is given, all of the records are displayed. This differs from the TABLE command, which requires an ordinalrange argument. For more information, see Specifying Display Ranges (ordinalrange). The online help lists this option as rangeopt. Restricts the selection to a subset of the WAL log records within the ordinal range based on the field content of the records. For more information, see Specifying a WHERE Clause. The WAL command displays fully formatted WAL log records, including Transient Journal (TJ) subtypes. It displays before/after image content in hexadecimal. WAL formats the WAL log records according to the WAL log record kind. For more information on the WAL log record kinds and sub-kinds, see Specifying a WHERE Clause. WAL displays WAL log records by ordinal record number. To choose records based on the WAL log sequence number (WLSN) stored inside the records, use the WREC command. For more information, see WREC. The whereclause for the WAL command is applied in conjunction with the ordinalrange argument. The whereclause is recorded as part of the context and remains active for subsequent DISPLAY, NEXT, and PREV commands. Like the TABLE command, WAL has an initial context and a resulting context. The initial context is used to interpret the ordinalrange argument for the WAL command. The resulting context is produced by executing the WAL command. The initial context is unrelated to prior commands as explained below. The initial context encompasses the first WAL log record to the last, and the ordinalrange for the WAL command is interpreted with respect to this initial context. Therefore, the BEG and END identify the first and last WAL log records. Similar to the TABLE command, positive integers without the BEG/END prefix are considered to be ordinal record numbers, where 1 is the first record. Negative integers without the BEG/END prefix are backwards displacements from the last record, where -1 is the second to last record. Note: This differs from the interpretation of an ordinalrange for the DISPLAY command, where integers without the BEG/END prefix always refer to objects outside the current context, so that positive values refer to objects past the end of the current context, and negative values refer to objects before the beginning of the current context. For more information, see TABLE, DISPLAY, and Specifying Display Ranges (ordinalrange). The resulting context from a WAL command is a WAL log record or WAL log record range that matches the records selected. The resulting context from a TABLE command is a whole subtable, a subtable range, or null, but never a row range. For the TABLE command, the ordinalrange can restrict the display to a subset of the resulting context, even though this ordinalrange has no effect on the resulting context. The WAL command does not adopt this behavior. 214 Support Utilities, Release 16.00

215 WCI Chapter 3: Filer Utility (filer) WCI Purpose The WCI (WAL Cylinder Index) command accesses the WCI of a given cylinder of the WAL log and displays the WCI in accordance with the display options. The FIX option of the WCI command should be used by trained Teradata personnel only, and the user must be absolutely sure that the option should be used. Inappropriate use of the FIX option may result in data loss. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /X, /V) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /H /X /V FORCE Positions to the WCI. If this cylinder is not a WAL cylinder, an error occurs. Nothing is displayed unless there is a problem. Displays the header, trailer, and the first and last WDBDs. Displays the same information as the /S option. Displays all of the WDBDs, in addition to the information displayed by the /S option. Displays the WCI in hexadecimal. Displays the timestamp and other details from the WDB for each WDBD, in addition to the information displayed by the /L option. Verifies that the WCI is logically correct, but does not display it. Attempts to format and display the segment as a WCI, even if the segment is not a WCI. If the FORCE option is not used, the command fails on a mismatched block. Support Utilities, Release

216 Chapter 3: Filer Utility (filer) WCI FIX cylid = This option should be used by trained Teradata personnel only, and the user must be absolutely sure that the option should be used. Inappropriate use of this option may result in data loss. Attempts automatic repair of the WCI. You must supply the /V option with the FIX option, and the Teradata Database must be down. The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. Specifies that Filer should use the most recently saved value of cylid. This option cannot be used unless an input value has already been defined for cylid. For more information, see Stored Input Values. 216 Support Utilities, Release 16.00

217 WCID Chapter 3: Filer Utility (filer) WCID Purpose The WCID (WAL Cylinder Index Descriptor) command displays the WCID associated with the requested WAL log cylinder. Syntax cylid = The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. Specifies that Filer should use the most recently saved value for cylid. This option cannot be used unless an input value has already been defined for cylid. For more information, see Stored Input Values. Usage Notes The prompt for this selector looks like this: MI:WCID ==> In addition to global and upper-level commands, the following lower-level commands are allowed at the MI:WCID prompt: DELETE DISPLAY NEXT PATCH PREV WRITE Support Utilities, Release

218 Chapter 3: Filer Utility (filer) WDB WDB Purpose The WDB (WAL Data Block) command displays WAL log data blocks. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /H, /X, /V) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /H /X /V FORCE cylid Positions to the WDB. If one does not exist, an error occurs. Nothing is displayed unless there is a problem. Displays the header, trailer, and the first and last generic WAL log record headers. Displays all of the generic WAL log record headers, in addition to the information displayed by the /S option. Displays the type-specific WAL log record headers for NONTJ and TJ records, in addition to the information displayed by the /M option. Displays the header followed by the entire WDB in hexadecimal. Displays the change data (before and after images) in hexadecimal, in addition to the information displayed by the /L option. Verifies that the WDB is logically correct, but does not display it. Attempts to format and display the segment as a WDB, even if the block type does not indicate a WDB. The 64-bit identifier of a cylinder, entered as a 16-character hexadecimal number. 218 Support Utilities, Release 16.00

219 sectornum sectorcount = Usage Notes The sector number. The sector count. Specifies that Filer should use the most recently saved values for cylid, sectornum, or sectorcount. This option cannot be used unless input values have already been defined for cylid, sectornum, or sectorcount. For more information, see Stored Input Values. Chapter 3: Filer Utility (filer) WDB The checksum in the WDB is intended to cover the time between writing the WDB to disk and having filled the cylinder, finally writing the WCI and its WDBs to disk. The WCI is written once when the cylinder is initialized and once when the cylinder is full. The WCI is not written as each WDBD is added to the WCI in memory. In reading the WDBs in the last log cylinder after a restart that lost cache, no WDBDs with external checksums exist on the disk; therefore, internal checksums in the WDBs are used to validate the WDBs. Support Utilities, Release

220 Chapter 3: Filer Utility (filer) WDBD WDBD Purpose The WDBD (WAL Data Block Descriptors) command shows a formatted display of the WDBDs in the current WCI. Syntax Note: The online help lists the display options (/S, /X, /H) as /dispopt. /S /X /H WLSN = WlsnNum Displays the WDBDs in a short format. This is the default. Displays the WDBDs in an extended format. Displays all of the WDBDs, followed by a display of the WDBDs in hexadecimal. Specifies a WAL log sequence number for which WDBDs will be displayed. WlsnNum is an 64-bit value, the format of which is determined by the radix setting. For more information, see RADIX. SECTOR = sectornum Usage Notes Specifies a sector number for which WDBDs will be displayed. sectornum is a hexadecimal or decimal value, subject to the radix setting. For more information, see RADIX. WDBDs are used to determine which WDBs exists and where they are located on a cylinder. The WDBD command must be invoked from a WCI or WCI:WDBD context. The resulting context is the selected range of WDBDs within the WCI. 220 Support Utilities, Release 16.00

221 WDBREC Chapter 3: Filer Utility (filer) WDBREC Purpose The WDBREC command displays WAL log records for a WAL data block (WDB). Syntax Note: The online help lists the row options (/S, /M, /L, /H, /X) as /rowopt. /S or /C /M /L /H /X Displays a count of the WAL log records in the WDB. Displays the generic WAL log record headers. Displays the type-specific WAL log record headers for NONTJ and TJ records, in addition to the information displayed by the /M option. Displays the WAL log record header in hexadecimal, in addition to the information displayed by the /L option. Displays the change data (before and after images) in hexadecimal, in addition to the information displayed by the /L option. Usage Notes The WDBREC command must be invoked from a WDB context (as indicated by the WDB> prompt). Support Utilities, Release

222 Chapter 3: Filer Utility (filer) WFIND WFIND Purpose The WFIND command displays the WAL log cylinders containing WAL log records in a specified range of WAL log sequence numbers or, if no range is specified, displays all WAL log cylinders. Syntax Note: The online help lists the display options (/S, /M, /L) as /dispopt. You can type the options at the beginning or end of the command. /S /M /L wlsnrange Displays the first WCID and first WDBD within the range. Displays the first WCID, the first WDBD, all intervening WCIDs, and the last WCID and last WDBD within the range. Displays all WCIDs and WDBDs within the range. Specifies one or a range of WAL cylinders and WAL data blocks based on the WAL log sequence number (WLSN) in the record header. To define wlsnrange, use this syntax: 222 Support Utilities, Release 16.00

223 WFLUSH Chapter 3: Filer Utility (filer) WFLUSH Purpose The WFLUSH command forces some or all of the WAL log records in memory to disk. Syntax wlsn Flushes the WAL log to disk, up to and including the memory buffer containing the specified WAL log sequence number (WLSN). Usage Notes Blocks of WAL log records are accumulated in memory buffers. When the current buffer fills, a write is initiated, and a new buffer becomes the current buffer (that is, a buffer switch occurs). A transaction can force the log to disk, up to a specified WLSN, to insure that the transaction is recorded safely. You can use the WFLUSH command to force some or all of the buffers to disk manually. If writes are already in progress, WFLUSH waits for those writes to complete. If the current buffer is flushed, a buffer switch occurs. Flushing the WAL log to disk is a very simple, low risk operation. One reason to flush the WAL log to disk is so that you can use normal Filer commands, such as WREC and WAL, to examine recent log records. This might be necessary because Filer does not examine the log buffers when displaying log records. For more information about the WREC or WAL commands, see WREC or WAL. Support Utilities, Release

224 Chapter 3: Filer Utility (filer) WMI WMI Purpose The WMI (WAL Master Index) command returns a formatted display of the WMI. Syntax Note: The online help lists the display options (/P, /S, /L, /H, /V) as /dispopt. /P /S /L /H /V Positions to the WMI. Nothing is displayed unless there is a problem. Displays the header, trailer, and the first and last WCIDs of the WMI. Displays all the WCIDs in the WMI, in addition to the information displayed by the /S option. Displays the WMI in hexadecimal. Verifies that the WMI is logically correct, but does not display it. Usage Notes The WMI command is also available when Filer is invoked from the System Debugger or Coroner. For information on invoking Filer from the System Debugger or Coroner, see Running Filer from the System Debugger and Coroner. 224 Support Utilities, Release 16.00

225 WREC Chapter 3: Filer Utility (filer) WREC Purpose The WREC command returns a formatted display of WAL log records based on the WAL log sequence number (WLSN) in the record header. If no range is given, all of the records are displayed. Syntax Note: The online help lists the display options (/P, /S, /M, /L, /X, /H, /V) as /dispopt. You can type the options at the beginning or end of the command. /P /S /M /L /X /H /V Positions to the WAL log record. Nothing is displayed unless there is a problem. Displays a count of the selected WAL log records. Displays the type, WLSN, and other generic WAL log record headers, in addition to the information displayed by the /S option. Displays the type-specific WAL log record headers for NONTJ and TJ records, in addition to the information displayed by the /M option. Displays the change data (before and after images) in hexadecimal, in addition to the information displayed by the /L option. Displays the header in hexadecimal, in addition to the information displayed by the /L option. Validates the WAL log records instead of displaying them. This option performs as much checking as possible without running SCANDISK, which requires starting the normal file system. For more information, see SCANDISK. Support Utilities, Release

