Upgrading a Cisco Meeting Server (CMS) database cluster is a straightforward process thanks to the simplicity of the solution. Thanks, Acano team!

The text below describes the steps for upgrading a three node cluster with two slaves and a single master. Two of the servers are Acano X3’s in a Call Bridge cluster and the other is a virtual appliance providing only database functions.

Before doing anything backups of each CMS should be taken. To take a backup of a Cisco Meeting Server / Acano use command backup snapshot. Using SFTP download the backup file, the license file cms.lic and any certificate files to a local directory.

1. SFTP the upgrade.img for the respective devices into the root directory. From Linux/OSX this can be done from the command line. For Windows, a utility such as WinSCP can be used.

$ sftp admin@192.168.44.12
sftp> put upgrade.img

1. Log into one of the devices via SSH, and confirm which one is the current database cluster master and which ones are the slaves with command database cluster status. Also, confirm there are no current issues with the environment and there are no errors reported in the Web Admin interface on any of the servers.

1. From the slave devices perform the upgrade with command upgrade one device at a time. Before proceeding to the next device ensure the prior one has upgraded successfully using command version and is back online in the database cluster with command database cluster status. Before proceeding to the next server ensure you can log in successfully to the Web Admin interface. It can take approx. 10 minutes after the upgrade completes before the login will be successful. A schema error after login is expected at this point.

Schema Error:

Upgrade:

1. Upgrade the master with command upgrade
2. Once the server comes back online confirm the current database master (it will have rotated when the old master was upgraded) and that all servers are online and in sync with command database cluster status.
3. From the new database master upgrade the schema with command database cluster upgrade_schema

1. Confirm the schema was successfully upgraded with command database cluster status. A status of enabled indicates a successful upgrade.

1. At this point verify the environment is operating correctly including, viewing spaces on each call bridge, dial plans, XMPP service (if relevant) and that there are no errors reported on the Web Admin interface or Syslog. If the deployment had edge servers now proceed to upgrade them.

To rollback an upgrade.

1. On each server upload the prior upgrade.img and run the upgrade command
2. Once each server has downgraded and been verified using version. On each server run command factory_reset app
3. Once each server has rebooted. Restore them from backups using backup rollback
4. Again confirm there are no errors and the environment is working as expected