Upgrading from Gateway 5.x

Owned by Rumena Zlatkova (Deactivated)
Last updated: Dec 26, 2023 by Bala Harish A(AC)
2 min read

The Content Gateway software components are packaged as RPMs in the Swarm distribution bundles downloaded from the Downloads section on the DataCore Support Portal. Download the bundles for both Swarm 11 and Swarm 12 to verify all versions needed to step through a migration from an unsupported version of Elasticsearch are available.

Elasticsearch Migration

Gateway 6 and higher cannot work with Elasticsearch 2.3.3, so be ready to switch over to the new search index before upgrading Gateway. Perform a rolling upgrade to the new Elasticsearch, taking advantage of the fact that each Gateway does not switch to the new Elasticsearch cluster until it is rebooted if more than one Gateway exists.

Work with DataCore Support to plan well and avoid downtime. See https://perifery.atlassian.net/wiki/spaces/public/pages/2443804878, Upgrading from Unsupported Elasticsearch.

  1. Review the upgrade impacts and known issues for the release being upgraded. See https://perifery.atlassian.net/wiki/spaces/public/pages/2443806175.

  2. Complete the upgrade of the Swarm Storage cluster.

  3. Complete the migration to Elasticsearch 6, so it is the primary search feed Gateway uses on restarting after upgrade. See https://perifery.atlassian.net/wiki/spaces/public/pages/2443809821.

  4. (Load balancer users only) In the load balancer, disable traffic for the specific Gateway being upgraded. Allow traffic to continue flowing to the other Gateways.

  5. Stop the Gateway service.

    systemctl stop cloudgateway

  6. Apply any operating system patches before upgrading the Gateway.

  7. Upgrade the RPMs for Gateway, Swarm UI, and Content UI:

    yum -y install caringo-gateway-VERSION.rpm yum -y install caringo-storage-webui-VERSION.rpm yum -y install caringo-gateway-webui-VERSION.rpm

  8. Reload the systemd control scripts:

    systemctl daemon-reload

  9. Fall back and retry if the upgrade failed:

    1. Revert the RPM back to Gateway 5.x.

    2. Restart the Gateway that failed the upgrade.

  10. Review and modify the gateway.cfg and the new logging.yaml configuration files in the /etc/caringo/cloudgateway directory. See .

    • gateway.cfg In the [storage_cluster] section, update the indexerHosts value in to the new Elasticsearch server.

    • logging.yaml  Edit the Syslog block and change the protocol to TCP if the server supports TCP and it is desired.

  11. The Gateway service does not automatically start after a system reboot, so re-enable the service if upgrading from version 5.2.1 or earlier.

  12. Start the Gateway service:

  13. (Load balancer users only) Re-enable client traffic to the newly upgraded Gateway in the load balancer.

  14. Repeat the process for the remaining Gateways.

© DataCore Software Corporation. · https://www.datacore.com · All rights reserved.