Before upgrading your Meeting Servers, take a backup of the current configuration on each of the servers. Use the backup snapshot <name> command. Save the backup safely to a local server. See the MMP Command Reference guide for full details. Do not rely on the automatic backup file generated by the upgrade/downgrade process as it may be inaccessible in the event of a failed upgrade/downgrade.
- Save the cms.lic and certificate files to the local drive.
Before upgrading, check the database cluster status. Check that all calls (SIP and clients) are working and no fault conditions are listed on the Web Admin pages.
Upgrade all Core servers one by one. Start with the non-database servers first, followed by the slave database servers, and the master database server last. Wait until each server has fully booted, and for database servers to connect to the database cluster before moving on to the next server.
For each server: wait approximately 10 to 12 minutes for the Web Admin to be available after upgrading the server. Log in to the Web Admin and go to Status > General and verify the new version is showing under System status. Do not proceed beyond this point until the Web admin login has been successful. If the version is not correct, upload a new image to this server, upgrade and wait to get back to this point.
Once all servers have upgraded, check that the database servers are connected and in sync (check this using database cluster status). Do not go onto the next step until the database servers are in sync.
- Run the database cluster upgrade_schema command on the master database server. Use the database cluster status command to check that the database schema has upgraded successfully. (For instructions on upgrading the database schema, see the Scalability & Resilience Deployment Guide.)
- Check that the Web Admin interface on each Call Bridge can display the list of coSpaces.
- Upgrade all Edge servers.
You have completed upgrading your resilent Meeting Server deployment.
Finally, check that:
dial plans are intact
- XMPP service is connected
- no fault conditions are reported on the Web Admin and log files
- you can connect using SIP and Cisco Meeting Apps (as well as Web Bridge if that is supported).
You have completed the upgrade.
Note: Rollback procedure. If anything unexpected happens after you upgrade the servers and you decide to downgrade to the previous release, downgrade each server in turn and then use the MMP command factory_reset app on each server. Once the server has rebooted from a factory reset, use the backup rollback <name> command to restore the backup files on each server. In the case of XMPP clustering, you will have to re-cluster XMPP (by reinitializing XMPP on the node and joining the others to it) as that will not successfully restore from the backup rollback <name> command. Providing you restore on the server the backup file that was created from the server, the license files and certificate files will match the server.