226 Chapter 3: Filer Utility (filer) WREC wlsnrange Selects one or a range of WAL log records based on the WAL log sequence number (WLSN) in the record header, instead of selecting the records by ordinal position within the log. If you do not specify a wlsnrange, the whole WAL log is displayed. To define wlsnrange, use this syntax: whereclause Usage Notes Restricts the selection to a subset of the WAL log records within the range based on the field content of the records. For more information, see Specifying a WHERE Clause. The wlsnrange selects records based on the WAL log sequence number (WLSN) recorded inside the log records. To select records by ordinal record number (such as 1st, 10th, or 5th WAL log record from the end), use the WAL command, which accepts an ordinalrange argument. For more information, see WAL. The whereclause for the WREC command is applied in conjunction with the wlsnrange. The whereclause is recorded as part of the context and remains active for subsequent DISPLAY, NEXT, and PREV commands. 226 Support Utilities, Release 16.00

227 WRITE Chapter 3: Filer Utility (filer) WRITE Purpose The WRITE command forces the currently selected segment to disk. Syntax /Y /NOCHECKSUM Prevents Filer from prompting for confirmation before executing the command. If checksums are enabled on the segment being modified or written, the checksum is not updated and written to the external segment that contains the checksum. Support Utilities, Release

228 Chapter 3: Filer Utility (filer) WSUMMARY WSUMMARY Purpose The WSUMMARY command scans the WAL log and provides a list of table identifiers referenced by non-tj WAL records within a specified range of WAL log sequence numbers (WLSNs). Syntax wlsnrange Selects one or a range of WAL log records based on the WAL log sequence number (WLSN) in the record header. If wlsnrange is not specified, WSUMMARY retrieves table IDs from all the WAL records. To define wlsnrange, use this syntax: 228 Support Utilities, Release 16.00

229 Filer Error Messages All Filer error messages are directed by default to your system console screen. Filer and file system error messages can be redirected through use of the ERRORS command. For information on file system messages, see Messages. Chapter 3: Filer Utility (filer) Filer Error Messages Support Utilities, Release

230 Chapter 3: Filer Utility (filer) Filer Error Messages 230 Support Utilities, Release 16.00

231 CHAPTER 4 Reconfiguration Estimator (reconfig_estimator) Overview Run the Reconfiguration Estimator utility, reconfig_estimator, before running the Reconfiguration utility described in Reconfiguration Utility (reconfig). Reconfiguration Estimator estimates the reconfiguration time based upon the number and size of tables on your current Teradata Database system. Reconfiguration Estimator provides time estimates for the following phases: Redistribution Deletion NUSI building Reconfiguration Estimator also validates the existence of tables enumerated in the ordering lists for the redistribution and deletion phases of reconfiguration. Note: Reconfiguration Estimator provides the most accurate estimates for add AMP operations. Estimates for other operations, such as moving or deleting AMPs may be less accurate. Runs From The Reconfiguration Estimator utility runs from Database Window or comparable interface to the Teradata Database console subsystem, such as cnsterm. Reconfiguration Estimator should never be stopped using the Database Window stop command. It can be stopped using the Reconfiguration Estimator ABORT command during the following phases: Entry of Reconfiguration Estimator parameters Pseudo Reconfig (the simulated reconfiguration performed to estimate processing time) Table level estimates. Reconfiguration Estimator cannot be stopped while data loading is in progress. The data loading phase is noted in the utility output screen. The user account used to run Reconfiguration Estimator should be empty, and must be granted the following privileges on itself: CREATE DATABASE DROP DATABASE CREATE TABLE DROP TABLE INSERT Support Utilities, Release

232 Chapter 4: Reconfiguration Estimator (reconfig_estimator) Usage Notes DELETE For example: GRANT CREATE DATABASE ON REuser TO REuser For general information on starting the utilities from different interfaces, see Starting the Utilities Usage Notes Reconfiguration Estimator time estimates are affected by how busy the system is while the estimates are being generated. The accuracy of estimates is improved when other system work is minimized, and best when Reconfiguration Estimator runs on an idle system. Reconfig Estimator performs a sample run of the Reconfiguration utility in order to determine estimation constants that are appropriate for the system. Because this process involves inserting and deleting sample data, the program requires the user name and password of an account that has rights to create and insert table rows. Create a user account exclusively for the use of Reconfiguration Estimator. Grant the account all privileges specified above. The account should have at least.5 to 1 GB of perm space per system AMP. Best estimates are obtained when Pseudo Reconfig (as noted in the utility output) runs for five to ten minutes on an idle system. The minimal time required to obtain reasonable estimates is proportional to how busy the system is while the estimates are being calculated. For example, if the system is only 10% idle, best estimates are obtained if the utility runs for 50 to 100 minutes. If Reconfiguration Estimator runs for less than an optimal time, run the utility again, after increasing the amount of perm space assigned to the account used by Reconfig Estimator. If Reconfiguration Estimator encounters any problems, the estimation process aborts and returns an error message. Note: Do not run multiple instances of Reconfiguration Estimator simultaneously. For more information on the reconfiguration process, see Reconfiguration Utility (reconfig). Example Reconfiguration Estimator Session Reconfig estimator Utility Enter the total number of AMPs in the old configuration:> 8 Enter the total number of AMPs in the new configuration:> 16 Enter the total number of Moved AMPs in the new configuration:> 0 Enter the total number of Nodes in the new configuration:> 2 Enter the number of tasks you want to run during reconfig:> 15 Enter an existing Username and Password for Loading Sample Data 232 Support Utilities, Release 16.00

233 or Press the Enter key to quit: Chapter 4: Reconfiguration Estimator (reconfig_estimator) Example Enter UserName:> rkpest Enter Password:> rkpest Loading data for running pseudo reconfig at 16/11/17 10:56:36.67 Please wait... Loading is complete at 16/11/17 10:57:41.31 Pseudo Reconfig Begins at 16/11/17 10:57:41.58 Please wait... Pseudo Reconfig Ends at 16/11/17 10:58:02.11 WARNING:Duration of Sample Reconfig is less than Optimal. WARNING:Estimates may not be correct. Value of Constant for New Estimator is: Compress DBlk Sort Redist Deletion TableSize Ratio RowCount NUSICount FB Size Flag Estimate Estimate Database.Table KB Y 126KB N 00:00: :00:00.00 DBC.ReconfigTableStatsTbl (0000H 00E3H) 10.00KB Y 31KB N 00:00: :00:00.00 SYSUIF.INSTALL_FILE (0000H 09FFH) 10.00KB Y 31KB N 00:00: :00:00.00 SYSUIF.REPLACE_FILE (0000H 0A00H) 10.00KB Y 31KB N 00:00: :00:00.00 SYSUIF.REMOVE_FILE (0000H 0A01H) 10.00KB Y 31KB N 00:00: :00:00.00 SYSUIF.REDISTRIBUTE_FILE (0000H 0A02H) 9.00KB Y 126KB N 00:00: :00:00.00 TD_SERVER_DB.TD_PshProInf (0000H 0AA5H) 11.00KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.TD_GEN_BINARY_POS (0000H 0AAFH) 11.00KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.TD_GEN_EXP_POS (0000H 0AB0H) 11.00KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.TD_MERGE_PP (0000H 0AB1H) 11.00KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.TD_MERGE_EXP_PP (0000H 0AB2H) 10.00KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.TD_CON_GET_SUPVER (0000H 0AB3H) KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.splrto_ovrwrprfle (0000H 0AB7H) KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.splrto_showexppro (0000H 0AB9H) 6.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.NUMFPFNS (0000H 03E9H) 6.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.TD_TUNABLE (0000H 03EAH) 6.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.NVP (0000H 03EBH) 6.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.NVP2 (0000H 03ECH) 6.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.NVP3 (0000H 03EDH) 6.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.NVP4 (0000H 03EEH) 6.00KB Y 126KB N 00:00: :00:00.00 SYSLIB.dem (0000H 03EFH) 6.00KB Y 126KB N 00:00: :00:00.00 SYSLIB.demddl (0000H 03F0H) 6.00KB Y 126KB N 00:00: :00:00.00 SYSLIB.dempart (0000H 03F1H) 11.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.installpkg (0000H 03F2H) KB Y 126KB N 00:00: :00:00.00 DBC.TempStatistics (0000H 0175H) 6.00KB Y 126KB N 00:00: :00:00.00 DBC.ReconfigRedistOrderTbl (0000H 00EAH) 6.00KB Y 126KB N 00:00: :00:00.00 DBC.ReconfigDeleteOrderTbl (0000H 00EBH) 11.21MB ,901 1 Y 126KB Y 00:00: :00:00.42 DBC.TVM (0000H 014FH) 56.00KB Y 126KB Y 00:00: :00:00.00 DBC.Dbase (0000H 0154H) 15.00KB Y 126KB N 00:00: :00:00.00 DBC.ReconfigInfoTbl (0000H 0173H) The current configuration has: 1 Nodes with 8 AMPs The new configuration will be: 2 Nodes with 16 AMPs There are 10 AMPs added to the new configuration The system has: 1053 tables using 70.69MB of data The estimated table redistribution time will be: 0.00 to 0.00 hours (for offline reconfig). The estimated table deletion time will be: 0.00 hours (for offline reconfig). This reconfig estimate is based upon 48XX/49XX/52XX OR LATER. Support Utilities, Release

234 Chapter 4: Reconfiguration Estimator (reconfig_estimator) Example 234 Support Utilities, Release 16.00

235 CHAPTER 5 Reconfiguration Utility (reconfig) Overview The Reconfiguration and Configuration utilities are used to define the AMPs and PEs that operate together as a Teradata Database. (Configuration is described in Configuration Utility (config).) Reconfiguration is the second step of the configuration and reconfiguration process that defines or modifies a Teradata Database system. Reconfiguration implements the Teradata Database system that is described in the new configuration map created in a previous Configuration utility session. The Reconfiguration utility is most often used when adding nodes to an existing configuration. In addition, use the Reconfiguration utility to change the following: Relationships of AMP and PE vprocs to nodes Clustering of AMP vprocs Other steps in configuring a system involve the Parallel Database Extensions (PDE) portions of Teradata Database, and are configured with the PUT utility. For more information on these configuration steps, see the Parallel Upgrade Tool (PUT) Reference, available from For more information on considerations when planning a system reconfiguration, see Teradata Knowledge Article KAP1AF072, available from Your Service ( Improper use of the Reconfiguration utility can result in data loss. Contact the Teradata Support Center before performing a system reconfiguration. Runs From The Reconfiguration Utility runs from Database Window or comparable interface to the Teradata Database console subsystem, such as cnsterm. Note: The reconfiguration process requires that no users be logged on to Teradata Database, and that the system be quiescent. Reconfiguration ends with a system restart. During reconfiguration, the system is unavailable to users. For general information on starting the utilities from different interfaces, see Starting the Utilities Configuration Maps and Reconfiguration The Configuration utility describes the complete Teradata Database system in the new configuration map area. Support Utilities, Release

