Zimbra Forums

After migrating my ZCS 6.0.10 Ubuntu 6.06 32bit installatioin to Ubuntu 10.04 64bit I am receiving this error on the Admin GUI Server status page:

Server Status data is not available. To see the server status, logger service must be installed.

I followed Moving ZCS to New Server - Zimbra :: Wiki to move the server and was successful in moving a 6.06 32 bit installation to 8.04 32 bit following these instructions. However, in moving to 10.04 64bit I received this error with every attempt I made. I was also having trouble creatin the self signed ssl which i though was related, see [URL=http://www.zimbra.com/forums/administrators/46631-solved-unable-load-certificate-when-creating-new-self-signed-cert.html[/URL] .
Eventually I completed the migration and then upgraded to 7.0.0 which solved the SSL issue and some stability issues (with ZCS 6.0.10 on Ubuntu 10.04 64 bit I would have to restart zimbra at least once a day).
However, my server status in the GUI is still not available. All services are reported as running via the CLI. In the logs, the only things of note appear to be:
zmstat.out

Warning: Not possible to monitor process stats

03/04/2011 14:57:00: ERROR: No such process: '8090'

03/04/2011 14:57:00: ERROR: No such process: '8091'

Warning: Not possible to monitor process stats

Warning: Not possible to monitor process stats

zmswatch.out

SMTP notification: Mar  4 16:40:00 u10-04-zimbra zimbramon[31310]: 31310:err: Disk warning: u10-04-zimbra.xxx.xx: /dev/mapper/vg00-opt at 87%

SNMP notification: Mar  4 16:40:00 u10-04-zimbra zimbramon[31310]: 31310:err: Disk warning: u10-04-zimbra.xxx.xx: /dev/mapper/vg00-opt at 87%

i: Missing type/value for variable

I am using a VM copy of this installtion (still at ZCS 6.0.10) to try and find the solution. I have gone through the troubleshooting process in the wiki Logger (ZCS 6.0.x and later) - Zimbra :: Wiki, and the only difference I can see between this installation and the working installation on Ubuntu 8.04 is on step 3, where I receive:

1299269700,0,0.00167280576877929,,2.90566362036962,0,0.00167280576877929,0.00334561153755857,0,0

1299270000,0,0.00167280576877929,,2.90566362036962,0,0.00167280576877929,0.00334561153755857,0,0

1299270300,,,,,,,,,

1299270600,,,,,,,,,

1299270900,,,,,,,,,

1299271200,,,,,,,,,

1299271500,,,,,,,,,

1299271800,,,,,,,,,

1299272100,,,,,,,,,

1299272400,,,,,,,,,

1299272700,,,,,,,,,

1299273000,,,,,,,,,
Host: zimbra.xxxx.xx
Can't use an undefined value as an ARRAY reference at /opt/zimbra/libexec/zmrrdfetch line 135.

All other steps appear to be ok.
I have tried, with no luck:

zmsyslogsetup

restart rsyslog

fixperms extended

Of note, prior to migrating my data to the new machine, the fresh install of ZCS was working fine.
I have spent the last week scouring the forums, and while I have found many similar problems I have yet to find one the exact same.

I have had similar issues after the 6 to 7 upgrade and found a post that lead me to also trying running the following as root, which fixed my problem:
/opt/zimbra/libexec/zmsyslogsetup
I also noted that the /var/log/zimbra.log and /var/log/zimbra-stats.log didn't exist until I ran that manually.
Doug

Thanks Doug. I came accross those as well, and did mention in my post that I had tried zmsyslogsetup. I had run it as root which you suggested, but I forgot to mention.
I just checked for zimbra.log and zimbra-stats.log and both are present and appear to be working ok. Here is the last bit from zimbra-stats.log:

Mar  5 13:41:04 u10-04-zimbra zimbramon[9045]: :::2026B9F6-4758-11E0-9B0D-FA03E4635087:::2629392, 576, 0, 904532, 35040, 300, 273552, 260816, 12736, 1896, 12312, 0, 0, 0, 3662556, 2617700, 34359738367, 274944, 34359453688, 0, 0, 0, 0, 0, 2048, 10240, 2086912, 0.90

Mar  5 13:41:05 u10-04-zimbra zimbramon[9394]: 9394:info: zmstat fd.csv: timestamp, fd_count, mailboxd_fd_count:: 03/05/2011 13:41:05, 1856, 274

Mar  5 13:41:06 u10-04-zimbra zimbramon[9043]: 9043:info: zmstat cpu.csv: timestamp, cpu:user, cpu:nice, cpu:sys, cpu:idle, cpu:iowait, cpu:irq, cpu:softirq, cpu0:user, cpu0:nice, cpu0:sys, cpu0:idle, cpu0:iowait, cpu0:irq, cpu0:softirq, cpu1:user, cpu1:nice, cpu1:sys, cpu1:idle, cpu1:iowait, cpu1:irq, cpu1:softirq:: 03/05/2011 13:41:06, 39.9, 0.0, 2.4, 57.5, 0.1, 0.0, 0.0, 39.9, 0.0, 2.5, 57.4, 0.2, 0.0, 0.0, 39.9, 0.0, 2.4, 57.7, 0.0, 0.0, 0.0

So that appears to be working ok....and I just checked, and still receiving the "server status data is not available" on the admin GUI.

