Disk stats graph missing following update to 5.0.4

[John Munro]

Following upgrade to 5.0.4 from 5.0.2 all is better with the exception of the disk stats page. /boot is showing, but / is missing (see screenshot). Prior to the upgrade, it was showing and recording fine.
New Picture.jpg

df -mlP is showing all part's fine with the root having the usual unusually long dev name of /dev/mapper/Ubuntu-root

Install is Ubuntu 6.04 LTS FOSS 5.0.4

I suspect it is an Ubuntu thing due to the long dev name, but this is only a suspicion.

Is anyone else experiencing this? Would like to confirm before filing a bz report.

[uxbod]

Are there any errors in /var/log/zimbra.log ? You could also try running :-

sh -X /opt/zimbra/libexec/zmgengraphs

to see if it generates any errors ?

[John Munro]

There are no errors or warnings in zimbra.log

Running 'sh -x /opt/zimbra/libexec/zmgengraphs' (not -X) from both root and zimbra account does produce errors saying '...line 19: use: command not found' etc. but this would be expected.

Running /opt/zimbra/libexec/zmgengraphs runs fine and generates the following gif files in logger/db/work (I had to change the extension to png to upload them here)

disk.mail3.dsty.ac.jp.hour.Disk_Usage_0.png

disk.mail3.dsty.ac.jp.hour.Disk_Usage_1.png

The timestamps are correct and the 2nd shows the backup of the new version performed at 3:30am.

[John Munro]

Also checked jetty/logs/access_log.2008-03-28 (most recent) for any 404's and found none related (lots of ...com_zimbra_wikipedia/blank_pixel.gif 404's, but that would be another matter).

So it seems the code is not even looking for the 2nd graph.

[John Munro]

Referencing Possible bug in disk statistics page generation

The script at /opt/zimbra/jetty-6.1.5/webapps/zimbraAdmin/js/zimbraAdmin/statistics/view/ZaServerDiskStatsPage.js looks ok (count still starts at 0) and doesn't appear to have changed in this version.

I am guessing the problem lies in the code that detects how many images to load. As the images are being generated fine, I would think that either this error is independent of server OS, or it is a faulty upgrade installation on this machine.

Can anyone confirm that they are able to see more that one graph in the disk stats?

[PeterH]

Hello,

Cannot confirm seeing more graph but can confirm exact same behaviour with 5.0.3NE and Ubuntu 6.06LTS

Regards,
Peter

[John Munro]

So it looks like it got broke in the 5.0.2 to 5.0.3 upgrade. Can anybody see more than one graph in the disks stats with or without Ubuntu 6.04 LTS?

[mstreet]

Same behavior here.

Disk Stats on /boot
No Disk Stats on /

These little things are annoying but not show stoppers yet..... still sitting on the fence....

[velde064]

Having the same problem on 5.0.3 on ubuntu

I am also getting at my admin account a lots of emails about 34.000 emails that my disk space has exceeded 80% which is not the case.

Anyone seeing also this behavoir?

Mar 29 08:00:10 mail zimbramon[8269]: 8269:err: Disk warning: mail.<server> /dev/md0 at 82%
Mar 29 07:50:11 mail zimbramon[5873]: 5873:err: Disk warning: mail.<server> /dev/md2 at 82%
etc...

[mmorse]

Bug 26216 - [patch] zmdisklog bad logic - reversed disk used with disk available > Bug 26060 - snmp disk checks report incorrect usage.
[SOLVED] ZCS 5.0.3 zmdisklog messed up (contains fix)