236 Chapter 5: Reconfiguration Utility (reconfig) Disk Space for Reconfiguration As the Teradata Database system grows and changes, use Configuration to revise the new configuration map to reflect the following types of changes to the Teradata Database system: Addition and deletion of vprocs and hosts Changes to cluster assignments Use the configuration maps to perform these tasks: Store the identification and status of each vproc in the Teradata Database system Identify the AMPs that comprise each AMP cluster Identify each PE and its associated host The Teradata Database system contains these two configuration maps: The current configuration map, which describes the current arrangement and status of vprocs in the Teradata Database system The new configuration map, which includes changes and additions to the configuration The new configuration map serves as input to the Reconfiguration utility. Typically, Reconfiguration is used to alter the number of AMPs in a Teradata Database system. The new configuration map includes the status of each AMP in a Teradata Database system. After Configuration builds a new configuration map, Reconfiguration redefines the Teradata Database system configuration according to the new map. Reconfiguration copies the new configuration map to the current configuration map on each node. During reconfiguration, data is redistributed among AMPs if AMPs were added or deleted, or if cluster assignments were changed. Redistribution causes hash bucket assignments to be recalculated. These assignments describe the AMPs responsible for primary data and the AMPs responsible for fallback data. Hash bucket arrays contain hash bucket assignments for each AMP: Primary data rows for the current configuration Fallback data rows for the current configuration Primary data rows for the new configuration Fallback data rows for the new configuration Information that defines the new configuration is used during reconfiguration. Disk Space for Reconfiguration When changing cluster assignments or hash bucket sizes, ample disk space must be available on all AMPs. If sufficient space is not available, Teradata Database stops. To recover, reinitialize Teradata Database using the sysinit utility. Note that reinitializing the system destroys all data, which would need to be restored from an archive backup. Before you start a reconfiguration that involves a change of clusters, currentperm space should be less than 53% of the total maxperm space (that is, at least 47% of system maxperm must be available). Before Starting Reconfiguration Before starting Reconfiguration: 236 Support Utilities, Release 16.00

237 Verify that hardware is in place, properly configured, and online. Hardware configuration is detailed in the hardware service manuals for your Teradata Database system. Use the Vproc Manager utility to ensure all AMP-to-PE vprocs are online and ready. For more information, see the Vproc Manager (vprocmanager) chapter of Utilities. Ensure adequate free space is available for reconfiguration workspace. Depending on the number of fallback tables in existence and changes in DBS clustering, reconfig can require a significant amount of free space to distribute data rows. This involves temporarily rewriting the rows to a new location. Preventing and Correcting Skew Chapter 5: Reconfiguration Utility (reconfig) Before Starting Reconfiguration To prevent skewing of a NoPI table or join index (JI) during a reconfiguration to fewer AMPs: If the data in a NoPI table does not need to be retained, delete the rows from the table prior to reconfiguration. If the data in a NoPI table needs to be retained, copy the NoPI table into a PI table and delete the rows from the NoPI table prior to reconfiguration. Reinsert the rows into the NoPI table after the reconfiguration. For a NoPI JI that is not empty (due to base table not being empty), if possible drop the JI prior to reconfiguration and recreate it after reconfiguration. To smooth out the data skew that occurs for a nonempty NoPI table or JI after a reconfiguration to more AMPs (or fewer AMPs for a NoPI table for which rows were not deleted prior to the reconfig): Copy a NoPI table to a new NoPI table using the HASH BY RANDOM option, then drop the old NoPI table after dropping any JIs and create new JIs as needed on the new NoPI table. Grant privileges on the new NoPI table as needed. A nonempty NoPI JI should be dropped and recreated after the reconfiguration in order to remove skew. To prevent skewing of a table or JI having a primary AMP index (a PA table or join index) during a reconfiguration to fewer AMPs: If the data in a PA table does not need to be retained, delete the rows from the table prior to reconfiguration. If the data in a PA table needs to be retained, copy the PA table into a PI table and delete the rows from the PA table prior to reconfiguration. Reinsert the rows into the PA table after the reconfiguration. For a PA JI that is not empty (due to base table not being empty), if possible drop the JI prior to reconfiguration and recreate it after reconfiguration. After reconfiguration, rows for a nonempty PA table are not necessarily assigned to the correct AMPs per the new configuration, and the data is skewed. Because of the incorrect distribution of the rows the PA table is marked for limited access. Only SELECTs are allowed on these tables. To smooth out the data skew that occurs for a nonempty PA table or JI after a reconfiguration to more AMPs (or fewer AMPs for a PA table for which rows were not deleted prior to the reconfig) and to have rows distributed to the AMPs based on the PA, copy the PA table to a new PA table, then drop the old PA table after dropping any JIs and recreate JIs as needed on the new PA table. Grant privileges on the new PA table as needed. Alternatively, you can alter an nonempty PA table to a NoPI table in order to quickly regain full access to the table. At some convenient later time, copy the NoPI table (that was formerly a PA table) to a new PA table, then drop the old NoPI table after dropping any JIs and recreate JIs as needed on the new PA table. Grant privileges on the new PA table as needed. This regains the PA and removes the skew. A nonempty PA JI should be dropped and recreated after the reconfiguration in order to remove skew and correctly distribute the data. Support Utilities, Release

238 Chapter 5: Reconfiguration Utility (reconfig) Before Starting Reconfiguration Specifying the Order of Table Processing Reconfiguration allows user specification of the order in which tables are processed during the redistribution and deletion phases of reconfiguration. During the deletion phase, Reconfiguration deletes duplicate copies of rows that have been redistributed. Teradata recommends explicit specification of this order. Although the rationale for specifying a processing order during these two phases is slightly different, the strategies are the same: The goal of ordering tables for the redistribution phase of reconfiguration is to ensure that the largest tables are started first, while avoiding starting too many large redistribution jobs at the same time. The goal of ordering tables for the deletion phase of reconfiguration is to begin the deletion of larger tables early in this phase, because it takes longer to delete larger tables. In both cases, the preferred strategy is to process the largest table first, followed by a small or empty table, then continue interleaving large and small tables for processing. The ordering lists for both phases should match. To specify the order of table processing during the redistribution phase, insert a row for each table to be redistributed into DBC.ReconfigRedistOrderV: The DatabaseName and TableName fields specify a table. The OrderNumber field should contain an integer, the value of which indicates when the table will be processed relative to other tables in the list. To specify the order of table processing during the deletion phase, create a separate list by inserting rows into DBC.ReconfigDeleteOrderV. For more information on the structure of these tables, see Data Dictionary. If the order for processing is not explicitly specified, tables are processed in table ID order. Run the CheckTable Utility Run the CHECK command of the CheckTable utility at the PENDINGOP level to verify that no tables are in pending status: CHECK ALL TABLES AT LEVEL PENDINGOP SKIPLOCKS PRIORITY = H; Note: Do not use the CHECK command with the PARALLEL option when using the PENDINGOP option and with users logged on. A hang will occur. For more information, see the CheckTable (checktable) chapter in Utilities. Ensure that online archive logging is not enabled for any database object. Reconfiguration will not run if online archive logging is enabled. Use the LOGGING ONLINE ARCHIVE OFF statement to turn off logging. You must specify the name of the database or the names of the tables for which logging is enabled. For more information, see SQL Data Definition Language. Disable logons, and verify that the Teradata Database system is in a quiescent state. After the reconfiguration is complete logons must be explicitly enabled again. Run the Reconfiguration Estimator Utility Use the Reconfiguration Estimator utility prior to running the Reconfiguration utility. Reconfiguration Estimator validates the existence of the tables in the ordering lists. For more information on Reconfiguration Estimator, see Reconfiguration Estimator (reconfig_estimator). 238 Support Utilities, Release 16.00

239 The Reconfiguration Process This section summarizes the reconfiguration process. Individual Reconfiguration commands are described in Reconfiguration Utility Commands. As Reconfiguration runs, it performs these functions: Checks the Teradata Database system status to ensure that reconfiguration is possible. Disk storage capacity is checked to ensure that the Teradata Database system has sufficient storage to accommodate the redistributed data in the event of a delete AMP reconfiguration. Reconfiguration terminates if the Teradata Database system does not have sufficient storage capacity. After Teradata Database system status is verified, new hash bucket arrays are calculated based on current and new configuration maps. Redistributes primary and fallback data. Unique secondary index subtables, if any, are redistributed also. For PA and NoPI tables, no rows are redistributed when adding AMPs. Deletes rows that were redistributed from AMPs on which they resided before the redistribution. Nonunique secondary indexes, if any, are rebuilt. Updates space accounting information, hash bucket arrays, and configuration maps. You start the Reconfiguration utility from Database Window or from a comparable interface to the Teradata Database console subsystem, such as the cnsterm Linux command-line tool. For general information on starting the utilities, see Starting the Utilities. After the utility is started, start the actual reconfiguration by typing reconfig at the Reconfiguration utility command prompt. For more information on the RECONFIG command, see RECONFIG. The WITH PAUSE option causes reconfiguration to pause before entering the auto-restart phase, the point beyond which reconfiguration will automatically restart if Teradata Database restarts. The point at which the auto-restart phase begins depends on the purpose of the reconfiguration: If reconfiguration is for adding, deleting, or moving AMPs, for cluster reassignments, the auto-restart phase begins at the beginning of table redistribution. For all other reconfiguration purposes, the auto-restart phase begins when that new primary hash map is saved. The reconfiguration proceeds sequentially in this order: Chapter 5: Reconfiguration Utility (reconfig) The Reconfiguration Process 1. During the initialization of reconfiguration, the existence of all Teradata Database system table headers on all AMPs is verified. If a table header is missing, the table ID and the corresponding vproc ID from which the header is missing are displayed in the Database Window, and at the end of the verification, reconfig displays the following message: DBS Table Header verification failed, the missing header is available on the Database Window. Repair all missing table headers before reattempting Reconfiguration. ***** Depress the ENTER key to reset DBS. At this point, you should switch to the Database Window to obtain a list of missing table headers before resetting the Teradata Database system. Note: If this error condition occurs, contact the Teradata Support Center for procedures to fix the missing headers. Support Utilities, Release

