What would you do if you knew?

Transcription

1 What would you do if you knew? Teradata Aster Execution Engine Aster Instance Installation Guide for Aster-on-Hadoop Only Release B K July 2017

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, 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, Amazon Elastic Compute Cloud, Amazon EC2, Amazon Simple Storage Service, Amazon S3, AWS CloudFormation, and AWS Marketplace 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 and CDH 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 either registered trademarks or trademarks of EMC Corporation in the United States and/or other countries. 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 of Quest Software, Inc. 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. 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. Veritas, the Veritas Logo and NetBackup are trademarks or registered trademarks of Veritas Technologies LLC or its affiliates in the U.S. 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 noninfringement. 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 CHAPTER 1 Introduction to Aster Instance Installation Early Planning For an Aster Instance Installation Use this information as part of the initial planning for installing an Aster instance on a Hadoop cluster. Network Preparation IP addresses need not be contiguous, but do need to be on the same network. Although the Aster instance will not manage allocation of IP addresses, it is still best to keep all addresses as part of the same subnet. Also, it is best to limit the number of network devices between nodes and to keep the interconnect as balanced as possible. For example, placing all machines on the same VLAN but having a bridge or router between machines may cause bandwidth skew among nodes, and you may suffer undesirable performance as a result. Ensure to make a large enough subnet because although it is possible to change the network settings (including IP addresses) of your nodes after installation, it is not recommended. Aster Instance Installation Overview Installing an Aster instance on a Hadoop cluster uses Teradata's Parallel Upgrade Tool (PUT) to install the Aster Execution Engine Base Module software package. The instructions in this guide require that before installing the Aster Execution Engine Base Module software package that you have already installed and will manage a supported OS and Hadoop distribution on the Hadoop cluster. See Supported OS and Hadoop Distribution Combinations or the Teradata Aster Execution Engine Release Notes for the supported OS and Hadoop distribution combinations. Basic Aster Instance Concepts Node Types There are two types of nodes in an Aster instance: queen nodes and vworker nodes. For more information about the Aster instance architecture, refer to the Teradata Aster Execution Engine Aster Instance User Guide. Queen Nodes Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

4 Chapter 1: Introduction to Aster Instance Installation Basic Aster Instance Concepts The queen is the central node in the system, and is responsible for managing the execution engine cluster configuration and coordinating queries. It is installed on a Hadoop cluster edge node. Responsibilities of the queen node include: Accept, optimize and plan SQL or SQL/MR statements from clients Distribute work across the vworkers in the cluster Aggregate results from workers Send results back to clients Vworker Nodes Vworkers are the daemons that are distributed over participating Hadoop cluster data nodes and are responsible for computation on distributed data. A vworker is the unit of parallelism in the execution engine working on a slice of the input data independently and coordinated by the queen. As your computational needs grow, you can add more vworkers to the execution engine cluster to ensure query execution times scale linearly with data volume and computational needs. Each participating Hadoop data node hosts one or more (typically four-twelve) vworkers. The vworkers receive phases of the execution plan that are relevant to a slice of the distributed data. Data that reside on HDFS is read in parallel by the vworkers via the Aster-Hadoop connector. Each vworker is also responsible for the storage of a slice of Aster transient data in its local temporary storage space. Vworkers have the capability to perform inter-vworker communication in case the plan requires shuffling of data between worker machines. Once the computation is complete, the results are sent back to the queen. The number of vworkers in the cluster determines the computational capacity of the execution engine. The engine capacity can scale to match required need by adding more vworkers. Adding vworkers can be achieved by increasing the number of vworkers per data node (up to the maximum number of vworkers per node recommended), or by increasing the number of participating Hadoop data nodes. The latter increase may require adding new data nodes to the Hadoop clusters if all data nodes are already participating. Node States The node state or status indicates the operational health or condition of a physical node in an Aster instance. In the AMC Nodes > Node Overview tab, each node in the node list has a color indicating its status. The possible node statuses are: New When a node is first added to the Aster instance, or registered, it is considered to be a New node. At this point, the Aster instance is aware of the node s existence, but the node has not yet contacted the queen in order to be prepared, or loaded with the Aster instance software. Nodes are also shown as New immediately following a restart of the Aster instance, before their state can be determined. Preparing After the node contacts the queen to be prepared, its status changes to Preparing. While in this status, it is loading the Aster instance software and preparing itself to become a participant in Aster Database. Prepared Once the node completes preparation, its status becomes Prepared. At this point, the node is ready to be incorporated into the Aster instance so that it can host virtual workers (vworkers). 4 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

5 Chapter 1: Introduction to Aster Instance Installation Aster Instance Tools Upgrading The Aster software on the node is being upgraded. Active Active is the acceptable state for nodes in a running cluster. Active nodes are nodes that are available immediately to process queries in the Aster instance. Passive The Passive state for nodes in a running cluster is not an expected state. If a node reflects a Passive state, contact your Teradata Customer Support Representative. Suspect Suspect nodes are nodes that have exhibited unusual behavior and are participating in the Aster instance in a limited capacity while being investigated for potential failures by the queen. Failed Failed nodes are nodes that are no longer participating in the Aster instance. Aster Instance Tools These tools are used in the installation, configuration, and management of the Aster instance: Aster Management Console (AMC) Aster Instance Command Line Interface (ncli) Teradata's Parallel Upgrade Tool (PUT) Aster instance command line management tool (asteryarn) AMC The Aster Management Console (AMC) is a browser-based management application that runs on the queen. You will be able to use the AMC after the queen is installed in the Aster instance. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

6 Chapter 1: Introduction to Aster Instance Installation Aster Instance Tools Logging in to the AMC Information on using the AMC is documented in the Teradata Aster Execution Engine Aster Instance User Guide. This guide is available on the Teradata Information Products website: Open the Aster Management Console (AMC) in a browser window. To do this, navigate to <public ip address of the queen>. 2. In the login window, type the username db_superuser and the password db_superuser. ncli The Aster instance command line interface (ncli) is a command line management tool for the Aster instance. Many of the functions available in ncli can also be performed using the AMC. ncli provides the ability to manage the instance when: A browser is not available, The queen is not running, or Network setup problems prevent you from accessing the AMC, but you can access a console window on the queen. Information on using ncli is documented in the Teradata Aster Execution Engine Aster Instance User Guide. This guide is available on the Teradata Information Products website: PUT Teradata's Parallel Upgrade Tool (PUT) is the graphical user interface (GUI) used to install the Aster instance. Information on using Teradata's Parallel Upgrade Tool (PUT) is documented in the Parallel Upgrade Tool (PUT) Reference. This guide is available on the Teradata Information Products website: Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

7 asteryarn asteryarn is a command line management tool that is used to manage an Aster instance. Some of the functions available in asteryarn can also be performed using ncli. asteryarn provides these cluster management capabilities: Get the status of the cluster. List all running clusters. Chapter 1: Introduction to Aster Instance Installation Resource Management with YARN Information on using asteryarn is documented in the Teradata Aster Execution Engine Aster Instance User Guide. This guide is available on the Teradata Information Products website: Resource Management with YARN For an Aster instance, Yet Another Resource Negotiator (YARN) is the resource management layer for the Apache Hadoop ecosystem that assists with the allocation of resources and the advancement of tasks. Information on configuring YARN is documented in Set YARN Configuration Value Changes, and additional information about YARN (and asteryarn) is documented in the Teradata Aster Execution Engine Aster Instance User Guide. This guide is available on the Teradata Information Products website: Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

8 Chapter 1: Introduction to Aster Instance Installation Resource Management with YARN 8 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

