Striim 3.10.3 documentation

In-place upgrade

An in-place upgrade installs the new version of Striim while keeping the current repository database data. You may do an in-place upgrade to Striim 3.10.3 from version 3.9.7 or later. From earlier versions, use the Export-import upgrade method.

Performing an in-place upgrade from 3.10.3.4 or earlier to 3.10.3.5 or later requires deleting all Elasticsearch data. If you have WActionStore data persisted to Elasticsearch that you want to preserve, use the Export-import upgrade method instead.

Before upgrading, see "Changes that may require modification of your TQL code or workflow" in the release notes (striim/Striim_release_notes_readme.html) so you will be aware of any changes that may be required to your applications before running them in the new version.

  1. Back up the cluster as described in Backing up and restoring a cluster.

  2. Quiesce and undeploy all running applications with persisted streams. Stop and undeploy all other running and deployed applications.

  3. Open the Striim console and enter the following commands to stop the hidden monitoring applications:

    stop application Global.MonitoringSourceApp;
    undeploy application Global.MonitoringSourceApp;
    stop application Global.MonitoringProcessApp;
    undeploy application Global.MonitoringProcessApp;
    exit;
  4. Stop all Forwarding Agents.

  5. If the metadata repository is hosted on PostgreSQL, export all dashboards and their applications, then drop their namespaces. Alternatively, use the Export-import upgrade method.

  6. If upgrading from 3.10.3.4 or earlier to 3.10.3.5 or later, delete all Elasticsearch data. For example, in Linux:

    rm -rf /opt/striim/elasticsearch/data
  7. Upgrade all servers:

    On each server in the cluster, move the Striim configuration file to /opt/striim/conf-backup (or, if you prefer, some other directory) so it will not be removed when you uninstall the old version of Striim:

    sudo mkdir /opt/striim/conf/conf-backup
    sudo mv /opt/striim/conf/startUp.properties /opt/striim/conf/conf-backup/
    

    for Centos: on each server in the cluster, download striim-node-3.10.3-Linux.rpm and enter the following commands (on CentOS 6, omit systemctl):

    sudo systemctl stop striim-node
    sudo rpm -e striim-node
    sudo rpm -ivh striim-node-3.10.3-Linux.rpm 

    for Ubuntu: on each server in the cluster, download striim-node-3.10.3-Linux.deb and enter the following commands (on Ubuntu 14.04, omit systemctl):

    sudo systemctl stop striim-node
    sudo dpkg --remove striim-node
    sudo dpkg -i striim-node-3.10.3-Linux. 
  8. Using a client for your metadata repository host, run the appropriate script in the /opt/striim/conf/ directory (### will match the version of Striim you're upgrading to):

    • for Derby: UpgradeMetadataReposDerbyTo###.sql

    • for Oracle: UpgradeMetadataReposOracleTo###.sql

    • for PostgreSQL: UpgradeMetadataReposPostgresTo###.sql

  9. On each server in the cluster, enter the following command to restore your Striim configuration (adjust as necessary if you backed up your files to a different location):

    sudo cp /opt/striim/conf/conf-backup/startUp.properties /opt/striim/conf/
  10. f you are upgrading from 3.9.8 or later, skip this step.

    On one server, run sudo su - striim /opt/striim/bin/sksConfig.sh and enter passwords for the Striim keystore and the admin and sys users. If hosting the metadata repository on Oracle or PostgreSQL, enter that password as well (see Configuring a DBMS to host Striim's metadata repository). If you are using a Bash or Bourne shell, characters other than letters, numbers, and the following punctuation marks must be escaped: , . _ + : @ % / -

  11. Copy sks.jks snf sksKsy.pwd from /opt/striim/conf/ on that server to /opt/striim/conf/ on all other servers and assign ownership of those files to Striim:

    sudo chown striim sks.jks
    sudo chown striim sksKey.pwd
  12. Start all servers.

  13. If the metadata repository is hosted on PostgreSQL, reload any applications and dashboards you dropped in step 3.

  14. Reload any open processors (see Loading and unloading open processors).

  15. Upgrade and start all Forwarding Agents (see Upgrading Forwarding Agents).