240 Chapter 5: Reconfiguration Utility (reconfig) The Reconfiguration Process 2. Before entering the hash map calculation phase, Reconfiguration checks all AMPs for active user sessions. If any users are logged on, Reconfiguration issues an error message and halts. 3. Reconfig calculates the hash map. 4. If you used the WITH PAUSE option to the RECONFIG command, Reconfiguration prompts: Reconfiguration pauses before the auto-restart phase. Do You Want to Abort? Type Y(es) or N(o). This is the last opportunity to stop the reconfiguration. 5. The Table Redistribution phase of Reconfiguration begins changes to data in the current configuration. Reconfiguration redistributes tables (including stored procedures, user-defined functions (UDFs), user-defined methods (UDMs), non-value-ordered join indexes, and non-value-ordered hash indexes). Tables are processed in the order specified in DBC.ReconfigRedistOrderV. In the table redistribution phase, Reconfiguration verifies that the table to be redistributed exists on all online AMPs before table redistribution is started. This integrity check identifies those database tables that are corrupted between initialization and this phase. If a database table is missing from one or more of the online AMPs, the application window shows this message: Table DBName. TBLName *** is missing from some of the online AMPs *** This table is skipped. The Table Redistribution phase is also the auto-restart phase of reconfiguration, beyond which the reconfiguration process will be restarted automatically if the system restarts. The auto-restart phase is the phase beyond which the reconfiguration must continue to completion and cannot be aborted. 6. Reconfiguration deletes moved rows from tables and rebuilds NUSIs. Tables are processed in the order specified in DBC.ReconfigDeleteOrderV, or in table ID order if no other order has been specified. 7. Reconfiguration saves the following: New primary hash map New fallback hash map Current primary hash map Current fallback hash map Current configuration map New configuration map Backup IDs 8. Reconfiguration deletes new hash maps. 9. Reconfiguration saves the bitmap hash table. 10. Reconfiguration updates the following: Disk space Vproc configuration 11. Reconfiguration displays this message: Restart DBS due to completion of Reconfiguration. System is about to reset. 240 Support Utilities, Release 16.00

241 IF... a Teradata Database system restart occurs when Reconfiguration is in the middle of the Hash Map calculation phase a restart occurs after the first message and before the second one you are adding new AMPs before Reconfiguration THEN... the following message appears: Hash Map Calculation Phase Begins. { Reconfig Phase one -- rcophas1 } Hash Map Calculation Phase Ends. Reconfiguration will not start automatically during Teradata Database system start up. a 6140 error message displays if you start Reconfiguration manually after the Teradata Database system comes up. 6140: RECONFIG aborted due to improper disk initialization procedure. In this case, do the following: Chapter 5: Reconfiguration Utility (reconfig) Reconfiguration Considerations a. Use the ctl utility to change the Start DBS setting on the Debug screen to Off. See the Control GDO Editor (ctl) chapter in Utilitiesfor more information. b. Restart Teradata Database. c. Use the Vproc Manager utility INITVDISK command to initialize the file system on the newly added AMPs. For more information, see the Vproc Manager (vprocmanager) chapter in Utilities. d. Use the ctl utility to set Start DBS back to On. e. Restart Teradata Database. f. Start the Reconfiguration utility from Database Window. 12. You must drop and rebuild value-ordered join indexes and value-ordered hash indexes before they can be used again. 13. If desired, you may convert tables that were converted from NoPI or PA before the reconfig back to their original type. 14. If desired, you may run CheckTable and SCANDISK on any or all tables. Reconfiguration Considerations MOVE AMP Operation In a MOVE AMP operation, Reconfiguration copies all rows from the Move-From AMP to the Move-To AMP. You can move a range of AMPs to another range of AMPs; however, a one-to-one correspondence between the Move-From AMPs and the Move-To AMPs must exist. The following sequence discusses the state of these AMPs before, during, and after the reconfiguration process. Support Utilities, Release

242 Chapter 5: Reconfiguration Utility (reconfig) Reconfiguration Considerations 1. The Parallel Upgrade Tool (PUT) is used to start the reconfiguration process. During this step, PUT identifies new AMPs on the system. Some of these new AMPs may have been added, and some may be targets of the MOVE AMP process. PUT must be used at this time to identify which AMPs are the moveto AMPs and which are the move-from AMPs. 2. The Configuration utility (config) is used to complete the DBS configuration of AMPs in regards to adding and moving AMPs, and defining the cluster configuration. The move-to AMPs and the movefrom AMPs specified in the config commands must match the MOVE AMP definitions specified during the PUT process 3. Prior to and during most of the reconfiguration process, the Move-To AMPs are configured as new AMPs, just as if you added them. Move-To AMPs have their own vproc numbers and associated storage. PUT sets up the relationship between an AMP and its storage. 4. The file system tracks which storage belongs to which AMPs by inserting the AMP number in the cylinder indexes written with each cylinder. This allows the file system to verify that multiple AMPs do not write to the same cylinder of storage. 5. When almost finished, Reconfiguration saves the Teradata Database configuration maps to indicate the new configuration. 6. Reconfiguration notifies the file system to update the cylinder indexes to indicate the change in storage ownership. 7. Reconfiguration sets a control flag to keep Teradata Database in a DOWN state and to initiate a restart. 8. While Teradata Database is in a DOWN state, PUT exchanges the storage mapping between the Move- From and Move-To AMPs. For additional information on the MOVE AMP command, see Configuration Utility (config). Fatal AMPs During a Move-AMP Operation When Reconfiguration completes a MOVE AMP operation, change the PUT configuration to associate the new storage and AMPs correctly. If the configuration is not changed, when Teradata Database restarts the Move-To AMPs are marked FATAL. Effects on Journal Tables Reconfiguration deletes all journal tables (active, saved, and restored subtables). Effects on PA and NoPI Tables There are special considerations for reconfiguring a system that includes PA or NoPI tables. Reconfiguring a system that includes PA or NoPI tables can result in data skew for those tables. Because the number of hash buckets per AMP in PA and NoPI tables is often small (usually only one per AMP), when a reconfiguration involves changes in the number of AMPs, some AMPs may end up with substantially more rows than other AMPs. Consequently, it is best to minimize the number of PA and NoPI tables on the system before performing a reconfiguration, especially in cases where NoPI staging tables are large. For more information on PA and NoPI tables, see Database Design and SQL Data Definition Language. Nonempty PA tables are marked as Reconfig-Aborted when: Changing hash bucket size for a system Changing the number of AMPs Redistributing rows of a PA or NoPI table can result in table skew, when some AMPs receive more rows than other AMPs. In those cases, permanent space can become an issue for those AMPs. The following can be done for skew and reconfig-aborted PA tables: 242 Support Utilities, Release 16.00

243 Chapter 5: Reconfiguration Utility (reconfig) Reconfiguration Utility Commands For a skewed NoPI table, create a new NoPI table, and use an INSERT...SELECT operation with HASH BY RANDOM clause to copy the rows from the original NoPI table to the new table. For a PA table marked as Reconfig-Aborted, create a new PA table and use an INSERT...SELECT operation to copy the rows from the original PA table to the new PA table. Note: Permanent space in the database is divided equally among the AMPs of the system. If the presence of PA or NoPI tables on a reconfigured system causes appreciable data skew, permanent space for the database may need to be increased to account for the AMPs that will need to store more data. For more information on PA and NoPI tables, see SQL Data Definition Language and Database Design. Reconfiguration Utility Commands A reconfiguration session consists of reconfiguring the Teradata Database system according to the new configuration map built by the Configuration utility. Reconfiguration presents a command-line environment that allows the entry of the following Reconfiguration commands, which are described in more detail in the sections that follow. Command RECONFIG STATUS STOP Function Begins a session that changes the Teradata Database system configuration according to the map provided by the Configuration utility. Pressing the F2 key issues a RECONFIG command if you run the Reconfiguration utility from Database Window. Determines the status of Reconfiguration. Stops Reconfiguration. Pressing the F3 key issues a STOP command if you run the Reconfiguration utility from Database Window. Help information is available only when Reconfiguration is run Database Window. To display help information, press the F7 key. Support Utilities, Release

244 Chapter 5: Reconfiguration Utility (reconfig) RECONFIG RECONFIG Purpose The RECONFIG command starts the reconfiguration process, which changes the Teradata Database system configuration. Prior to performing a reconfiguration, a new configuration map must have been produced using the Configuration utility. The command can also be used to change certain reconfiguration parameters after the process has begun. Syntax DISPLAY n TASKS whether to enable or disable output of Reconfiguration status (but not statistics): ON enables output. (Default) OFF disables output. the limitation of the number of Reconfiguration sessions running in parallel. n is a number from 0 through 10. The default is 10. Specifying 0 tasks pauses the reconfiguration. [CPU IO] PRIORITY p the job settings to use for the reconfiguration. The default is to assign the reconfiguration process a medium timeshare access level. Use the CPU and IO keywords to specify that the priority set is specific to either CPU or I/O time. If PRIORITY is specified with neither CPU nor IO modifiers, the priority specified is applied to both CPU and IO time. Available settings for p are determined by workload settings. For more information about setting workload priorities, see the Workload Designer section in the Teradata Viewpoint User Guide. Search for About Console Utilities. 244 Support Utilities, Release 16.00

245 PAUSE STATISTICS whether Reconfiguration is to pause before entering the auto-restart phase. Each AMP stores the current phase of reconfiguration. whether to enable or disable output of Reconfiguration statistics: Usage Notes ON enables statistics. (Default) OFF disables statistics. The RECONFIG command can be used throughout the reconfiguration process to change the values of TASK, PRIORITY, JOURNAL MAXPERMSPACE, and STATISTICS. Example: Running RECONFIG suppressing statistics output To disable statistics output, type: Chapter 5: Reconfiguration Utility (reconfig) RECONFIG reconfig with statistics off Example: Running RECONFIG with 10 parallel sessions To reconfigure with 10 control tasks and statistics output disabled, type: reconfig with 10 tasks statistics off Example: Running RECONFIG supressing all display output To reconfigure with eight control tasks and output display disabled, type: reconfig with 8 tasks display off Example: Typical RECONFIG output This example shows some output from the RECONFIG command. Reconfiguration output displays the following: Start and end time of each table processed during redistribution Run-time statistics of the table process that includes statistics for all AMPs This information is included with the deletion and NUSI building. The output might become excessive for a large number of tables, so you might want to disable DISPLAY. If you disable DISPLAY, you can still obtain status using the STATUS command. Support Utilities, Release