9 CHAPTER 2 Early Preparation For an Aster Instance Installation System Requirements Supported Hardware Supported hardware for a new installation: Teradata Hadoop Appliance 3 Teradata Hadoop Appliance 4 Teradata Hadoop Appliance 5 Teradata Hadoop Appliance 6 Third party Hadoop hardware (for example, Dell or HP) For installations using third party Hadoop hardware, Teradata recommends using Data Nodes with 512GB memory and 24 cores. Supported OS and Hadoop Distribution Combinations The supported operating system (OS) and Hadoop distribution combinations for a new installation of an Aster instance on a Hadoop cluster are shown here: Hadoop Distribution CDH 5.5.1, 5.8 and 5.9 HDP 2.3.4, 2.4.2, and Teradata Hadoop Appliance OS Version Teradata SUSE Linux Enterprise Server (SLES) 11 SP3 Teradata SUSE Linux Enterprise Server (SLES) 11 SP3 Commodity Hardware OS Version RedHat 6.6, 6.7, 6.8 and 7.3 SLES 11 SP3 RedHat 6.6, 6.7, 6.8 and 7.3 SLES 11 SP3 Hadoop Cluster Configuration Recommendations Teradata Aster recommends these Hadoop cluster configurations in order to run Aster-Analytics-on- Hadoop on a supported Hadoop distribution: Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

10 Chapter 2: Early Preparation For an Aster Instance Installation System Requirements For an Aster instance installation the requirements are: Recommended configuration: Three Master nodes, One Edge node, and Eight Data nodes Minimum configuration: Three Master nodes, One Edge node, and Three Data nodes Node size: For each vworker, Aster requests a container of 32GB memory and four CPU cores. If the number of vworkers is X and the number of nodes is Y, Aster uses max(x/y, 1)*32 GB of memory and max(x/y, 1)*4 CPU cores inside YARN on each node. Supported Hadoop Cluster Security Modes An Aster instance installation on a Hadoop cluster supports these Hadoop security modes: Simple Kerberos Aster vworker Calculation Information Use this information to help you determine the number of vworkers to configure per worker node. An Aster-YARN container runs on an Aster worker and contains Aster vworkers. For additional information on the Aster-YARN container, refer to the Teradata Aster Execution Engine Aster Instance User Guide. Memory Calculation A single Aster vworker handles 500 GB of data when configured with a minimum of 32 GB of memory. The formula for the total number of vworkers per Aster-YARN container (according to memory) is: Xm = (Max container memory limit in GB)/32 CPU Calculation These CPU intensive processes run inside the container: Postgres Teradata Aster SQL-MapReduce and Teradata Aster SQL-GR jobs The formula for the total number of vworkers per Aster-YARN container (according to CPU) is: Xc = ((Max container CPU limit) - 1)/4 The number 4 in the above equation represents the number of CPU cores that are requested for one Aster vworker on YARN. vworkers per Container Calculation Using the above calculations, the formula for the recommended number of vworkers per container, where each vworker can handle ten concurrent users, is: X = min(xc, Xm) 10 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

11 Total Number of vworkers Calculation Chapter 2: Early Preparation For an Aster Instance Installation System Requirements The output from this calculation can be used when determining the number of vworkers under Setting HDP YARN Configuration Changes, and as the NUM_PARTITIONS parameter when you configure the /home/beehive/.cluster_config file. The conclusion from the above calculations is that one vworker with 32 GB memory and four CPUs can handle 500 GB of data and ten concurrent users. Using this information, the formula to calculate the total number of vworkers is: N = y/10 TN = N * (x/500) where x is the database size in GB and y is the concurrency For example, if the user requirement is 10 TB and 40 users: N = 40/10 = 4 TN = 4 * (10 TB / 0.5 TB) = 4 * 20 = 80 total vworkers Using this information and calculations for vworker per container, you can calculate the number of YARN containers needed by Aster. Aster requests 1 container per node, so this estimation will help you determine the number of nodes. Client and Tools Compatibility Aster Client and Teradata Tools compatibility information for the Teradata Aster Execution Engine is provided in the Teradata Aster Execution Engine Release Notes. Supported Drivers and Utilities For a list of the supported Aster drivers and utilities, see the Teradata Aster Client Guide. AMC Supported Web Browsers AMC Generation Google Chrome Microsoft Internet Explorer Firefox Apple Safari Opera AMC and up 8.0, 9.0, 10.0, 11.0 (Microsoft Windows 7 and Microsoft Windows Server 2008 R2 and up) 35 and up N/A N/A Teradata recommends that you upgrade to the latest version of your browser. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

12 Chapter 2: Early Preparation For an Aster Instance Installation Cluster Network Prerequisite Support for SSLv3 is discontinued in response to the POODLE vulnerability. Because this vulnerability is inherent in the protocol, the only way to address it is to remove support for it. Aster requires that browsers support TLS1.1 or later. Cluster Network Prerequisite Gathering Network Settings 1. Collect this information for the node that will be the Aster instance queen, and have it available during the installation of the Aster instance on a supported Hadoop cluster. hostname domain name IP address (byn0) TCP Ports Used by the Aster Instance Informational only. You do not need to perform any special configuration to enable the Aster instance to function using these ports, which are set up automatically by the installer. In the event of a port conflict, see Resolve Service Port Conflicts or Resolving PORT_POSTGRES (5432) Port Conflict. If you encounter an issue with lingering ports, see Mitigating Port Lingering Issue. TCP Ports for Core Components Port Service Used on Nodes 1953 PORT_STATSSERVER Worker 1985 PORT_WORKERDAEMON Worker 2008 PORT_LOG_SERVER Worker 2115 PORT_ICE_SERVER Queen, Worker 2117 PORT_ICE_SERVER_DATA Queen, Worker 2195 PORT_MONITORING Worker 5432 PORT_POSTGRES Worker TCP Ports for Local Database Instances Port Service Used on Nodes 8000 local instance at the queen 12 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

13 Port Service Used on Nodes 8001 Model worker (when activated) and vworker instances TDput may use port 9000 during an installation operation, but any use of the port by TDput occurs when the vworkers are down PORT_WORKERBASEPORT Worker Chapter 2: Early Preparation For an Aster Instance Installation TCP Ports Used by the Aster Instance TCP Ports for Transient Inter-node Data Transfer Ports Port Service Used on Nodes CreateWorkerViaCloning LogCollector Extensibility shared Java VM TCP Ports for UI Web Servers Port Service Used on Nodes 80 Apache HTTP Port Queen 443 Apache HTTPS Port Queen 8080 Tomcat HTTP Port Queen 8443 Tomcat HTTPS Port Queen 8005 Tomcat "Shutdown" Port Queen 8009 Tomcat AJP Integration Port Queen TCP Ports for Non-Aster Services Port Service Used on Nodes 22 sshd Queen, Worker 8649 gmond Queen, Worker gmetad Queen Resolve Service Port Conflicts To resolve port conflicts on the service ports listed in the following table, System Administrators can directly change the conflicted port in /home/beehive/config/ports or by using the ncli netconfig Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

14 Chapter 2: Early Preparation For an Aster Instance Installation TCP Ports Used by the Aster Instance setport command. The cluster must be restarted in order for changes to the port number information to take effect. A conflict on the WORKERDAEMON (1985) service port can cause the system to be unresponsive after issuing the ncli system softstartup command. For information on the ncli netconfig setport command, refer to the Command Line Interface (ncli) Section in the Teradata Aster Execution Engine Aster Instance User Guide. Service Ports Service Name Default Port Number PORT_ICE_SERVER 2115 PORT_ICE_SERVER_DATA 2117 PORT_LOG_SERVER 2008 PORT_MONITORING 2195 PORT_STATSSERVER 1953 PORT_WORKERBASEPORT 9010 PORT_WORKERDAEMON 1985 PORT_POSTGRES 5432 Resolving PORT_POSTGRES (5432) Port Conflict To resolve a port conflict on the PORT_POSTGRES (5432) service port, System Administrators can directly change the conflicted port by performing this procedure: 1. Execute this command on the queen node as the root user: ncli system softshutdown 2. Change the port parameter in /home/beehive/bin/utils/init/postgresql.conf to an available port value for Postgres. For example, Execute this command on the queen node: ncli node clonefile /home/beehive/bin/utils/init/postgresql.conf 4. Update the PORT_GLOBAL parameter in /home/beehive/bin/utils/init/ InitializeWorkers.sh to the port value selected in Step 2. For example, Execute this command on the queen node: ncli node clonefile /home/beehive/bin/utils/init/initializeworkers.sh 6. Execute this command as the root user on the queen node to start the Aster instance: ncli system softstartup 14 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

