Db2 Big SQL cluster administration utility

Authorization

To run this utility, you must be logged on as root, a non-root user with sudo privileges, or the Db2 Big SQL service user. In this release, logging on as a user who is not the default Db2 Big SQL user (that is, bigsql) is not supported.

Prerequisites

• The bigsql-admin utility must be run on the Db2 Big SQL head node. You can find the utility in /usr/ibmpacks/IBM-Big_SQL/7.1.0.0/bigsql-cli.

Syntax

super-user-options

bigsql-user-options

Command parameters

-enableHA "standby-host"

Specifies that a second head node is to be added and that high availability disaster recovery (HADR) is to be enabled. If you are enabling HA on the same node where the Db2 Big SQL console is installed, you must first stop the console service. You can combine this option with the -failoverController Enable and -startFC options.

-disableHA

Specifies that the second head node is to be removed and that HADR is to be disabled.

-failoverController [ Enable | Disable | Update ]

Specifies whether failover automation for HA is to be enabled or disabled. Specify Update after changes to the ZooKeeper configuration. You can combine the -failoverController Enable option with the -enableHA and -startFC options.

-addWorker "host", ...

Specifies that one or more worker nodes are to be added to the Db2 Big SQL cluster. Disable the failover controller before using this option (-failoverController Disable).

-removeWorker "host", ...

Specifies that one or more worker nodes are to be removed from the Db2 Big SQL cluster. Disable the failover controller before using this option (-failoverController Disable).

-removeDeadNode "host"

Specifies that a dead node is to be removed from your cluster.

-updateNonBigsqlMasters

Specifies that the bigsql user is to be created on non-Big SQL masters and that required Jar files are to be copied.

-register

Specifies that Db2 Big SQL is to be registered with Cloudera Manager.

-unregister

Specifies that Db2 Big SQL is to be removed from Cloudera Manager. Use this option before upgrading Cloudera Manager. Then, use the -register option to register Db2 Big SQL with the upgraded Cloudera Manager application.

-updateTopology

Specifies that the Db2 Big SQL topology is to be updated in Cloudera Manager.

-logs

Specifies that Db2 diagnostic (db2diag) log files are to be collected. This option invokes the bigsql-support.sh script. For more information, see Db2 Big SQL diagnostic log collection.

-triggerFailover

Specifies that the standby head node is to issue a TAKEOVER HADR command. After triggering a failover on CDP, it will take several minutes for the administrative task scheduler (ATS) to begin invoking the auto-analyze feature and automatic Hive catalog synchronization to the Db2 Big SQL catalog. The normal delay for Db2 Big SQL is 5 minutes, but after a failover, wait at least 10 minutes before changing any ATS task list entries by running ADMIN_TASK_* stored procedures or by using the -enableHcatSync or -disableHcatSync option on the bigsql-admin utility.

If a triggered failover does not complete successfully (the primary head node does not become the standby head node), stop and then restart the Db2 Big SQL service (that is, run the utility with the -stop option, and then run it again with the -start option).

-setLwn count

Specifies that the number of worker node partitions is to be changed. The value of count can be any integer number between 1 and 7 inclusive. Disable the failover controller before using this option (-failoverController Disable).

-smoke

Specifies that a Db2 Big SQL smoke test is to be run on your cluster. You can run a smoke test to validate your installation.

Tip: To run data load tests, run the command /usr/ibmpacks/IBM-Big_SQL/5.0.4.0/bigsql-cli/BIGSQL/package/scripts/bigsql-smoke.sh -l.

-health

Specifies that a Db2 Big SQL health check is to be run on your cluster. This checks whether the Db2 Big SQL database and all Db2 Big SQL nodes are running. It verifies database connections, adds missing database partitions, and registers the new database partitions.

-checkDeadNodes

Checks whether your cluster has any dead nodes. Dead nodes are Db2 Big SQL nodes that are not accessible or that are unable to communicate over the network. Ensure that any dead nodes are removed before checking cluster health.

-enableLogPruning

Specifies that automatic Db2 diagnostic (db2diag) log file pruning is to be enabled.

-disableLogPruning

Specifies that automatic Db2 diagnostic (db2diag) log file pruning is to be disabled.

-logPrune

Specifies that health check logs are to be pruned. You can find the cluster health check log files at /tmp/bigsql-health-check-*.log.

-autoConfigure

Specifies that the Db2 AUTOCONFIGURE command is to be run against the BIGSQL database.

-enableHcatSync

Specifies that automatic Hive catalog syncing to the Db2 Big SQL catalog is to be enabled.

-disableHcatSync

Specifies that automatic Hive catalog syncing to the Db2 Big SQL catalog is to be disabled.

-hcatSync

Specifies that the HCAT_SYNC_OBJECTS stored procedure is to be run.

-restart

Specifies that the Db2 Big SQL service is to be restarted.

-start

Specifies that the Db2 Big SQL service is to be started.

-stop

Specifies that the Db2 Big SQL service is to be stopped.

-startFC

Specifies that the failover controller is to be started on both head nodes, if high availability is enabled. You can combine this option with the -enableHA and -failoverController Enable options, but these options must be specified in a specific order, as shown in the following example:

bigsql-admin -enableHA -failoverController Enable -startFC

-syncStatus

Specifies that node status is to be synchronized with Cloudera Manager. Specify this option when Cloudera Manager shows a different status for the Db2 Big SQL head node and worker nodes than what you see when you run bigsql status from the command line.

-h

Specifies that the help menu is to be displayed.