246 Chapter 5: Reconfiguration Utility (reconfig) RECONFIG Note: You can disable the table begin/completion output with the RECONFIG WITH DISPLAY OFF command. You can disable the statistics output with the RECONFIG WITH STATISTICS OFF command. 08/05/13 09:04:35 Task 01 Begin redistribution DBC.Global (0000H 0019H). 08/05/13 09:04:35 Task 00 Begin redistribution DBC.TransientJournal (0000H 001AH). 08/05/13 09:04:35 Task 00 End redistribution DBC.TransientJournal (0000H 001AH). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: /05/13 09:04:36 Task 02 Begin redistribution DBC.Owners (0000H 001BH). 08/05/13 09:04:36 Task 01 End redistribution DBC.Global (0000H 0019H). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: ,564 08/05/13 09:04:36 Task 00 Begin redistribution DBC.Parents (0000H 001CH). 08/05/13 09:04:36 Task 02 End redistribution DBC.Owners (0000H 001BH). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: ,831 08/05/13 09:04:36 Task 00 End redistribution DBC.Parents (0000H 001CH). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: ,943 08/05/13 09:04:36 Task 01 Begin redistribution DBC.ErrorMsgs (0000H 001EH). 08/05/13 09:04:36 Task 00 Begin redistribution DBC.ChangedRowJournal (0000H 0020H). 08/05/13 09:04:36 Task 00 End redistribution DBC.ChangedRowJournal (0000H 0020H). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: Support Utilities, Release 16.00

247 STATUS Chapter 5: Reconfiguration Utility (reconfig) STATUS Purpose The STATUS command allows you to determine the status of Reconfiguration at any time during the process. Syntax Usage Notes STATUS is available at any time during Reconfiguration and includes the status of the following: Total number of tables processed for the redistribution, deletion, and NUSI phases Total number of tables remaining to be processed for the redistribution, deletion, and NUSI phases Completion percentage according to the total number of bytes processed Status of each non-idle, parallel reconfiguration control task Completion percentage and name of table being processed for each non-idle, parallel reconfiguration control task Example: STATUS display during reconfiguration hash map calculation The following is an example of STATUS output during hash map calculation. Current reconfig phase: Hash Map Calculation Example: STATUS output during reconfiguration redistribution phase The following is an example of STATUS output during redistribution. Current reconfig phase: Redistribution Estimated time remaining: 00:19:21 Redistribution status - tables processed: 152 tables remaining: 26 70% of total bytes processed Deletion/NUSI status - tables processed: 0 tables remaining: 178 0% of total bytes processed Reconfig task 0 Processed 69% of RKPMED.SIW_XXXX_FB_12 (0000H 04C4H) Reconfig task 1 Processed 69% of RKPMED.SIW_XXXX_FB_7 (0000H 04BFH) Reconfig task 2 Processed 69% of RKPMED.SIW_XXXX_FB_11 (0000H 04C3H) Reconfig task 3 Processed 68% of RKPMED.SIW_XXXX_FB_15 (0000H 04C7H) Reconfig task 4 Processed 69% of RKPMED.SIW_XXXX_FB_8 (0000H 04C0H) Reconfig task 5 Processed 68% of RKPMED.SIW_XXXX_FB_13 (0000H 04C5H) Reconfig task 6 Processed 64% of RKPMED.SIW_XXXX_FB_14 (0000H 04C6H) Reconfig task 7 Processed 69% of RKPMED.SIW_XXXX_FB_9 (0000H 04C1H) Support Utilities, Release

248 Chapter 5: Reconfiguration Utility (reconfig) STATUS Reconfig task 8 Processed 69% of RKPMED.SIW_XXXX_FB_6 (0000H 04BEH) Reconfig task 9 Processed 69% of RKPMED.SIW_XXXX_FB_10 (0000H 04C2H) Example: STATUS output during reconfiguration deletion phase The following is an example of STATUS output during Deletion/NUSI rebuilding. Current reconfig phase: Deletion/NUSI Rebuild Current estimates are based upon current number of tasks. Estimated redistribution time remaining: 0.00 hours. Estimated deletion time remaining: 0.42 hours. Redistribution status - tables processed: 1,427 tables remaining: 0 100% of total bytes processed Deletion/NUSI status - tables processed: 1,217 tables remaining: % of total bytes processed Reconfig task 0 Processed 48% of RP_31864.tbl_col20_1_src (0000H 0BFAH) Reconfig task 1 Processed 43% of RP_31864.tbl_col20_74_src (0000H 0BFBH) Reconfig task 2 Processed 28% of RP_31864.tbl_col20_7 (0000H 0BFEH) Reconfig task 3 Processed 7% of RP_31864.tbl_col20_69_src (0000H 0C01H) Reconfig task 4 Processed 9% of RP_31864.tbl_col20_51_src (0000H 0C00H) Reconfig task 5 Processed 52% of RP_31864.tbl_col20_82_src (0000H 0BF9H) Reconfig task 6 Processed 2% of RP_31864.tbl_col20_86_src (0000H 0C02H) Reconfig task 7 Processed 26% of RP_31864.tbl_col20_61_src (0000H 0BFFH) Reconfig task 8 Processed 0% of RP_31864.tbl_col20_66_src (0000H 0C03H) Reconfig task 9 Processed 42% of RP_31864.tbl_col20_71_src (0000H 0BFCH) 248 Support Utilities, Release 16.00

249 STOP Chapter 5: Reconfiguration Utility (reconfig) STOP Purpose The STOP command stops Reconfiguration. Syntax Usage Notes The STOP command can be used only before the RECONFIG command is issued. Support Utilities, Release

250 Chapter 5: Reconfiguration Utility (reconfig) Example of a Reconfiguration Session Example of a Reconfiguration Session The following example shows the complete output from a system reconfiguration. Much of the detailed table redistribution text is eliminated for brevity. 250 Support Utilities, Release 16.00

251 Chapter 5: Reconfiguration Utility (reconfig) Example of a Reconfiguration Session Reconfiguration Session The RECONFIG program provides the user with a facility to redistribute data when the configuration of a DBS is changed by pressing the appropriate function key. System Time (Reconfiguration): 16/11/16 16:21: The following contiguous maps are defined on this system: TD_GlobalMap (Map Number 1) has Amps 0-3 and 2 PEs defined. TD_DataDictionaryMap (Map Number 2) has Amps 0-3 and 2 PEs defined. TD_Map1 (Map Number 1025) has Amps 0-3 and 2 PEs defined (System Default Map). TD_Map4 (Map Number 1033) has Amps 0-5 and 2 PEs defined (New Global Map). Enter command or Press <F2> for Reconfig command <F3> for Stop command <F4> for Query or Status command <F5> for Display Retry List command <F7> for help> reconfig Reconfig will use 20-bit bucket size. AMPs are added into configuration, do you want to adjust space for databases/users whose SpaceMap is set to TD_GLOBALMAP? (yes or no) <1:no no Compress DBlk Sort Redist Deletion TableSize Ratio RowCount NUSICount FB Size Flag Estimate Estimate Database.Table MB ,428 0 Y 126KB N 00:00: :00:00.12 DBC.ReconfigTableStatsTbl (0000H 00E3H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSUIF.INSTALL_FILE (0000H 09FFH) 8.00KB Y 31KB N 00:00: :00:00.00 SYSUIF.REPLACE_FILE (0000H 0A00H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSUIF.REMOVE_FILE (0000H 0A01H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSUIF.REDISTRIBUTE_FILE (0000H 0A02H) 6.00KB Y 126KB N 00:00: :00:00.00 TD_SERVER_DB.TD_PushProfileInfo (0000H 0AA5H) 9.00KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.TD_GEN_BINARY_PUSHOPSET (0000H 0AAFH) 9.00KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.TD_GEN_EXP_PUSHPROFILE (0000H 0AB0H) 9.00KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.TD_MERGE_PUSHPROFILE (0000H 0AB1H) 9.00KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.TD_MERGE_EXP_PUSHPROFILE (0000H 0AB2H) 8.00KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.TD_CONN_GET_SUPP_VERSION (0000H 0AB3H) KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.splrto_overwrite_profile (0000H 0AB7H) KB Y 31KB N 00:00: :00:00.00 TD_SERVER_DB.splrto_showexpanded_profile (0000H 0AB9H) 4.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.NUMFPFNS (0000H 03E9H) 4.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.TD_TUNABLE (0000H 03EAH) 4.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.NVP (0000H 03EBH) 4.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.NVP2 (0000H 03ECH) 4.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.NVP3 (0000H 03EDH) 4.00KB Y 126KB N 00:00: :00:00.00 TD_SYSFNLIB.NVP4 (0000H 03EEH) 4.00KB Y 126KB N 00:00: :00:00.00 SYSLIB.dem (0000H 03EFH) 4.00KB Y 126KB N 00:00: :00:00.00 SYSLIB.demddl (0000H 03F0H) Support Utilities, Release

252 Chapter 5: Reconfiguration Utility (reconfig) Example of a Reconfiguration Session 4.00KB Y 126KB N 00:00: :00:00.00 SYSLIB.dempart (0000H 03F1H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.installpkg (0000H 03F2H) 93.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.installsp (0000H 03F4H) KB Y 31KB N 00:00: :00:00.00 SYSLIB.savepkg (0000H 03F6H) KB Y 31KB N 00:00: :00:00.01 SYSLIB.loadpkg (0000H 03F8H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.GetTimeZoneDisplacement (0000H 03F9H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.GetISFDName (0000H 03FAH) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.GetQueryBand (0000H 03FBH) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.GetQueryBandPairs (0000H 03FCH) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.GetQueryBandPairs2 (0000H 03FDH) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.GetQueryBandValue (0000H 03FEH) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.GetQueryBandValue2 (0000H 03FFH) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.DBQLRowConcat (0000H 0400H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.FeatureNames_TBF (0000H 0404H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.GetQueryBandSP (0000H 0406H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.GetQueryBandValueSP (0000H 0407H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.AbortSessions (0000H 0408H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.AbortListSessions (0000H 0409H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.IdentifyDatabase (0000H 040AH) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.IdentifySession (0000H 040BH) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.IdentifyTable (0000H 040CH) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.IdentifyUser (0000H 040DH) 12.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorAWTResource (0000H 040EH) 12.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorAWTResource2 (0000H 040FH) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorAMPLoad (0000H 0410H) 24.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorMySessions (0000H 0411H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorPhysicalConfig (0000H 0412H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorSystemPhysicalConfig (0000H 0413H) 12.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorPhysicalResource (0000H 0414H) 11.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorPhysicalSummary (0000H 0415H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorQueryBand (0000H 0416H) 24.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorSession (0000H 0417H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorSessionRate (0000H 0418H) 10.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorSQLCurrentStep (0000H 0419H) 10.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorSQLSteps (0000H 041AH) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorSQLText (0000H 041BH) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorVersion (0000H 041CH) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorVirtualConfig (0000H 041DH) 11.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorVirtualResource (0000H 041EH) 13.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorVirtualSummary (0000H 041FH) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.SetResourceRate (0000H 0420H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.SetSessionAccount (0000H 0421H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.SetSessionRate (0000H 0422H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorWDRate (0000H 0423H) 14.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.MonitorWD (0000H 0424H) 9.00KB Y 31KB N 00:00: :00:00.00 syslib.gettvsudfviewpointsummary (0000H 0425H) 12.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.ExportTSET (0000H 0426H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.QueryBandReservedNames_TBF (0000H 0427H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.DBSRetryableErrors_TBF (0000H 0429H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.DBSDataRelatedErrors_TBF (0000H 042BH) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.SQLRestrictedWords_TBF (0000H 042DH) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.SpaPrintTrackInfoLocal (0000H 0432H) 8.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.SpaPrintTrackInfoGlobal (0000H 0433H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.SpaAggrTrackInfoGlobal (0000H 0434H) 9.00KB Y 31KB N 00:00: :00:00.00 SYSLIB.ExtractRoles (0000H 0435H) KB Y 126KB Y 00:00: :00:00.00 DBC.GlobalDBSpace (0000H 0171H) 53.00KB Y 126KB Y 00:00: :00:00.00 DBC.EventLog (0000H 0172H) 252 Support Utilities, Release 16.00