15 Mitigating Port Lingering Issue Chapter 2: Early Preparation For an Aster Instance Installation Set Up Passwordless Authentication with SSH Port lingering is a known and expected behavior in Linux. The listening ports will linger for a configured amount of time even after processes have died. If you encounter this issue, Teradata recommends taking these actions to mitigate the problem: 1. Reduce the FIN timeout, which causes sockets to linger for a shorter time. For example: sysctl net.ipv4.tcp_fin_timeout=30 Reducing the timeout to near-zero risks dropped packets during normal operations. 2. Enable socket reuse, which allows the next server process to bind to a lingering socket. For example: sysctl net.ipv4.tcp_tw_reuse=1 Set Up Passwordless Authentication with SSH Set Up Passwordless Authentication for Root User To install the Aster instance, you need to create an SSH authorized key and set up passwordless root SSH. Use this Aster-recommended method to set up passwordless authentication of the root user. Depending on whether or not you already have an SSH key, proceed with one of these options: If you already have an SSH key: Copying the SSH Key to the Queen. If you need to generate a new SSH key: Generating a New SSH Key. Generating a New SSH Key 1. Log in as root on the node where you want to generate the SSH key. This could be the queen or another node. 2. Execute this command to generate a new key: # ssh-keygen -t dsa 3. Press Enter to accept the default directory for the location where the key will be saved. Enter file in which to save the key (/root/.ssh/id_dsa): Press [Enter] key 4. Press Enter when prompted for a passphrase. The installation will fail if you create a passphrase. Enter passphrase (empty for no passphrase): Press [Enter] key Enter same passphrase again: Press [Enter] key Your identification has been saved in /root/.ssh/id_dsa. Your public key has been saved in /root/.ssh/id_dsa.pub. The key fingerprint is: 04:be:15:ca:1d:0a:1e:e2:a7:e5:de:98:4f:b1:a6:01 root@auser-queen 5. Execute this command to change to the directory where your key file was created: Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

16 Chapter 2: Early Preparation For an Aster Instance Installation Set Up Passwordless Authentication with SSH # cd.ssh 6. Execute this command to copy your DSA key file to authorized_keys to make it an SSH authorized key: # cat id_dsa.pub >> authorized_keys 7. Execute one of these commands to change the permissions on the authorized key file: # chmod 600 authorized_keys # chmod og-rwx authorized_keys 8. Proceed with one of these options: If you generated the key on the queen: Testing the SSH Key on the Queen. If you generated the key on another machine: Copying the SSH Key to the Queen. Copying the SSH Key to the Queen 1. Log in as root on the node that has the SSH key you want to use. 2. Verify that the key file works: a) Issue an SSH command to localhost: # ssh localhost date b) If you are asked to confirm, type yes. c) Press Enter. The command should succeed without prompting for a password. 3. Make sure the queen is powered on and booted into its OS 4. Change the working directory to your home directory on the node with the SSH key: # cd 5. Copy the key file to the queen: a) Run this command, specifying the IP address of the queen to which you are copying the key: # scp -pr.ssh root@<queen-ip>: b) If you are asked to confirm, type yes. c) Press Enter. d) If you are prompted for a password, enter it. 6. Verify that you can SSH to the queen without a password. # ssh <queen-ip> # exit 7. Proceed to Testing the SSH Key on the Queen. Testing the SSH Key on the Queen 1. Log in as root on the queen. 2. Verify that the key file works: a) Issue an SSH command to localhost: # ssh localhost date b) If you are asked to confirm, type yes. c) Press Enter. The command should succeed without prompting for a password. 16 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

17 Enabling Passwordless Root SSH for All Nodes 1. Log in as root on the queen. 2. Verify that the key file works: a) Open an SSH session to localhost: # ssh localhost b) If you are asked to confirm, type yes. c) Press Enter. The command should succeed without prompting for a password. 3. Exit the SSH session: # exit 4. Make sure the worker nodes are powered on and booted into their OS. 5. Back on the queen, change the working directory to your home directory: # cd 6. Copy the key file to each of the Aster instance nodes: a) Run this command, specifying the IP address of the node to which you are copying the key: # scp -pr.ssh root@<node-ip>: b) If you are asked to confirm, type yes. c) Press Enter. d) If you are prompted for a password, enter it. 7. Verify that you can SSH to each worker node without a password. # ssh <queen-ip> # exit 8. Test the passwordless connections among all nodes: a) SSH from each of the nodes to all the other nodes. b) If you are asked to confirm, type yes. c) Press Enter. Chapter 2: Early Preparation For an Aster Instance Installation Set Up Passwordless Authentication with SSH 9. If you wish to do so, you may disable passwordless SSH when you no longer need it (for example, after adding a node, upgrading, and so on). Disabling Passwordless Root SSH for All Nodes After you ve completed the procedure that required passwordless root SSH, you may disable it for all nodes. 1. Log in as root on the queen. 2. Issue the command to disable passwordless SSH, providing the IP address of each worker node in the instance: # ncli sshkeys disable -n <IP1,IP2,...> When specifying multiple node IP addresses using the -n option, they must be separated by a comma only, without spaces. 3. Whenever you need to re-enable passwordless SSH (for example, when adding a node, upgrading, and so on) the procedure will direct you to do so. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

18 Chapter 2: Early Preparation For an Aster Instance Installation Set Up Passwordless Authentication with SSH Passwordless Authentication for Beehive User Setting up passwordless authentication for the Beehive user is performed after completing installation of the Aster instance. See Setting Up Passwordless Authentication for Beehive User for information on performing this procedure. 18 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

19 CHAPTER 3 Install an Aster Instance Introduction Use this information to install (stop, restart, uninstall, and change the partition count of) an Aster instance on a supported OS and Hadoop distribution combination. Perform all the steps unless the text directs you to skip a step for your particular environment. Prerequisites Supported Operating Systems These instructions require that you have already installed and will manage the operating system on all nodes in the Hadoop cluster. See Supported OS and Hadoop Distribution Combinations for the supported OS and Hadoop distribution combinations. When you perform a fresh install of an Aster instance, the installation software expects a clean machine. Pre-Installation Reminders Make sure you are ready to install the Aster instance: For an Aster instance installation, the requirements are: Install the Aster queen node on a Hadoop cluster edge node. Install Aster worker nodes on only Hadoop cluster data nodes. Review Early Planning For an Aster Instance Installation to ensure that networking for your cluster is correctly setup for the installation of an Aster instance. These installation instructions describe installing a single Aster instance on a supported operating system and Hadoop distribution combination. Currently, only a single Aster instance is supported for installation on a Hadoop cluster. If you are reinstalling an Aster instance, you must first uninstall the existing Aster instance before proceeding with the installation. See Uninstalling the Aster Instance. If Apache Knox or Apache Ranger software is running on your Hadoop cluster, contact your Account Representative before installing an Aster instance. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