OK. I just did I comparison between the old ZCS 6.0.10 VM that I working with trying to fix this problem and the production ZCS 7. Turns out that zmsyslogsetup did partially fix some of the problems.
On the ZCS 7 machine there was no stats data in the zimbra-stats.log file. After running:

root@ziimbra:/opt/zimbra/libexec# ./zmsyslogsetup

updateRsyslogd: Updating /etc/rsyslog.d/50-default.conf...done.

root@ziimbra:/opt/zimbra/libexec# restart rsyslog

rsyslog start/running, process 19940

root@ziimbra:/opt/zimbra/libexec# ./zmfixperms --extended

zimbra@ziimbra:/home/rob$ zmcontrol restart

Stats are now starting to shop up in zimbra-stats.log on the ZCS 7 machine. So on both machines now, stats seem to being collected and I can view them in the Admin GUI (though there hasn't been enough time to collect stats on the ZCS 7 machine, it is no longer giving "no data available" when looking at cpu0:idle etc).
However, the Server Status page in the GUI is still reporting as not abailable on both...

Perhaps another clue to the problem. On my 6.0.10 machine in the admin GUI I can view global server statistics, but if I select the individual server u10-04-zimbra.xxx.xx I receive the error:
[QUOTE]This Server is not an MTA and will not have any MTA statistics[/QUOTE]
Also, when viewing the global Advanced Statistics, I am able to select between two servers. Which I shouldn't be as this is a stand alone installation.
I suspect something happened in my moving process to cause this. So, here is what I have done that got me to here:
My original server was 6.0.10 running on Ubuntu 6.06LTS with hostname zimbra.xxx.xx
I moved this server to a new server running Ubuntu 10.04LTS 64 with hostname u10-04-zimbra.xxx.xx following this quide: Moving ZCS to New Server - Zimbra :: Wiki. I made sure to run

zmsetservername -n u10-04-zimbra.xxx.xx

After not being able to solve this problem, I then moved the server again to a new server(still 10.04LTS 64) with the same hostname as the original: zimbra.xxx.xx. Again following the quides instructions. I then upgraded this install to ZCS 7. This is running as the production server now, while I have kept u10-04-zimbra to try and sort out this problem.
On both servers, both hostnames appear in the Advanced statistics tab, though they only have data available about themselves, the other hostname reporst no data available. So it seems like there is a hostname somwhere in the ldap or config that was not changed/removed in the move process...Any ideas?

Well, I spoke too soon. While stats are appearing in zimbra-stats.log, the admin GUI is still showing no data on the ZCS 7 installation....
After running the zmsyslogsetup on the ZCS 7 machine, I received the following errors from the Database integrity check on the weekend:

Database errors found.

/opt/zimbra/mysql/bin/mysqlcheck --defaults-file=/opt/zimbra/conf/my.cnf -S /opt/zimbra/db/mysql.sock -A -C -s -u root --password=xxx

mysql.general_log

 Error    : You can't use locks with log tables.

 mysql.slow_log

 Error    : You can't use locks with log tables.

Database errors found.

/opt/zimbra/mysql/bin/mysqlcheck --defaults-file=/opt/zimbra/conf/my.cnf -S /opt/zimbra/db/mysql.sock -A -C -s -u root --password=xxx

mysql.general_log

Error : You can't use locks with log tables.

mysql.slow_log

Error : You can't use locks with log tables.
That 2 lines are mysql db error which can be safely ignore.
general_log and slow_log are mysql archive table which cannot use db lock.

Thanks, John. Good to know.

Well, I've managed to get it working on my ZCS 6 VM. Basically I copied the data from /opt/zimbra/store, /opt/zimbra/index, /opt/zimbra/db/data (I also made sure I had my ldap, DB_CONFIG, and localconfig.xml backups to use), then uninstalled ZCS 6 completely.
Following the instructions here Network Edition: Moving from 32-bit to 64-bit Server - Zimbra :: Wiki I reinstalled ZCS 6. I noticed when installing this time for some reason the install had Zimbra-Logger = Disabled in the config menu, so I changed this to enabled. Before I copying any of the old data back I checked and everything was working OK on the new install. Every other time this has been the case as well, so not sure if the Zimbra-Logger setting was a factor.
I then followed the rest of the instrctions to put the data back in. On restarting Zimbra, mailboxd was not running. I thought the might have something to do with the self signed certs again, so I tried to recreate them, but it failed same as: [URL=http://www.zimbra.com/forums/administrators/46631-solved-unable-load-certificate-when-creating-new-self-signed-cert.html[/URL]. At that point I just decided to restart Zimbra, and voila - it seems to be working now....
So, I am going to back up my productions ZCS 7 VM tonight and try the same thing and see what happens...
In my attempts to move to 64 bit ZCS I had also tried the instructions here: Moving ZCS to New Server - Zimbra :: Wiki, which involves copying the whole config directory. I am wondering if copying the old localconfig.xml might have had something to do with the problems...

Well, I did the same on my ZCS 7 VM and all is working now. I think it had to do with the localconfig.xml, but here was one other thing I did do and that was run:

dpkg-reconfigure sysstat

I don't think that should have anything to do with the server status, but I don't know.