253 Support Utilities, Release KB Y 126KB N 00:00: :00:00.00 DBC.ResUsageSmhm (0000H 0174H) 4.00KB Y 126KB N 00:00: :00:00.00 DBC.TempStatistics (0000H 0175H) 4.00KB Y 126KB N 00:00: :00:00.00 DBC.ReconfigRedistOrderTbl (0000H 00EAH) 4.00KB Y 126KB N 00:00: :00:00.00 DBC.ReconfigDeleteOrderTbl (0000H 00EBH) 11.20MB ,888 1 Y 126KB Y 00:00: :00:00.46 DBC.TVM (0000H 014FH) 41.00KB Y 126KB Y 00:00: :00:00.00 DBC.Dbase (0000H 0154H) 16.00KB Y 126KB N 00:00: :00:00.00 DBC.ReconfigInfoTbl (0000H 0173H) The current configuration has: 1 Nodes with 4 AMPs The new configuration will be: 1 Nodes with 6 AMPs There are 2 AMPs added to the new configuration The system has: 1053 tables using 71.38MB of data The estimated table redistribution time will be: 0.00 hours (for offline reconfig). The estimated table deletion time will be: 0.00 hours (for offline reconfig). This reconfig estimate is based upon 48XX/49XX/52XX OR LATER. Reconfig waiting for Recovery to complete... Recovery has been stopped 16/11/16 16:21:24.08 Hash Map Calculation Phase Begins 16/11/16 16:21:25.33 Hash Map Calculation Phase Ends 16/11/16 16:21:25.34 Table Redistribution Phase Begins 16/11/16 16:21:25.37 Task 00 Begin redistribution DBC.ReconfigTableStatsTbl (0000H 00E3H)(No Sort). 16/11/16 16:21:25.46 Task 00 End redistribution DBC.ReconfigTableStatsTbl (0000H 00E3H). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: 2, MB ,256 16/11/16 16:21:25.47 Task 00 Begin redistribution SYSUIF.INSTALL_FILE (0000H 09FFH)(No Sort). 16/11/16 16:21:25.50 Task 00 End redistribution SYSUIF.INSTALL_FILE (0000H 09FFH). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB ,205 16/11/16 16:21:25.52 Task 01 Begin redistribution SYSUIF.REPLACE_FILE (0000H 0A00H)(No Sort). 16/11/16 16:21:25.54 Task 00 Begin redistribution SYSUIF.REMOVE_FILE (0000H 0A01H)(No Sort). 16/11/16 16:21:25.62 Task 00 End redistribution SYSUIF.REMOVE_FILE (0000H 0A01H). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB ,609 16/11/16 16:21:25.64 Task 01 End redistribution SYSUIF.REPLACE_FILE (0000H 0A00H). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB ,180,069,165 16/11/16 16:21:25.61 Task 02 Begin redistribution SYSUIF.REDISTRIBUTE_FILE (0000H 0A02H)(No Sort). 16/11/16 16:21:25.65 Task 00 Begin redistribution TD_SERVER_DB.TD_PushProfileInfo (0000H 0AA5H)(No Sort). 16/11/16 16:21:25.72 Task 02 End redistribution SYSUIF.REDISTRIBUTE_FILE (0000H 0A02H). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB , /11/16 16:22:03.80 Task 02 Begin redistribution DBC.Dbase (0000H 0154H)(Sort). 16/11/16 16:22:03.88 Task 00 Begin redistribution DBC.ReconfigInfoTbl (0000H 0173H)(No Sort). 16/11/16 16:22:03.93 Task 00 End redistribution DBC.ReconfigInfoTbl (0000H 0173H). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB ,624 16/11/16 16:22:03.94 Task 02 End redistribution DBC.Dbase (0000H 0154H). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB ,152 16/11/16 16:22:03.98 Task 01 End redistribution DBC.TVM (0000H 014FH). Statistics: RowCount ByteCount TotSecs CPUSecs IOCount AllAmps: 2, MB ,165 Chapter 5: Reconfiguration Utility (reconfig) Example of a Reconfiguration Session

254 254 Support Utilities, Release /11/16 16:22:04.36 Table Redistribution Phase Ends 16/11/16 16:22:04.38 Old Table Deletion Phase Begins 16/11/16 16:22:04.82 Task 00 Begin deletion DBC.RCConfiguration (0000H 0007H). 16/11/16 16:22:04.94 Task 01 Begin deletion DBC.Global (0000H 0019H). 16/11/16 16:22:05.04 Task 00 End deletion DBC.RCConfiguration (0000H 0007H). Statistics: FSysCallCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB ,434 16/11/16 16:22:05.05 Task 02 Begin deletion DBC.TransientJournal (0000H 001AH). 16/11/16 16:22:05.07 Task 00 Begin deletion DBC.Owners (0000H 001BH). 16/11/16 16:22:05.15 Task 01 End deletion DBC.Global (0000H 0019H). Statistics: FSysCallCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB ,389 16/11/16 16:22:05.17 Task 03 Begin deletion DBC.Parents (0000H 001CH). 16/11/16 16:22:05.18 Task 01 Begin deletion DBC.ErrorMsgs (0000H 001EH). 16/11/16 16:22:05.20 Task 04 Begin deletion DBC.ChangedRowJournal (0000H 0020H). 16/11/16 16:22:05.28 Task 05 Begin deletion DBC.SavedTransactionStatusTable (0000H 0024H). 16/11/16 16:22:05.36 Task 06 Begin deletion DBC.LocalTransactionStatusTable (0000H 0025H). 16/11/16 16:22:05.42 Task 07 Begin deletion DBC.UtilityLockJournalTable (0000H 0027H). 16/11/16 16:22:05.46 Task 08 Begin deletion DBC.LocalSessionStatusTable (0000H 0028H). 16/11/16 16:22:05.49 Task 02 End deletion DBC.TransientJournal (0000H 001AH). Statistics: FSysCallCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB , AllAmps: KB ,841 16/11/16 16:23:35.10 Task 03 End deletion TD_SERVER_DB.TD_CONN_GET_SUPP_VERSION (0000H 0AB3H). Statistics: FSysCallCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB ,903 16/11/16 16:23:35.11 Task 13 End deletion DBC.ReconfigDeleteOrderTbl (0000H 00EBH). Statistics: FSysCallCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB /11/16 16:23:35.12 Task 09 End deletion DBC.ReconfigRedistOrderTbl (0000H 00EAH). Statistics: FSysCallCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB /11/16 16:23:35.13 Task 11 End deletion DBC.Dbase (0000H 0154H). Statistics: FSysCallCount ByteCount TotSecs CPUSecs IOCount AllAmps: KB ,334 16/11/16 16:23:35.22 Task 10 End deletion DBC.TVM (0000H 014FH). Statistics: FSysCallCount ByteCount TotSecs CPUSecs IOCount AllAmps: MB ,218 16/11/16 16:23:35.28 Old Table Deletion Phase Ends 16/11/16 16:23:35.28 Saving Backup IDs Phase Begins 16/11/16 16:23:35.33 Saving Backup IDs Phase Ends 16/11/16 16:23:35.34 Insert/Update DBC.Maps Table Phase Begins 16/11/16 16:23:35.48 Insert/Update DBC.Maps Table Phase Ends 16/11/16 16:23:35.49 Saving Current Configuration Map Phase Begins 16/11/16 16:23:36.82 Saving Current Configuration Map Phase Ends 16/11/16 16:23:36.83 Saving New Configuration Map Phase Begins 16/11/16 16:23:36.84 Saving New Configuration Map Phase Ends 16/11/16 16:23:36.84 Saving Bitmap Hash Table Phase Begins 16/11/16 16:23:37.05 Saving Bitmap Hash Table Phase Ends 16/11/16 16:23:37.07 Updating Disk Space Phase Begins 16/11/16 16:23:37.43 Updating Disk Space Phase Ends 16/11/16 16:23:37.44 Updating Vproc Configuration Begins 16/11/16 16:23:37.46 Updating Vproc Configuration Ends System Time (Reconfiguration): 16/11/16 16:23: : Restarting DBS due to completion of reconfiguration. System is about to be reset for reconfig offline. Chapter 5: Reconfiguration Utility (reconfig) Example of a Reconfiguration Session

255 CHAPTER 6 System Initializer (sysinit) Overview The System Initializer utility, sysinit: Initializes the Teradata Database Creates or updates the DBS Control Record and other Globally Distributed Objects (GDOs) Initializes or updates configuration maps Allows you to set the hash function value in the DBS Control Record Runs From System Initializer runs from Database Window or comparable interface to the Teradata Database console subsystem, such as cnsterm. Type start sysinit to start System Initializer from Database Window. System Initializer completes after you respond to the last prompt in the user dialog. You can terminate sysinit early by entering Quit in response to certain prompts. For a description of these prompts and messages, see Running System Initializer. For general information on starting the utilities from different interfaces, see Starting the Utilities Before You Begin Before you run sysinit, you should understand the following concepts. Globally Distributed Objects Globally Distributed Objects, (GDOs) are binary files of structured data that are kept consistent across all nodes and vprocs in a Teradata Database system. GDOs typically store system settings and configuration information that is shared by all nodes of the system. Sysinit initializes several Teradata Database GDOs. Depending on the state of the DBS Control Record, GDOs, and on the current configuration when System Initializer is run, sysinit prompts for certain information, such as the hash function to use for row distribution among AMPs, and the type of language support required. For information on these prompts, see Running System Initializer. Configuration Maps Configuration maps define the current and new configuration of the Teradata Database system vprocs. A configuration map: Stores the identification and status of each vproc in the Teradata Database system Support Utilities, Release