20 Chapter 3: Install an Aster Instance Prerequisites Teradata has not tested this release of the Teradata Aster Execution Engine for Aster-on-Hadoop in an environment where Apache Knox or Apache Ranger software is running. If you must deploy the Teradata Aster Execution Engine in such an environment, contact your Account Representative. Review Early Preparation For an Aster Instance Installation topics, which include: System Requirements: Ensure you are installing the Aster instance on a supported OS and Hadoop distribution combination, follow Hadoop cluster configuration recommendations, and determine the number of vworkers per worker node. Cluster Network Prerequisites: Will you be using multi-nic machines? Ensure you have collected the necessary information to have it available during the installation. TCP Ports Used by the Aster Instance: This section is informational. However, you should review that ports that are required by the Aster Instance in order to avoid an unnecessary port conflict during installation. Set Up Passwordless Authentication with SSH: Ensure you have set up passwordless authentication with SSH for the root user. The standard Firefox browser no longer supports Java beginning with version 52. As a result, when you attempt to load PUT using the Firefox browser PUT displays a blank screen instead of the expected login screen. To workaround this issue, you must install the Extended Support Release version of Firefox, and then manually upgrade your Windows client machine (the machine you use to connect to PUT) to Java 1.8. Contact your Customer Support Representative for information on obtaining the Extended Support Release version of Firefox. Obtain Installation Files Work with your Account Team to order the Aster Execution Engine product bundle for the supported OS and Hadoop distribution combination on to which you will install the Aster instance and other Aster software. The Teradata At Your Service (T@YS) website, is an additional resource that is available to customers and Teradata employees. This is a secure site from which you can access Orange Books, technical alerts, and knowledge repositories; view and join forums; and download software packages and documentation. Aster Execution Engine Product Bundle The Aster Execution Engine product bundle comprises this software package and software products: Product / Software Package For installing... Product Installation Instructions Aster Execution Engine Base Module An Aster instance on a Hadoop cluster This Guide Teradata Aster Analytics Aster Analytics Teradata Aster Analytics Foundation User Guide Teradata Aster Client Aster Client Aster Client Guide Teradata Aster AppCenter Aster AppCenter Aster AppCenter User Guide 20 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

21 Chapter 3: Install an Aster Instance Prerequisites This guide only describes installing the Aster Execution Engine Base Module software package. For information on installing Teradata Aster Analytics, Teradata Aster Client or Teradata Aster AppCenter, see the product guide specified above. Aster Execution Engine Base Module Package The Aster Execution Engine Base Module package includes the Dependency Files and Required Files used to install an Aster instance on a Hadoop cluster. Dependency Files tdicu noarch.rpm TeraGSS_linux_x noarch.rpm cliv noarch.rpm tptbase noarch.rpm Required Files These files are required in order to install an Aster instance, and belong in /var/opt/teradata on your Hadoop cluster: Ensure that all required files for a specified version of an installation are located in a single directory and that a separate directory is used for each subsequent version of required installation files. aster-clients-linux rxxxx.x86_64.rpm aster-db-common rxxxx.x86_64.rpm aster-db-config-common rxxxx.x86_64.rpm aster-db-config-redhat rxxxx.x86_64.rpm aster-db-config-suse rxxxx.x86_64.rpm aster-db-coordinator rxxxx.x86_64.rpm aster-db-pkg rxxxx.x86_64.rpm aster-db-toolchain rxxxx.x86_64.rpm aster-db-worker rxxxx.x86_64.rpm TDput Installer Package Use the latest TDput installer package compatible with the supported OS and Hadoop distribution combination. To install an AX version of an Aster instance, the minimum recommended version of TDput is: For Teradata Hadoop Appliances running a supported OS and Hadoop distribution combination: TDput or later build for Hadoop. For commodity hardware running a supported OS and Hadoop distribution combination: TDput or later build for commodity. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

22 Chapter 3: Install an Aster Instance Prerequisites Set YARN Configuration Value Changes Perform the required HDP or CDH YARN configuration changes associated with the installed Hadoop distribution. These required changes must be performed before running ncli system softstartup, ncli system softrestart or asteryarn commands. asteryarn commands are described in Teradata Aster Execution Engine Aster Instance User Guide. Setting HDP YARN Configuration Changes If you have previously installed and then uninstalled an Aster instance and the Hadoop cluster configuration has not changed, you do not need to repeat these YARN configuration changes, but you may want to confirm that the required parameters are still present and in effect. 1. Access the ambari-server. 2. Determine if the YARN configuration settings meet the memory requirements for running an Aster instance on the Hadoop cluster: a) Locate the memory allocated for all YARN containers on a node. In the Ambari web interface, select Yarn > Configs > Settings. In the Memory section under the Node heading, note the memory allocated for all YARN containers on a node value as value A, which is used later in this procedure for computation purposes. In the example below, the value of A is 12GB. b) Determine the max memory required for map-reduce jobs. In the Ambari web interface, select MapReduce2 > Configs > Settings. In the MapReduce section under the MapReduce Framework heading, determine the max(mapmemory, ReduceMemory). Note the max(mapmemory, ReduceMemory) value as value B, which is used later in this procedure for computation purposes. In the example below, the value of B is 5GB. 22 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

23 Chapter 3: Install an Aster Instance Prerequisites c) Determine the number of vworkers you want to configure per Worker node. Note the number of vworkers value as value C, which is used later in this procedure for computation purposes. As an example, the value of C is 2. This number is also used later to calculate the NUM_PARTITIONS parameter when you configure the /home/beehive/.cluster_config file. d) Determine the memory to be configured per vworker. Note the memory to be configured per vworker value as value D which is used later in this procedure for computation purposes. As an example, the value of D is 2GB. Teradata Aster recommends (32GB) per vworker. If you can not allocate 32GB per vworker due to memory limitations, try a smaller value. This value is also used later as the containermemoryinmb parameter when you define values in the/ home/beehive/config/asteryarn.cfg file. e) The Aster YARN application Master runs in a container on one of the data nodes and needs memory allocated. We request 512MB of memory for the Aster YARN application Master, but what gets allocated is max(512mb, X), where X is the value of the Minimum Container Size (Memory). To locate the value for X, access Yarn > Configs > Settings in the Memory section under the Container heading. Note the max(512mb, X) value as value E, which is used later in this procedure for computation purposes. In the example below, X is 512MB, so the value of E (max(512mb, X)) is 512MB. f) Using the values for A, B, C, D, and E in the below memory requirements equation, determine if the values satisfy the equation, so that memory requirements are met for running an Aster instance on the Hadoop cluster. A >= (B + C*D + E) For example: 12000>=( *2+512) 3. If memory requirements are not met for running an Aster instance on the Hadoop cluster, perform one or more of the following actions in order to satisfy the equation: Reduce the value for Minimum Container Size (Memory) at Yarn > Configs > Settings in the Memory section under the Container heading. When determining the Minimum container size, consider if you will run other jobs in addition to Aster YARN and Hive (Map-reduce). Increase the value for Memory allocated for all YARN containers on a node, which is the A value, at Yarn > Configs > Settings in the Memory section under the Node heading. Reduce the memory to be configured per vworker, which is the D value. 4. Confirm the YARN configuration settings meet the CPU requirements for running an Aster instance on the Hadoop cluster: Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

24 Chapter 3: Install an Aster Instance Prerequisites a) Locate the number of virtual cores. In the Ambari web interface, select Yarn > Configs > Settings. In the CPU section under the Node heading, note the number of virtual cores as value A, which is used later in this procedure for computation purposes. b) The Minimum container size (virtual cores) is 1. c) Locate the maximum number of virtual cores. In the Ambari web interface, select Yarn > Configs > Settings. In the CPU section under the Container heading, note the Maximum Container Size (VCores) as value B, which is used later in this procedure for computation purposes. d) Determine the number of vworkers you want to configure per Worker node. Note the number of vworkers value as value C, which is used later in this procedure for computation purposes. This number is also used later to calculate the NUM_PARTITIONS parameter when you configure the /home/beehive/.cluster_config file. e) Determine number of virtual cores you want to configure per vworker. Note the number of virtual cores value as value D, which is used later in this procedure for computation purposes. Teradata recommends four virtual cores per vworker and that one vworker configured with four virtual cores can manage 500GB of data and ten concurrent users on a supported Hadoop cluster. If you cannot allocate four virtual cores per vworker due to limitations, allocate a fewer number of virtual cores per vworker. The number of virtual cores is also used later as the containervcore parameter when you configure the /home/beehive/config/asteryarn.cfg file. f) Using the values for A, B, C and D, in the below CPU requirements equations, verify that the values evaluate correctly in the equations. A>= C*D B >= C*D Note that (B - C*D) cores are used for the remaining applications. If Hive (Map-reduce) becomes hung, try decreasing C and D. 5. Restart all YARN services. 24 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

