Quote:
|
Originally Posted by DGS_David Em, glad to see Zimbra feel this is important, no reply or response. This really inspires me as I start to rebuild things that a future upgrade will not return me to oblivion at some point.
Oh well, guess I'll have better backups next time although I might edit the edit the funtion removeExistingInstall() in util/utilfunc.sh before I upgrade next 
My fault not good enough backups, Zimbra fault that the upgrade managed to bork and kill my install! |
We feel this is important but this is not something we've had happen ever since the GA (Feb 2006). So hard to believe it *just happened*. A couple things could have made this better although not much can be done at this time.
- move backups to somewhere else. this is basic DR practice. ie if the drive(s) would have failed you'd be in the same boat. We use rsync twice a day to push backups to a different server. We also take tape of that backup off-site in regular intervals.
- when something goes wrong try to find out what it was. look in the logs, look at the state of your install.
We will make sure to add a prompt on the /opt/zimbra removal but this won't help those that aren't moving backups off disk or off site. If they allow it to be removed.
If you can send your install logs from /tmp that would be helpful. Maybe we can figure out why the 1st install failed.