[SOLVED] zimbraRedoLogRollover settings ignored

[sblandford]

I've just changed a few zimbraRedoLogRollover related settings to get the redolog to roll-over every 10 minutes.

This doesn't seem to be happening although it has worked in the past on another server.

I can't easily restart the whole server (would have to schedule with client) so I would like to know if there is a single part of the server I have to restart to get the changes to come into effect.


Settings are:
zimbraRedoLogArchiveDir: redolog/archive
zimbraRedoLogCrashRecoveryLookbackSec: 10
zimbraRedoLogDeleteOnRollover: FALSE
zimbraRedoLogEnabled: TRUE
zimbraRedoLogFsyncIntervalMS: 10
zimbraRedoLogLogPath: redolog/redo.log
zimbraRedoLogRolloverFileSizeKB: 1
zimbraRedoLogRolloverHardMaxFileSizeKB: 50000
zimbraRedoLogRolloverMinFileAge: 10


Zimbra version:
7.0.1_GA_3105.RHEL5_64-20110304210645 on Centos 5.x.

Thanks & regards...

Update: It's working on a server I made as a mirror. So it works after a Zimbra restart. Restarting the whole of Zimbra just isn't an option for the first server.

[fyd]

So whats the incremental backup settings like ..

[sblandford]

Actually I got impatient and just restarted zmmailboxctrl. Fortunately it didn't take more than a few seconds (on 16GB of mailbox data) and no one noticed.

The archive directory is getting filled up as expected now.

Thanks & regards...

[fyd]

Originally Posted by sblandford

Actually I got impatient and just restarted zmmailboxctrl. Fortunately it didn't take more than a few seconds (on 16GB of mailbox data) and no one noticed.

Restarting with zmmailboxdctl won't take that long no matter how large your mailbox is.

[sblandford]

Originally Posted by fyd

Restarting with zmmailboxdctl won't take that long no matter how large your mailbox is.

Thanks, that's good to know!