25 Chapter 3: Install an Aster Instance Prerequisites Setting CDH YARN Configuration Changes If you have previously installed and then uninstalled an Aster instance and the Hadoop cluster configuration has not changed, you do not need to repeat these YARN configuration changes. 1. Access the Cloudera Manager. 2. Determine if the YARN configuration settings meet the memory requirements for running an Aster instance on the Hadoop cluster: a) In the Cloudera Manager interface, select Clusters > YARN > Configuration. b) Locate the memory allocated for all YARN containers on a node by using the Search option in the Configuration pane to search for Container memory. Note the Container memory value as value A, which is used later in this procedure for computation purposes. c) Determine the max memory required for map-reduce jobs by using the Search option in the Configuration pane to search for Map Task memory and Reduce Task memory. Note the max(map Task memory, Reduce Task memory) value as value B, which is used later in this procedure for computation purposes. d) Determine the number of vworkers you want to configure per Worker node. Note the number of vworkers value as value C, which is used later in this procedure for computation purposes. This number is also used later to calculate the NUM_PARTITIONS parameter when you configure the /home/beehive/.cluster_config file. e) Determine the memory to be configured per vworker. Note the memory to be configured per vworker value as value D which is used later in this procedure for computation purposes. Teradata recommends (32GB) per vworker. If you can not allocate 32GB per vworker due to memory limitations, try a smaller value. This value is also used later as the containermemoryinmb parameter when you define values in the/ home/beehive/config/asteryarn.cfg file. f) Locate the minimum container memory allocation by using the Search option in the Configuration pane to search for Container Memory Minimum. Note the max(512mb, X) value, where X is the value of the Container Memory Minimum, as value E, which is used later in this procedure for computation purposes. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

26 Chapter 3: Install an Aster Instance Prerequisites The Aster YARN application Master runs in a container on one of the data nodes and needs memory allocated. 512MB of memory is requested for the Aster YARN application Master, but what is allocated is max(512mb, X). g) Using the values for A, B, C, D, and E in the memory requirements equation, determine if the values satisfy this equation: A >= (B + C*D + E) 3. If memory requirements are not met for running an Aster instance on the Hadoop cluster, perform one or more of the following actions in order to satisfy the equation: Reduce the value for "Container Memory Minimum". When determining the Container Memory Minimum, consider if you will run other jobs in addition to Aster YARN and Hive (Map-reduce). Increase the value for Memory allocated for all YARN containers on a node, which is the A value, at Clusters > YARN > Configuration and use the Search option to search for Container memory. Reduce the memory to be configured per vworker, which is the D value. 4. Confirm the YARN configuration settings meet the CPU requirements for running an Aster instance on the Hadoop cluster: a) In the Cloudera Manager interface, select Clusters > YARN > Configuration. b) Locate the number of virtual cores by using the Search option in the Configuration pane to search for Container Virtual CPU Cores. Note the number of virtual cores as value A, which is used later in this procedure for computation purposes. c) The Minimum container size (virtual cores) is 1. d) Locate the maximum number of virtual cores by using the Search option in the Configuration pane to search for Container Virtual CPU Cores Maximum. Note the maximum number of virtual cores as value B, which is used later in this procedure for computation purposes. e) Determine the number of vworkers you want to configure per Worker node. Note the number of vworkers value as value C, which is used later in this procedure for computation purposes. This number is also used later to calculate the NUM_PARTITIONS parameter when you configure the /home/beehive/.cluster_config file. f) Determine number of virtual cores you want to configure per vworker. Note the number of virtual cores value as value D, which is used later in this procedure for computation purposes. 26 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

27 Teradata recommends four virtual cores per vworker and that one vworker configured with four virtual cores can manage 500GB of data and ten concurrent users on a supported Hadoop cluster. If you cannot allocate four virtual cores per vworker due to limitations, allocate a fewer number of virtual cores per vworker. The number of virtual cores is also used later as the containervcore parameter when you configure the /home/beehive/config/asteryarn.cfg file. g) Using the values for A, B, C and D, in the below CPU requirements equations, verify that the values evaluate correctly in the equations. A>= C*D B >= C*D Note that (B - C*D) cores are used for the remaining applications. If Hive (Map-reduce) becomes hung, try decreasing C and D. 5. Restart YARN and any dependent services as suggested by Cloudera Manager. Chapter 3: Install an Aster Instance Install an Aster Instance Verifying Nodes Managed by YARN 1. For a secure Hadoop cluster, the beehive user must be defined on all nodes managed by YARN. 2. To verify all nodes managed by YARN, issue the yarn node --list --all command. The output of the command should list all worker nodes, and indicate 0 for all nodes when the Aster instance or any other YARN application is not up and running. For example: # yarn node --list --all 16/06/01 17:19:02 INFO impl.timelineclientimpl: Timeline service address: 16/06/01 17:19:02 INFO client.rmproxy: Connecting to ResourceManager at hrh015m1.labs.teradata.com/ :8050 Total Nodes:3 Node-Id Node-State Node-Http-Address Number-of-Running- Containers hrh015d3.labs.teradata.com:45454 RUNNING hrh015d3.labs.teradata.com: hrh015d1.labs.teradata.com:45454 RUNNING hrh015d1.labs.teradata.com: hrh015d2.labs.teradata.com:45454 RUNNING hrh015d2.labs.teradata.com: Testing the Health of the Hadoop Cluster 1. Use your Hadoop cluster management software to check the health of the cluster. 2. Confirm that no errors are reported before proceeding with the installation. Install an Aster Instance Use these instructions to install an Aster Instance on a supported OS and Hadoop distribution combination. See Supported OS and Hadoop Distribution Combinations for a list of the supported combinations. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

28 Chapter 3: Install an Aster Instance Install an Aster Instance Installation Prerequisites If Apache Knox or Apache Ranger software is running on your Hadoop cluster, contact your Account Representative before installing an Aster instance. Teradata has not tested this release of the Teradata Aster Execution Engine for Aster-on-Hadoop in an environment where Apache Knox or Apache Ranger software is running. If you must deploy the Teradata Aster Execution Engine in such an environment, contact your Account Representative. Systems running a version of the Teradata SLES operating system have these packages installed: libasound libstdc++.so.6 net-snmp (requires installation of perl-snmp) Systems running a version of the Red Hat Enterprise Linux operating system have net-snmp installed on all nodes. All nodes to be included in the Aster instance must contain the same /etc/hosts file with correct and complete information. PUT and ncli use the /etc/hosts files to group nodes and to add workers. Max open files is minimally and identical on all worker nodes. You can use the ulimit -a command on each node to check the set limit. If you change the OS configuration parameter that specifies the maximum number of open files, you do not need to restart the server or the Aster instance at this time. The Aster instance is restarted later in the procedure, so the configuration change will take effect at that time. Java version 1.7 or 1.8 is installed on the computer that will be used to connect to PUT. Java 1.8 is the preferred installed version. Installing an Aster Instance Perform these tasks to install an Aster instance using the Parallel Upgrade Tool (PUT). To install an AX version of an Aster instance, the minimum recommended version of TDput is Information on using Teradata's Parallel Upgrade Tool (PUT) is documented in the Parallel Upgrade Tool (PUT) Reference Guide. This guide is available on the Teradata Information Products website: Identify PUT master node Identify a node to be the PUT master node. The master node: Is typically an edge node in the Hadoop cluster. Must be located on the same network as all nodes in the Hadoop cluster. Hosts the software RPM files to be installed on all nodes in the Hadoop cluster. Determining PUT installation status 28 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

