The Upgrade instructions https://confluence.atlassian.com/crowd/upgrading-crowd-via-automatic-database-upgrade-213519474.html have step to take up the database backup, however we are not touching the database, so why is this needed?
I have used these steps in my non-production environment to upgrade Crowd version, and it worked.
1. Stop Crowd service.
2. Backup Crowd home directory.
3. Rename Crowd installation directory.
4. Remove Crowd Service.
5. Unzip Crowd installation file to D:\Atlassian\Crowd.
6. Copy crowd-init.properties file from backed up Crowd installation directory to new.
7. Install and start Crowd service.
Hi,
The application upgrade may well change database schema/values as part of the startup process. If this happens, and then you need to roll back, you’ll want to have the database backup to revert to. Even if there’s no changes during the upgrade, things can go wrong and you may end up with corrupt data.
I’ve been running Crowd since 2.3 or so, and have been through many upgrades. While I’ve never had to revert the database cause of a failed user upgrade, ensuring there’s a backup to restore from is always still part of the process.
CCM
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.