256 Chapter 6: System Initializer (sysinit) Before You Begin Identifies the AMPs that constitute each AMP cluster Identifies each PE and its associated host The Teradata Database system contains two configuration maps: The current configuration map describes the current arrangement and status of vprocs in the Teradata Database system The new configuration map describes changes and additions to the configuration Sysinit can create a new configuration map and keep or revise a current map. For new configuration maps initialized by sysinit from the current configuration, the current configuration can retain one AMP and all existing PEs. The new configuration can retain one or all AMPs and all existing PEs. New Versus Previously Initialized Systems Running sysinit is part of the overall Teradata Database system initialization process. Sysinit handles Teradata Database system configuration differently, depending on whether the system is being initialized for the first time, or has been previously initialized on the same, or a previous version of Teradata Database: First-Time Initialization for New Systems When used to initialize a new Teradata Database system, sysinit initializes a configuration that includes all the vprocs defined in a vprocconfig GDO. On a new, uninitialized, Teradata Database system (or on a corrupted Teradata Database system whose configuration map cannot be read), sysinit creates a full configuration based on the vprocconfig GDO. All vprocs are set to an online, operational state. If a problem exists with any vproc, then a 0 PE, 1 AMP configuration is created. If a full configuration was created, then you are given options later for selecting a particular configuration for the new and old config maps. Previously Initialized Systems On a previously initialized Teradata Database system, sysinit retains the current configuration map and prompts you to do one of the following: Retain the current configuration in the new configuration map Select a different configuration for the current and the new configuration maps Notice: Sysinit will destroy all user and dictionary data currently in the Teradata Database system. Sysinit may reset the configuration settings stored in various GDOs and normally set using various system utilities to default values for the new version of Teradata Database. If you need to run sysinit on an existing system, first collect information from the following to ensure that customized settings can be restored and set as required after the sysinit: System Utilities: Control GDO Editor (ctl) DBS Control TDN Tuner (tdntune) Vproc Manager System Control Files: GDOs vconfig.out file UDF libraries (optional) 256 Support Utilities, Release 16.00

257 Running System Initializer Chapter 6: System Initializer (sysinit) Running System Initializer Notice: Running sysinit on a previously initialized Teradata Database system destroys all existing data and deletes all tables. Starting System Initializer Start system initializer by typing start sysinit from the Supervisor Window of Database Window or from a comparable Linux command-line console tool, such as cnsterm. The following message appears: W A R N I N G This program will destroy all user and dictionary data on the system. SYSINIT Master AMP Vproc is 0 at 11:25:42 on 04/04/11 If the Teradata Database system is running, sysinit displays: The DBS is currently running!!! SYSINIT cannot execute while the DBS is running. Would you like to restart the system without the DBS (YES/NO)? At the prompt, answer one of the following: IF you answer... THEN sysinit... YES sets the Start DBS configuration option to Off. The Start DBS setting can also be manually changed using the Control GDO Editor (ctl) utility. It determines whether the database software is started when PDE is started. For more information, see the Control GDO Editor (ctl) chapter in Utilities. displays the following message and exits: NO SYSINIT terminated without updating disks. automatically restarts PDE, the low level of Teradata Database, without starting the higher-level DBS database software. You must start sysinit again when PDE has restarted. displays the following message and exits: Set the "Start DBS" flag to "Off" using the CTL utility program and restart the system prior to running SYSINIT again. sysinit terminated without updating disks. The startup process for Teradata Database involves first starting parallel database extensions (PDE). These extensions create an operating environment that supports the parallel features of Teradata Database. Support Utilities, Release

258 Chapter 6: System Initializer (sysinit) Running System Initializer IF you answer... THEN sysinit... After PDE has started, the Teradata Database DBS software is usually automatically started. The Start DBS flag can be set to prevent the DBS software from starting after PDE is started. This flag is set using the Control GDO Editor (ctl) utility. For more information about the ctl utility, see the Control GDO Editor (ctl) chapter in Utilities. The following message appears when you restart sysinit: W A R N I N G This program will destroy all user and dictionary data on the system. A display indicating the console AMP vproc number, the current time, and the current date, as shown below: SYSINIT Master AMP Vproc is 0 at 11:25:42 on 04/04/11 Sysinit begins by attempting to read the DBS Control Record. The DBS Control Record is a Teradata Database globally distributed object (GDO) that contains system settings for Teradata Database. Sysinit handles the DBS Control Record differently, depending on whether the Teradata Database system was initialized previously or is being initialized for the first time. IF the DBS Control Record GDO... is corrupted or does not exist exists THEN sysinit displays... A new DBS Control Record will be initialized to default values. The existing DBS CONTROL GDO will be used. Sysinit displays information from the current DBS Control settings, and may prompt for further information or action based on those settings. Note: If the existing GDO is from the Teradata Database release immediately prior to the release being installed, the fields will be upgraded as necessary for the version installed. Choosing the Hash Algorithm and Language Support One of the DBS Control settings reflects the hash function Teradata Database will use for row distribution amongst AMPs. New Universal hash is the default, and is recommended for all new Teradata Database installations. 258 Support Utilities, Release 16.00

259 For existing Teradata Database installations, if the hash had previously been set to something other than New Universal or Pre Universal, sysinit asks for confirmation, and offers to change the hash type and language support mode. For new installations, sysinit asks if you wish to use New Universal Hash. If you respond No, you are able to select one of the older hash functions for backward compatibility. The following table describes the available hash types. Chapter 6: System Initializer (sysinit) Running System Initializer Hash Type New Universal Pre Universal Kanji International Description The recommended hash for most Teradata Database installations. New Universal hash can be used with standard (non-japanese) language support or with Japanese language support. This corresponds to the HashFuncDBC setting of 6 in the DBS Control utility. This hash type is supported for compatibility with earlier releases of Teradata Database. This hash type can be used in either Standard or Japanese Language Support Mode. This corresponds to the HashFuncDBC setting of 5 in the DBS Control utility. Hash that supports Japanese language. This hash type is supported for compatibility with earlier releases of Teradata Database. For non-character data, it produces more hash collisions than either of the Universal hashes. This corresponds to the HashFuncDBC setting of 3 in the DBS Control utility. Hash that supports international languages, excluding Japanese. This hash type is supported for compatibility with earlier releases of Teradata Database. For non-character data, it produces more hash collisions than either of the Universal hashes. This corresponds to the HashFuncDBC setting of 4 in the DBS Control utility. If you choose either New Universal or Pre Universal Hash, System Initializer displays: Enable Japanese language support (YES/NO/QUIT)? Japanese and Standard language support modes are included for compatibility with earlier releases of Teradata Database. If you are migrating an existing system to the current release, Teradata recommends that you maintain the language support mode of the prior installation. This allows limiting object names to the rules of prior releases. The following table describes the differences between Standard and Japanese language support. Standard Language Support The character set of the locale used for the TRANSLATE functions Unicode_to_Locale and Locale_to_Unicode is LATIN. Japanese Language Support The character set of the locale used for the TRANSLATE functions Unicode_to_Locale and Locale_to_Unicode is Kanji1. Support Utilities, Release

260 Chapter 6: System Initializer (sysinit) Running System Initializer Standard Language Support The compatibility encoding used for the processing of object names specified in internal hexadecimal form (XN) is LATIN. The value of the Language Support Mode column of DBCInfoTbl in the Data Dictionary is Standard. Japanese Language Support The compatibility encoding used for the processing of object names specified in internal hexadecimal form (XN) is Kanji1. The value of the Language Support Mode column of DBCInfoTbl in the Data Dictionary is Japanese. For both Language Support Modes: Object names are processed and stored in the Data Dictionary as UNICODE. Object name validation rules used by Teradata Database are determined by the NameValidationRule setting in DBS Control. For more information see the DBS Control utility chapter of Utilities. Reading the Current Configuration Map and Destroying Old Data Sysinit next attempts to read the current configuration map. IF a valid configuration map... THEN sysinit... does not exist or is corrupted initializes the configuration to a full configuration based on vprocconfig GDO and displays: Config Map will be initialized. does exist retains the current configuration and displays: The current configuration map has been read. Sysinit displays a message about destroying user and dictionary data and prompts you to confirm this action: SYSINIT is about to destroy all user and dictionary data. Are you sure that you want to do this? (YES/NO/QUIT)? IF you answer... YES NO or QUIT THEN sysinit displays Deleting all tables... before terminating: SYSINIT terminated without updating disks. 260 Support Utilities, Release 16.00

261 Verifying Vprocs Chapter 6: System Initializer (sysinit) Running System Initializer Sysinit checks to see if AMP vproc 0 is defined or if the hardware required to run AMP vproc 0 is available. If not, one of the following messages is displayed before sysinit terminates: or AMP vproc 0 is not operational. SYSINIT terminated without updating disks. AMP vproc 0 is not defined. SYSINIT terminated without updating disks. Sysinit will not initialize any AMPs that are not operational, that is, their VprocState is NONODE. The following message is displayed if any are found: The following AMPs are defined in the current configuration map and the physical hardware required to run them is not available. Hence, they will not be initialized by SYSINIT. nnnn nnnn nnnn... where nnnn nnnn nnnn... represent the vproc numbers of the AMPs in question. For a previously initialized Teradata Database system where sysinit was able to read the current configuration, sysinit first displays the current number of AMPs and PEs, as follows: Accessing the current configuration map... The current configuration map includes: 4 PE(s) and 8 AMP(s) Specifying the Configuration Map to Use If more than one AMP and one PE exist, sysinit displays four options for the current and new configuration maps: The current configuration map includes: 4 PE(s) and 8 AMP(s) Enter a value for the current/new configuration maps: 1 for current: 4 PE(s) and 8 AMP(s) new: 4 PE(s) and 8 AMP(s) 2 for current: 4 PE(s) and 1 AMP(s) new: 4 PE(s) and 1 AMP(s) 3 for current: 0 PE(s) and 1 AMP(s) new: 4 PE(s) and 8 AMP(s) 4 for current: 0 PE(s) and 1 AMP(s) new: 0 PE(s) and 1 AMP(s) Type a value for the current/new configuration maps (for most situations, choose option 1) Support Utilities, Release

262 Chapter 6: System Initializer (sysinit) Configuration and Reconfiguration Utilities System Initializer displays the following messages to confirm your selection: Updating the new configuration to: 4 PE(s) and 8 AMP(s) Updating the current configuration to: 4 PE(s) and 8 AMP(s) Updating current hash maps... Concluding Sysinit Before terminating, sysinit prompts you for the following action: Would you like to continue with startup (YES/NO)? IF you answer... THEN... YES IF you answer... NO sysinit restarts the PDE and DBS components of the system. THEN the following message appears... Don't forget to set the "Start DBS" flag to "On" using CTL utility program prior to restarting the system. Removing old UDF, Java libraries and UIFs if any... SYSINIT complete. After displaying the message, sysinit quits. Configuration and Reconfiguration Utilities If Teradata Database is initialized with fewer than the full complement of AMP and PE vprocs (something other than option 1 was selected for step ), run the Configuration and Reconfiguration utilities to define the AMPs and PEs that will operate together as a Teradata Database system. The Configuration utility adds, changes, modifies, or displays vprocs or hosts in the new configuration map. The Reconfiguration utility then redefines the Teradata Database system configuration according to the new map. For more information on the Configuration utility, see Configuration Utility (config). For more information on the Reconfiguration utility, see Reconfiguration Utility (reconfig). 262 Support Utilities, Release 16.00