29 1. Determine if PUT is installed on the master node by issuing this command on the master node: rpm -qa grep TDput Chapter 3: Install an Aster Instance Install an Aster Instance a) If TDput-xxxx is returned, PUT is installed on the master node. Issue the /etc/init.d/tdput start command to start TDput services and continue to Connecting to the master node. b) If TDput-xxxx is returned, but you want to upgrade the installed version of PUT, continue to Upgrading TDput. c) If TDput-xxxx is not returned, PUT is not installed on the master node. Continue to Installing TDput. Installing TDput Perform an initial installation of TDput on all Hadoop nodes enabled with root ssh. 1. Execute this command on the master node: rpm -Uvh TDput*.rpm 2. Open a supported web browser. 3. Enter the URL for the system in the browser bar. To run PUT as a Java Web Start application, enter a URL similar to one of these URLs in the browser bar: The Java Web Start application launches PUT from the web browser and runs it in a new window separate from the web browser. The Java Web Start application cannot be launched on a Mac, but you can use the Java applet to launch PUT through Firefox on a Mac. To run PUT as a Java applet, enter a URL similar to one of these URLs in the browser bar: The Java applet launches PUT from the web browser and runs it inside the web browser. 4. When the PUT Logon screen appears: a) Enter the root ID and password for the master node to which you are connecting. b) Select Teradata Support Mode (QuickLook ID). c) Enter a valid and active QuickLook ID. 5. Use the PUT install and upgrade operation to install PUT on the remaining Hadoop nodes in the cluster. Refer to the Parallel Upgrade Tool (PUT) Reference for information about the PUT install and upgrade operation. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

30 Chapter 3: Install an Aster Instance Install an Aster Instance Upgrading TDput To upgrade an existing installation of TDput, use the PUT install and upgrade operation to upgrade all the Hadoop nodes at the same time. Refer to the Parallel Upgrade Tool (PUT) Reference for information about the PUT install and upgrade operation. Connecting to the master node 1. Once PUT is installed on the master node, you can connect to the master node using HTTP on port 8080 or HTTPS on port For example: Creating beehive user 1. Create the beehive user on the queen (edge) node: groupadd beehive useradd -m -g beehive -d /home/beehive beehive Distributing vworkers across data disks 1. Optional - To distribute Aster vworkers across multiple data disks, create a /home/beehive/config/ vworkerdisklist file on every worker node in the cluster. The file must be created before the instance is started, and should contain a comma-separated list of disks found on the individual worker node. Ensure that there are no spaces between the list of disks. For example: tdh233e1:~ # cat /home/beehive/config/vworkerdisklist /data1,/data2,/data3,/ data4,/data5,/data6,/data7,/data8 Distribution of vworker nodes across multiple data disks will take effect only if a non-empty vworkerdisklist file is present under the required directory path. The vworkerdisklist file can be added after a softshutdown and, in that case, the distribution of vworker nodes will then take effect following a softstartup or a softrestart. If the file is not present, the vworkers for that worker node will not be distributed. Creating.cluster_config file 1. Create a.cluster_config file in the /home/beehive directory on the master (queen) node in order to specify the makeup of the Aster instance. 2. Include this content in the file: PRIMARY_IP=ip addr Required The IP Address of the coordinator (queen) node for the Aster instance on the network. Installation fails in any other case. NUM_WORKERS=integer Required The number of worker nodes in the Aster instance. NOTE: By default, if the number of data nodes listed in the output of the yarn node --list --all command is equal to the NUM_WORKERS argument, asteryarn attempts to bring up the workers on all nodes listed in the output of the yarn node --list --all command. In addition, there is also a dependency on the NUM_PARTITIONS argument as well. Aster attempts 30 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

31 to allocate the minimum number of partitions on each node, so if the number of partitions is less than the number of workers, some nodes will not be used. NUM_PARTITIONS=integer Required The total number of active partitions (vworkers) in the Aster instance, which should be an integral multiple of the number of worker nodes. If it is not an integral multiple it will be rounded down to the nearest multiple. DB_CHECKSUMS=on Optional Indicates the database is checksummed. The values are off or on. The default value is on. TARGET_RF=1 Optional Replication factor. The value is 1. DB_STORAGE=hdfs Required Persistent storage is stored in HDFS. The value is hdfs. HDFS_NAMENODE=value Required This parameter is required if running on HDFS. If the High Availability (HA) feature is NOT enabled on HDFS, use the IP address of the HDFS namenode. If the HA feature is enabled on HDFS, use the contents of the dfs.nameservices <value> tag as the HDFS_NAMENODE value instead of the HDFS namenode IP address. For Example: WORKER_FQDN_LIST=string comma separated list Optional Chapter 3: Install an Aster Instance Install an Aster Instance <property> <name>dfs.nameservices</name> <value><your HDFS service name></value> </property> dfs.nameservices is located in the hdfs-site.xml file on any HDFS node. The hdfs-site.xml file is typically located in the /etc/hadoop/conf directory. If you don't want to deploy Aster workers on some of the nodes listed in the output of the yarn node --list --all command, use this parameter to specify on what nodes the Aster workers are deployed. Valid values include any combination of FQDNs, Hostnames or an '*'. For example: WORKER_FQDN_LIST=hrh012d1,* WORKER_FQDN_LIST=hrh012d1, hrh012d2, hrh012d3 WORKER_FQDN_LIST=hrh012d1.labs.teradata.c om, hrh012d2.labs.teradata.com, hrh012d3.labs.teradata.com,* WORKER_FQDN_LIST configuration parameter validation considerations: The number of entries in the comma separated list must be less than or equal to the NUM_WORKERS argument. If the number of entries are less than the NUM_WORKERS argument, the comma separated list must have at least one '*'. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

32 Chapter 3: Install an Aster Instance Install an Aster Instance When invalid hostnames are used, the Application master will fail at the cluster startup time. If the Application master fails due to an invalid hostname, you can change the comma separated list by issuing the ConfigureNCluster command with a corrected comma separated list on the queen. For example: /home/beehive/bin/lib/configure/ ConfigureNCluster.py -- allowed_host_list=<correct comma separated list for workers> OR * NTP_SERVER=value Optional This is an optional parameter that, when specified, identifies the contents of the <value> tag as the NTP server. For example: NTP_SERVER= SQLMR_JAVA_MAX_MEMORY_IN _GB=value Optional This is optional parameter that represents the amount of memory to be used for JVM that would run on all worker to run java-based Teradata Aster SQL- MapReduce or Teradata Aster SQL-GR functions. When an Aster instance is activating, it will distribute this information to all Aster workers. A lower value for this parameter helps increase concurrency in the YARN environment resulting in more queries that can be run at the same time. An Aster instance uses this formula for calculating concurrency:asteryarnconcurrency = (YarnContainerMemoryInMB)/ (sqlmrjavamaxmemoryinmb)asterconcurrency = min(asteryarnconcurrency, asterinitialconcurrency) The asterconcurrency result becomes the Aster concurrency following execution of the ncli system softstartup or the ncli system softrestart commands. Any attempt to set more concurrency than what is automatically set may result in the shutdown of the Aster instance, in the AX version of the Aster Execution Engine software, when the Aster yarncontainer uses more resource than it is allocated. Example.cluster_config file contents: PRIMARY_IP= TARGET_RF=1 DB_CHECKSUMS=on DB_STORAGE=hdfs HDFS_NAMENODE= NUM_WORKERS=3 32 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

