I have one site installed with Zimbra Collaboration Network Edition in a single server setup and am planning to set up a second physical site with a hot stand bye Zimbra slave server if the master server goes down or is not longer accessible. If that happens I have to do some things manually, like importing a backup and redo logs and changing DNS etc.
To achieve this I am rsyncing the backup folder on a daily basis and I am also rsyncing the redolog folder every 10th minutes from the master to the slave. Both backups and redologs are stored outside Zimbra installation on the slave, in a "remote" folder.
I have some question about this.
1. As I understand I must setup Zimbra om my slave server just as the first server. Using the same passwords etc and also activate the license. Do I need a second Zimbra license for this?
2. Using the new NG modules for backup, on the slave server, can I simply import a daily backup from the "remote" folder?
3. And how to import the redo logs after importing a daily backup? Using CLI?
Or is there any other way to have a Zimbra server stand-by, ready to serve if the master server goes down or is not longer accessible?
- Zimbra Collaboration 9.0.0 now available. Read the release notes.
- Zimbra Collaboration 8.8.15 LTS now available. Read the release notes.
- Are you a Zimbra Developer? You can find some interesting stuff in our Official GitHub, Blog and the Community Github.
- Zimbra is Open Source! Read the FAQ. You can also contribute and build binary from source!
Discuss your pilot or production implementation with other Zimbra admins or our engineers.
2 posts • Page 1 of 1
There is no currently supported HA solution for ZCS (mailstores), you'll have to wait for ZCS 9 (if it's still targetted for that version) for the 'always on' feature. If you search the forums for HA & DRBD you'll find further information on the possibility of creating a HA solution but, IMO, it's more trouble than it's worth. I believe you'd be far better concentrating on a good Backup/Restore and disaster recovery solution and planning for that (including testing the failed and recovery situation) - make sure it's well documented and well tested and you should be OK with that.
Who is online
Users browsing this forum: No registered users and 9 guests