263 Sysinit Examples Chapter 6: System Initializer (sysinit) Sysinit Examples Example: Sysinit cannot run while the DBS is running This example shows that after the user started sysinit, a message stating that the Teradata Database system is running appears, and sysinit terminates. SYSINIT Utility ********************************************************************* ********************************************************************* ***** *** ***** W A R N I N G *** ***** *** ***** This program will destroy all user and dictionary data on *** ***** the system. *** ***** *** ********************************************************************* ********************************************************************* SYSINIT Master AMP Vproc is 0 at 10:40:47 on 11/04/14. The DBS is currently running!!! SYSINIT cannot execute while the DBS is running. Would you like to restart the system without the DBS (YES/NO)? no Set the "Start DBS" flag to "Off" using the CTL utility program and restart the system prior to running SYSINIT again. SYSINIT terminated without updating disks. Example: Sysinit complete session The following example shows a complete sysinit session, including prompts, user responses, and messages. The example assumes a previously configured Teradata Database with an existing DBS Control Record and an existing current configuration map. New Universal Hash is accepted. SYSINIT Utility *********************************************************************** *********************************************************************** ***** ***** ***** W A R N I N G ***** ***** ***** ***** This program will destroy all user and dictionary data on ***** ***** the system. ***** ***** ***** *********************************************************************** *********************************************************************** SYSINIT Master AMP Vproc is 0 at 16:32:15 on 11/10/22. The existing DBSCONTROL GDO will be used. System is being initialized to 20-bit hash buckets. The HashFuncDBC value in the DBSCONTROL GDO is Universal New Universal hash (6) is recommended for all new installations Do you wish to use HashFuncDBC (6) (YES/NO/QUIT)? yes Enable Japanese language support (YES/NO/QUIT)? no SYSINIT is about to destroy all user and dictionary data!!! Are you sure that you want to do this (YES/NO/QUIT)? yes Deleting all tables... Accessing the current configuration map... The current configuration map includes: 1 PE(s) and 4 AMP(s) Support Utilities, Release

264 Chapter 6: System Initializer (sysinit) Sysinit Examples Enter a value for the current/new configuration maps: 1 for current: 1 PE(s) and 4 AMP(s) new: 1 PE(s) and 4 AMP(s) 2 for current: 1 PE(s) and 1 AMP(s) new: 1 PE(s) and 1 AMP(s) 3 for current: 0 PE(s) and 1 AMP(s) new: 1 PE(s) and 4 AMP(s) 4 for current: 0 PE(s) and 1 AMP(s) new: 0 PE(s) and 1 AMP(s) 1 Updating the new configuration to: 1 PE(s) and 4 AMP(s) Updating the current configuration to: 1 PE(s) and 4 AMP(s) Updating current hash maps... Would you like to continue with start up (YES/NO)? yes SYSINIT complete. 264 Support Utilities, Release 16.00

265 APPENDIX A How to Read Syntax Diagrams Overview This appendix describes the conventions that apply to reading the syntax diagrams used in this book. Syntax Diagram Conventions Notation Conventions Item Definition and Comments Letter An uppercase or lowercase alphabetic character ranging from A through Z. Number A digit ranging from 0 through 9. Do not use commas when typing a number with more than 3 digits. Word Keywords and variables. UPPERCASE LETTERS represent a keyword. Syntax diagrams show all keywords in uppercase, unless operating system restrictions require them to be in lowercase. lowercase letters represent a keyword that you must type in lowercase, such as a Linux command. Mixed Case letters represent exceptions to uppercase and lowercase rules. The exceptions are noted in the syntax explanation. lowercase italic letters represent a variable such as a column or table name. Substitute the variable with a proper value. lowercase bold letters represent an excerpt from the diagram. The excerpt is defined immediately following the diagram that contains it. UNDERLINED LETTERS represent the default value. This applies to both uppercase and lowercase words. Spaces Punctuation Use one space between items such as keywords or variables. Type all punctuation exactly as it appears in the diagram. Paths The main path along the syntax diagram begins at the left with a keyword, and proceeds, left to right, to the vertical bar, which marks the end of the diagram. Paths that do not have an arrow or a vertical bar only show portions of the syntax. The only part of a path that reads from right to left is a loop. Support Utilities, Release

266 Appendix A: How to Read Syntax Diagrams Syntax Diagram Conventions Continuation Links Paths that are too long for one line use continuation links. Continuation links are circled letters indicating the beginning and end of a link: A A When you see a circled letter in a syntax diagram, go to the corresponding circled letter and continue reading. Required Entries Required entries appear on the main path: SHOW If you can choose from more than one entry, the choices appear vertically, in a stack. The first entry appears on the main path: SHOW CONTROLS VERSIONS Optional Entries You may choose to include or disregard optional entries. Optional entries appear below the main path: SHOW CONTROLS If you can optionally choose from more than one entry, all the choices appear below the main path: READ SHARE ACCESS Some commands and statements treat one of the optional choices as a default value. This value is UNDERLINED. It is presumed to be selected if you type the command or statement without specifying one of the options. Strings String literals appear in apostrophes: 'msgtext ' Abbreviations If a keyword or a reserved word has a valid abbreviation, the unabbreviated form always appears on the main path. The shortest valid abbreviation appears beneath. SHOW CONTROLS CONTROL In the above syntax, the following formats are valid: 266 Support Utilities, Release 16.00

267 SHOW CONTROLS SHOW CONTROL Appendix A: How to Read Syntax Diagrams Syntax Diagram Conventions Loops A loop is an entry or a group of entries that you can repeat one or more times. Syntax diagrams show loops as a return path above the main path, over the item or items that you can repeat:, 3, 4 ( cname ) Read loops from right to left. The following conventions apply to loops: Item Description Example maximum number of entries allowed minimum number of entries allowed separator character required between entries delimiter character required around entries The number appears in a circle on the return path. The number appears in a square on the return path. The character appears on the return path. If the diagram does not show a separator character, use one blank space. The beginning and end characters appear outside the return path. Generally, a space is not needed between delimiter characters and entries. In the example, you may type cname a maximum of four times. In the example, you must type at least three groups of column names. In the example, the separator character is a comma. In the example, the delimiter characters are the left and right parentheses. Excerpts Sometimes a piece of a syntax phrase is too large to fit into the diagram. Such a phrase is indicated by a break in the path, marked by ( ) terminators on each side of the break. The name for the excerpted piece appears between the terminators in boldface type. The boldface excerpt name and the excerpted phrase appears immediately after the main diagram. The excerpted phrase starts and ends with a plain horizontal line: Support Utilities, Release

268 Appendix A: How to Read Syntax Diagrams Syntax Diagram Conventions LOCKING excerpt HAVING con where_cond excerpt, cname, col_pos Multiple Legitimate Phrases In a syntax diagram, it is possible for any number of phrases to be legitimate: DATABASE TABLE VIEW dbname tname vname In this example, any of the following phrases are legitimate: dbname DATABASE dbname tname TABLE tname vname VIEW vname 268 Support Utilities, Release 16.00

269 Appendix A: How to Read Syntax Diagrams Syntax Diagram Conventions Sample Syntax Diagram, CREATE VIEW viewname AS A CV cname LOCKING LOCK A dbname ACCESS B DATABASE FOR SHARE MODE tname IN READ TABLE WRITE vname EXCLUSIVE VIEW EXCL,, B SEL expr FROM tname qual_cond C.aname C HAVING cond ; qual_cond WHERE cond, GROUP BY cname, col_pos Support Utilities, Release

270 Appendix A: How to Read Syntax Diagrams Syntax Diagram Conventions 270 Support Utilities, Release 16.00

271 APPENDIX B Starting the Utilities Overview Teradata Database offers several interfaces from which the utilities may be started and run. Interface Database Window (DBW) Description DBW is a graphical tool that connects to the Teradata Database console subsystem (CNS). CNS provides console services to utility programs that operate on the database level of Teradata Database. Console utilities should be started from DBW. Note: Operators must be members of the tdtrusted user group to run console utilities, or must be logged in as root. Non-tdtrused users may be explicitly granted access to the console using the CNS GRANT command. For more information on the GRANT command, see the Database Window chapter of Utilities. For low bandwith connections, command-line interfaces to CNS are available, such as cnsterm and cnstool. Online documentation is available for cnsterm and cnstool in the form of Linux man pages. A subset of the console utilities can be run from the Remote Console portlet of Teradata Viewpoint. For more information, see Teradata Viewpoint User Guide. Linux command line Utilities that run directly from the command line are primarily those that operate on the PDE level of Teradata Database. Chapter names in the Utilities and Support Utilities manuals reflect the utility common name followed by the name of the executable utility program enclosed in parentheses. For example, Control GDO Editor (ctl). Use the executable program name to start the utility from the command line or Database Window. Note: Not all utilities support all available user interfaces. For a listing of supported user interfaces for a utility, see the documentation for that utility. When started, some utilities present their own interactive command-line or graphical user interfaces. These utilities allow browsing and entering information, and continue running until they are explicitly stopped. Many utilities that present their own command environment are stopped by entering the QUIT command. Some utilities that run from DBW can be stopped by issuing the stop window_number command from the DBW Supervisor window, where window_number is the numeric identifier of the DBW application window in which the utility is running. For more information on DBW, see the "Database Window (xdbw)" chapter in Utilities. Support Utilities, Release

272 Appendix B: Starting the Utilities Starting a Utility from Database Window Starting a Utility from Database Window Database Window (DBW) is an X client program that requires an X server to be running on the local machine. DBW supports standard X Windows display forwarding. To ensure that the graphical user interface displays properly, you can use the standard -display option to specify the host name or IP address of the local machine. To start a utility from Database Window: 1. If not already done, set up the Teradata Database environment by typing: tdatcmd at the Linux command line. 2. Open DBW from the Linux command line by typing: xdbw display displayspec & where displayspec is the name or IP address of the local machine, followed by a colon and the server number, typically 0 or 0.0. For example: xdbw -display myworkstation.mycompany.com:0.0 & or xdbw -display :0.0 & The DBWmain window opens. 3. Click the Supvr button to open the Supervisor (supv) window. 272 Support Utilities, Release 16.00

273 Appendix B: Starting the Utilities Starting a Utility from Database Window 4. Under Enter a command type: start utilityname [ options ] where utilityname is the name of the utility, and options can include any of the available command-line options and arguments of that utility. utilityname is not case-sensitive. The following message appears: Started ' utilityname ' in window x where x is the number of one of the four available application windows DBW provides. Note: Utilities started from DBW run in the control vproc, usually vproc 0, unless the VPROC option of the START command is used to specify a different vproc. For more information, see the Database Window utility documentation in Utilities. Each utility started runs in one of the four application windows. The title bar of the application window and the corresponding button in the DBW main window change to reflect the name of the running utility. When the utility stops running, the application window and main window button revert to the default text title (that is, Appl1, Appl2, and so forth). Support Utilities, Release