33 NUM_PARTITIONS=6 WORKER_FQDN_LIST=hrh012d1,* NTP_SERVER= SQLMR_JAVA_MAX_MEMORY_IN_GB=16 Chapter 3: Install an Aster Instance Install an Aster Instance Verifying /sbin/ifconfig permissions 1. For Hadoop appliances, verify permissions for /sbin/ifconfig by confirming that /sbin/ifconfig has a minimum of 0555 permissions on all the nodes. 2. If there is not a minimum of 0555 permissions on all the nodes, run this command on all YARN nodes to set the permissions: chmod ugo+rx /sbin/ifconfig Create /aster in HDFS If there isn t an /aster directory in HDFS, create the directory. For Hadoop clusters that are not Kerberos-enabled, see Creating /aster on clusters that are not Kerberosenabled. For Hadoop clusters that are Kerberos-enabled, see Creating /aster on clusters that are Kerbros-enabled. Creating /aster on clusters that are not Kerberos-enabled 1. Execute these commands from any node in the cluster to create /aster in HDFS: su - hdfs -c "hdfs dfs -ls /" su - hdfs -c "hdfs dfs -mkdir /aster" su - hdfs -c "hdfs dfs -chown -R beehive:beehive /aster" su - hdfs -c "hdfs dfs -ls /" Creating /aster on clusters that are Kerbros-enabled Prerequisite Ensure Kerberos is correctly configured on the Hadoop cluster. For CDH 5.5.1, and on Teradata SLES 11 SP3 clusters, work with your account team to obtain these Kerberos setup instructions, and confirm that the instructions are implemented: +Setup For HDP 2.3.4, 2.4.2, and on Teradata SLES 11 SP3 clusters, work with your account team to obtain these Kerberos setup instructions, and confirm that the instructions are implemented: For HDP 2.3.4, 2.4.2, and on Redhat clusters, confirm these instructions in the Ambari Guide are implemented: launching_the_kerberos_wizard_automated_setup.html Kerberos clients must be installed on all nodes in the cluster. If the Kerberos clients (kinit and kdestroy) are not present, the installation will fail. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

34 Chapter 3: Install an Aster Instance Install an Aster Instance For TD SLES, the clients are installed by implementing the above instructions for TD SLES 11 SP3. For RHEL, you must install the client packages on all nodes while configuring Kerberos. 1. Execute steps 2 through 5 on the Key Distribution Center (KDC) server node. To locate the KDC server hostname, look in /etc/krb5.conf on the edge node for the kdc = <kdc_server_hostname> parameter. 2. Log into the KDC server node. 3. Determine if the HDFS principal is present in KDC: kadmin.local # listprincs a) If present, the output of the listprincs command will show an HDFS principal similar to hdfs@<domain-name>. For example: hdfs@cdh220.hadoop.teradata.com b) If the HDFS principal does not exist, use the kadmin console to create the HDFS principal. If the - randkey option is not specified, you will be prompted for a password. For example: kadmin.local: addprinc -randkey hdfs@<domain-name> WARNING: no policy specified for hdfs@<domain-name>; defaulting to no policy Principal "hdfs@<domain-name>" created. 4. Using the kadmin console, create the HDFS principal for the beehive user. For example: Replace the domain name CDH220.HADOOP.TERADATA.COM with the domain name identified in the previous step. kadmin.local: addprinc -randkey beehive@cdh220.hadoop.teradata.com WARNING: no policy specified for beehive@cdh220.hadoop.teradata.com; defaulting to no policy Principal "beehive@cdh220.hadoop.teradata.com" created. 5. Using the kadmin console, extract the hdfs.keytab and the beehive.keytab files to your current working directory. For example: Replace the domain name CDH220.HADOOP.TERADATA.COM with the domain name used in the previous step. kadmin.local: xst -k hdfs.keytab hdfs@cdh220.hadoop.teradata.com Entry for principal hdfs@cdh220.hadoop.teradata.com with kvno 4, encryption type Triple DES cbc mode with HMAC/sha1 added to keytab WRFILE:hdfs.keytab. Entry for principal hdfs@cdh220.hadoop.teradata.com with kvno 4, encryption type ArcFour with HMAC/md5 added to keytab WRFILE:hdfs.keytab. Entry for principal hdfs@cdh220.hadoop.teradata.com with kvno 4, encryption type DES with HMAC/sha1 added to keytab WRFILE:hdfs.keytab. Entry for principal hdfs@cdh220.hadoop.teradata.com with kvno 4, encryption type DES cbc mode with RSA-MD5 added to keytab WRFILE:hdfs.keytab. kadmin.local: xst -k beehive.keytab beehive@cdh220.hadoop.teradata.com Entry for principal beehive@cdh220.hadoop.teradata.com with kvno 4, 34 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

35 encryption type Triple DES cbc mode with HMAC/sha1 added to keytab WRFILE:beehive.keytab. Entry for principal with kvno 4, encryption type ArcFour with HMAC/md5 added to keytab WRFILE:beehive.keytab. Entry for principal with kvno 4, encryption type DES with HMAC/sha1 added to keytab WRFILE:beehive.keytab. Entry for principal with kvno 4, encryption type DES cbc mode with RSA-MD5 added to keytab WRFILE:beehive.keytab. 6. Use scp to transfer the hdfs.keytab and beehive.keytab files to the edge node under the /root directory. 7. Copy the hdfs.keytab file to the HDFS user s home directory. For HDP, the home directory is / home/hdfs. For CDH, the home directory is /var/lib/hadoop-hdfs. For example: cp /root/hdfs.keytab /var/lib/hadoop-hdfs/ # /home/hdfs for HDP and /var/lib/hadoop-hdfs for CDH chown hdfs:hdfs /var/lib/hadoop-hdfs/hdfs.keytab 8. Obtain a Kerberos ticket by issuing this command: su - hdfs -c "kinit -kt hdfs.keytab hdfs" 9. Verify the Kerberos ticket by issuing this command: su - hdfs -c "klist" For example: cdh220e1:~ # su - hdfs -c "klist" Ticket cache: FILE:/tmp/krb5cc_105 Default principal: hdfs@cdh220.hadoop.teradata.com Valid starting Expires Service principal 06/10/16 00:42:15 06/11/16 00:42:15 krbtgt/ CDH220.HADOOP.TERADATA.COM@CDH220.HADOOP.TERADATA.COM renew until 06/17/16 00:42:15 Kerberos 4 ticket cache: /tmp/tkt105 klist: You have no tickets cached 10. Create /aster in HDFS by executing these commands: su - hdfs -c "hdfs dfs -ls /" su - hdfs -c "hdfs dfs -mkdir /aster" su - hdfs -c "hdfs dfs -chown -R beehive:beehive /aster" su - hdfs -c "hdfs dfs -ls /" 11. On the edge (queen node), copy the beehive.keytab file and set appropriate permissions by executing these commands: mkdir /home/beehive/.keytab chown beehive:beehive /home/beehive/.keytab cp /root/beehive.keytab /home/beehive/.keytab chown beehive:beehive /home/beehive/.keytab/beehive.keytab Chapter 3: Install an Aster Instance Install an Aster Instance Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

36 Chapter 3: Install an Aster Instance Install an Aster Instance Starting PUT Start PUT using either an HTTP or HTTPS connection and run it as either a Java applet or a Java Web Start application. The Java applet launches PUT from the web browser and runs it inside the web browser. The Java Web Start application launches PUT from the web browser and runs it in a new window separate from the web browser. The Java Web Start application cannot be launched on a Mac, but you can use the Java applet to launch PUT through Firefox on a Mac. 1. Open a supported web browser. 2. To run PUT as a Java applet, enter one of these URLs in the browser bar: To run PUT as a Java Web Start application, enter one of these URLs in the browser bar: In the PUT log in screen, login using one of these options: If a window opens asking if you want to run the TDput Loader application, click Run Enter the root ID and password for the Hadoop node to which you are connecting. If you have a valid, active Teradata QuickLook ID, select Teradata Support Mode (QuickLook ID), and enter a valid, active QuickLook ID. Installing Aster Execution Engine Base Module Package This procedure installs the Aster Execution Engine Base Module package using PUT. 36 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

