My overnight backup is causing issues with the zimbra mta.
My script is based on those commonly found here in the forum and it involves zmcontorl stop , local rsync to another folder and then zmcontrol start. Recently a problem has developed where the zmcontrol start process is having heartburn.
The first event in Zimbra log to indicate problems is during zmcontrol stop where the logger process is not shutting down. Assuming that isn't the culprit. when zmcontrol start begins I am seeing slapd attempting to execute a STARTTLS, it establishes and then does a BIND, and a search of LDAP. The search result tag=101 err=0 nentries=1 test=
and then I see closed(connection lost) This happens over and over. Eventually I see
mail postfix/postqueue: fatal: Queue report unavailable - mail system is down
then you see more of the STARTTLS messages and eventually
zimbramon .....zmdailyreport: Failed to email report: Connection refused at /opt/zimbra/zimbramon/lib/Mail/Mailer.pm line 150.
(which I assume is bacause the mta is down)
So what can I do? double start and shutdown ZImbra? Add something in my script to check mta status and execute zmmtactl start if it is stopped?