statistics and monitoring

[hardaur]

So I have Zimbra installed, and honestly, I'm pretty impressed. Simple installation (sans the library link problem with FC4) and easy to get up and running. Only one problem. . .

SNMP refuses to start, no errors in the logs or zmcontrol. Looks like it tries to start from the logs:

Sep 29 22:57:26 zimbra zimbramon[351]: 351:info: Starting services
Sep 29 22:57:27 zimbra zimbramon[351]: 351:info: STARTING ldap
Sep 29 22:57:27 zimbra zimbramon[351]: 351:info: start service ldap
Sep 29 22:57:27 zimbra zimbramon[351]: 351:info: start app ldap
Sep 29 22:57:27 zimbra zimbramon[351]: 351:info: Starting child ldap: (20050929225727)
Sep 29 22:57:35 zimbra zimbramon[351]: 351:info: STARTING services
Sep 29 22:57:35 zimbra zimbramon[351]: 351:info: STARTING snmp
Sep 29 22:57:35 zimbra zimbramon[351]: 351:info: start service snmp
Sep 29 22:57:35 zimbra zimbramon[351]: 351:info: start app swatch
Sep 29 22:57:35 zimbra zimbramon[351]: 351:info: Starting child swatch: (20050929225735)
Sep 29 22:57:35 zimbra zimbramon[351]: 351:info: STARTING antivirus
Sep 29 22:57:35 zimbra zimbramon[351]: 351:info: start service antivirus
Sep 29 22:57:35 zimbra zimbramon[351]: 351:info: start app amavisd
Sep 29 22:57:36 zimbra zimbramon[351]: 351:info: Starting child amavisd: (20050929225736)


So I'm getting no monitoring or statistics from the admin interface. Any suggestions?

Thanks!
H

P.S. The zmcontrol output:


[zimbra@zimbra ~]$ zmcontrol -d start
Calling start (zimbra.mydomain.com)

perl /opt/zimbra/bin/zmmon &

RESPONSE: (Zimbra::Mon::serviceInfo)
antispam
status Running
antivirus
status Running
ldap
status Running
mailbox
status Running
mta
status Running
snmp
status Stopped
host
ip xxx.xxx.xxx.xxx
name zimbra.mydomain.com
[zimbra@zimbra ~]$

[marcmac]

check /tmp/swatch.out for errors - I suspect that zmsnmpinit failed - you can re-run that, and check for errors as well.

[hardaur]

Ok, the /tmp/swatch.out gave me what I needed. Aparently the suite was started at some point as root and swatch didn't have the access to /var/log/zimbra.log that it wanted. chmod zimbra:zimbra /var/log/zimbra.log took care of that.

So, now, everything seems to be running, however, I'm still not seeing any statistics (been running for 3 or 4 days), all stats still read as not yet available. Any ideas?

Thanks!

H

[marcmac]

Many things could be wrong - the stats package in this build is not in the best shape - we're currently working on making it more robust.

Check the zimbra crontab, and you'll see many scripts for gathering and processing stats, and graph generation - check the output files for those scripts for errors. It could be we're not getting the data into the db, or that the graph generation is failing.

[hardaur]

zmaggregatestats complains that it has no data:

SN: Slice:/opt/zimbra/store
P: 1
select time, value from server_stat where name = 'Slice:/opt/zimbra/store' order by time desc limit 1
Slice:/opt/zimbra/store
No data available to aggregate

zmaggregatestats then complains of a duplicate key:

./zmaggregatestatsdaily
Aggregating LmtpRcvdData:total:bytes for 2005-09-30

select time, value from aggregate_stat where name = 'LmtpRcvdData:total:bytes' and time like '2005-09-30%' order by time desc
Found 2005-09-30 00:00:00 0
CV: 0
insert into aggregate_stat (time, name, value, period) values (?,?,?,?)
DBD::mysql::st execute failed: Duplicate entry '2005-09-30 00:00:00-LmtpRcvdData:total:bytes-24' for key 1 at ./zmaggregatestatsdaily line 188.
Undefined subroutine &Zimbra::Logger::Log called at ./zmaggregatestatsdaily line 189.

[marcmac]

The duplicate key just means that it's already aggregated that period. What did zmfetchstats output? It sounds like the disk slice data isn't making it into the table.

[hardaur]

Ok, here's what transpired (I know, dirty fix, but I'm just running on a test platform).

[zimbra@zimbra ~]$ zmfetchstats
Can't locate Soap.pm in @INC (@INC contains: /opt/zimbra/zimbramon/lib/i386-linux-thread-multi /opt/zimbra/zimbramon/lib /opt/zimbra/zimbramon/lib/Zimbra/Mon/SOAP /opt/zimbra/zimbramon/lib/Zimbra/Mon /opt/zimbra/zimbramon/lib/i386-linux-thread-multi/ /usr/lib/perl5/site_perl/5.8.6/i386-linux-thread-multi /usr/lib/perl5/site_perl/5.8.5/i386-linux-thread-multi /usr/lib/perl5/site_perl/5.8.4/i386-linux-thread-multi /usr/lib/perl5/site_perl/5.8.3/i386-linux-thread-multi /usr/lib/perl5/site_perl/5.8.6 /usr/lib/perl5/site_perl/5.8.5 /usr/lib/perl5/site_perl/5.8.4 /usr/lib/perl5/site_perl/5.8.3 /usr/lib/perl5/site_perl /usr/lib/perl5/vendor_perl/5.8.6/i386-linux-thread-multi /usr/lib/perl5/vendor_perl/5.8.5/i386-linux-thread-multi /usr/lib/perl5/vendor_perl/5.8.4/i386-linux-thread-multi /usr/lib/perl5/vendor_perl/5.8.3/i386-linux-thread-multi /usr/lib/perl5/vendor_perl/5.8.6 /usr/lib/perl5/vendor_perl/5.8.5 /usr/lib/perl5/vendor_perl/5.8.4 /usr/lib/perl5/vendor_perl/5.8.3 /usr/lib/perl5/vendor_perl /usr/lib/perl5/5.8.6/i386-linux-thread-multi /usr/lib/perl5/5.8.6 .) at /opt/zimbra/bin/zmfetchstats line 44.
BEGIN failed--compilation aborted at /opt/zimbra/bin/zmfetchstats line 44.
[zimbra@zimbra ~]$ locate Soap.pm
/opt/zimbra/zimbramon/lib/Zimbra/SOAP/Soap.pm
[zimbra@zimbra ~]$ ln -s /opt/zimbra/zimbramon/lib/Zimbra/SOAP/Soap.pm /opt/zimbra/zimbramon/lib/Soap.pm
[zimbra@zimbra ~]$ zmfetchstats
Fetching from host zimbra from 0000000000 till 0000000000, period

I don't see zmfetch stats in the cron, is that spawned from another process? Guessing that it might be trying to run now but missing some command line args.

Thanks yet again!

H

[marcmac]

zmfetchstats is called from one of the other zm* processes - if you've fixed the Soap.pm include (my bug, btw the stats gathering should work better now.