37 For an Aster instance installation, the installation location requirements for queen and worker nodes are: Install the Aster queen node on a Hadoop cluster edge node. Install Aster worker nodes on only Hadoop cluster data nodes. Chapter 3: Install an Aster Instance Install an Aster Instance The subset of topics described in this section are intended to supplement the detailed information available in the corresponding topics in Install/Upgrade Software Section in the Parallel Upgrade Tool (PUT) Reference, and provide information specific to an Aster instance software installation. Detailed information on using Teradata's Parallel Upgrade Tool (PUT) is documented in theparallel Upgrade Tool (PUT) Reference. This guide is available on the Teradata Information Products website: PUT Interface Screen 1. Click Install/Upgrade Software. Configuration Mode 1. Select Typical. 2. Click Next. Find All Subnets This action can take several minutes to complete, and returns the IP address of all subnets where PUT daemons are running. 1. Select the subnet that corresponds to the cluster s subnet. 2. Click Next. Select Nodes 1. Select the IP address of the nodes in the cluster from the Available list and move them to the Selected list by clicking >>. 2. Click Next. If all nodes in the cluster do not have a shared subnet, select the subnet on which the nodes you want to install are located; multiple runs of the TDput operation will be required. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

38 Chapter 3: Install an Aster Instance Install an Aster Instance Tip: If you have previously installed and uninstalled an Aster instance on this Hadoop cluster and the configuration of the Hadoop cluster has not changed, this screen will already have the previous node selection under Selected. If the node selection is correct, click Next. System Information Warning 1. The information is advisory and no action needs to be taken. Click Next. Select Spool Area 1. Use the default location. Click Next. Media Sources 1. Browse to the source for the RPM packages downloaded in Aster Execution Engine Base Module Package. 2. Click Next. The Media Source Confirmation screen appears, displaying the selected media sources. 3. Review and, if applicable, add or remove sources: If you downloaded more than one RPM package, perform this step until you have entered the source for all downloaded RPM packages. To add media sources, select Yes, and repeat the previous steps; otherwise, select No. To remove a media source, select Remove It; otherwise, select Keep It. 4. When you are done specifying media sources, in the Media Source Confirmation screen, click Next. 5. In the Media Source(s) To Be Scanned screen, click Next. 38 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

39 Chapter 3: Install an Aster Instance Install an Aster Instance Group Nodes Assign every node in the Aster instance to one of two groups (queen or workers). For each group (queen or workers) you will loop through this task. Ensure that every node in the Aster instance is assigned to a group. The action to assign every node in the Aster instance to a group is mandatory whether or not you initially plan to actively use the node as part of the Aster instance. During an initial installation of the Aster instance, all nodes should initially be shown in one group until you assign them to different groups. If the Aster instance nodes are improperly grouped before any action is taken or if PUT will not allow groups to be assigned or combined into a single group, refer to the Parallel Upgrade Tool (PUT) Reference or Troubleshoot Installation Issues for troubleshooting information. 1. In the Group Nodes screen, select the node(s) for each group (queen or workers). Press Ctrl-click to select more than one node. Nodes categorized as HDEDGE nodes are Aster Queen nodes, and HDDATA nodes are Aster worker nodes. 2. Click Modify Group. The Modify Group screen opens. 3. In the Modify Group screen: a) Under the Choose Group heading, Select New Group. b) Under the Node Type heading, select the appropriate Node Type: ADQueen or ADWorkr. If the Node Type heading is not available: Confirm that the installed version of PUT is minimally and update to a newer version if necessary. For new installations only, refer to the Troubleshooting PUT Appendix in the Parallel Upgrade Tool (PUT) Reference. c) Click Submit Changes. You are returned to the Group Nodes screen where your changes are displayed. 4. Repeat these steps until all cluster nodes are in the correct groups. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

40 Chapter 3: Install an Aster Instance Install an Aster Instance 5. Click Next. Set Maintenance Window 1. Select the default value, No. 2. Click Next. Select Packages The Select Packages screen lists the nodes and the software package name and version that will be installed. To select the software packages if the nodes and software packages are not auto selected, see Selecting Software Packages. To accept the software package installation recommendations, click Next. The software package name and version must match the required and dependency file names and versions included in the Aster Execution Engine product bundle for the supported OS and Hadoop distribution combination on to which you will install the Aster instance and other Aster software. Selecting Software Packages In the Select Packages screen, associate the package with the group on which it should be installed. Perform this task for each group. 1. Select the appropriate package name for the node type: 40 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

41 For the Queen, select aster-db-coordinator version 7.xx.xx.xx Chapter 3: Install an Aster Instance Install an Aster Instance where 7.xx.xx.xx is the version of the file downloaded in Aster Execution Engine Base Module Package. For the worker group, select aster-db-worker version 7.xx.xx.xx where 7.xx.xx.xx is the version of the file downloaded in Aster Execution Engine Base Module Package. 2. In the Modify screen, select the group onto which the package should be installed. 3. Click Next. Ensure that you do not click the Select All option. The Install/Upgrade Software: Package Dependencies Exist screen opens. 4. In the Install/Upgrade Software: Package Dependencies Exist screen: a) Scroll through the list and verify that the package status for every package under the Status column indicates one of these statuses: If, instead of the Install/Upgrade Software: Package Dependencies Exist screen, the Package Dependencies Not Met screen opens, ensure that the necessary software packages are available and then restart the procedure. Installed Available but not selected for upgrade Available but not selected for upgrade - version specific b) Click Yes. You are returned to the Select Packages screen. 5. In the Select Packages screen, click Next. Confirm Package Selections 1. Verify that the packages to be installed and the number of nodes that each package will be installed on are correct. 2. Click Next. The software package name and version must match the required and dependency file names and versions included in the Aster Execution Engine product bundle for the supported OS and Hadoop distribution combination on to which you will install the Aster instance and other Aster software. Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

42 Chapter 3: Install an Aster Instance Install an Aster Instance Select Boot Environment (BE) for Upgrade: Systems running a version of the Teradata SLES operating system: a) Select Non-VM&F upgrade - use Primary BE (BE1). b) Click Next. Systems running a version of the Red Hat Enterprise Linux operating system: no user action is required. The procedure bypasses the Select Boot Environment (BE) for Upgrade screen. Distribute Packages No user interaction required. The time for this step to complete is dependent on the size of the packages and the number of nodes to which those packages must be distributed. Begin Installing Packages Now? 1. Click Next. Begin Install? 1. Click Next. Package Install 1. Click Next. This step will take time to complete. The Overall Package Install Results screen opens. 2. In the Overall Package Install Results screen: a) Verify that all packages were installed successfully. 42 Aster Instance Installation Guide for Aster-on-Hadoop Only, Release

43 Chapter 3: Install an Aster Instance Install an Aster Instance b) Click Next. Pause for Change Control Activities An intentional stopping point is provided at this point in the operation in order to record Change Control activities. 1. Click Next. Finished! Installation is now complete; temporary files are cleaned up. 1. Click Finish to complete the operation prior to performing another PUT operation. 2. Click Next to return to the PUT Interface screen. Bringing Up the Aster Instance To bring up the Aster instance following installation of the rpm files, perform these tasks: Manually "source" the asterenv.sh file Manually "source" the asterenv.sh file to ensure that your current shell will have the correct environment variables in order to complete the installation. From ncli, see Sourcing the asterenv.sh file from ncli. From AMC, see Sourcing the asterenv.sh file from AMC (Optional). Sourcing the asterenv.sh file from ncli 1. Locate and open either of these files: ~/.profile or ~/.bashrc 2. In the opened file, copy and paste this text as a single line into the file: Aster Instance Installation Guide for Aster-on-Hadoop Only, Release