Item Update interval not working

[ptader]

An Item I have setup to check every 120 seconds, doesn't. Here is the most current data:

2008-12-11 16:44:49 1229035489 0
2008-12-11 16:09:01 1229033341 0
2008-12-11 15:30:24 1229031024 0
2008-12-11 14:52:17 1229028737 0
2008-12-11 14:12:57 1229026377 0
2008-12-11 13:33:33 1229024013 0
2008-12-11 12:55:23 1229021723 0
2008-12-11 12:13:49 1229019229 0
2008-12-11 11:28:41 1229016521 0
2008-12-11 10:49:21 1229014161 0

Anybody else see items not following their configured Update Intervals?

Thanks
-pt

[MrKen]

What does zabbix_server.log show?
What version of zabbix?

[ptader]

server and client version 1.4.4

Nothing in the server log pertaining to this Item, but it's not set to verbose.

[MrKen]

Is there anything in the log that doesn't look right? Maybe something else is slowing things down.

Also this thread might help: http://www.zabbix.com/forum/showthread.php?p=27399

MrKen

[ptader]

Thanks for the link, it had some helpful tips. Regardless, I'm still faced with an item that only runs once every 10 minutes when it's configured to run every 2 minutes.

The client log, set to verbose, also shows that it's not running every 2 minutes.


15728:20081212:120608 Requested [myItem]
15728:20081212:120608 Run remote command [/usr/bin/sudo /usr/local/bin/check_myItem >/dev/null 2>&1; echo $?] Result [1] [0]...
15727:20081212:121255 Requested [myItem]
15727:20081212:121256 Run remote command [/usr/bin/sudo /usr/local/bin/check_myItem >/dev/null 2>&1; echo $?] Result [1] [0]...
15727:20081212:122157 Requested [myItem]
15727:20081212:122158 Run remote command [/usr/bin/sudo /usr/local/bin/check_myItem >/dev/null 2>&1; echo $?] Result [1] [0]...
15725:20081212:122858 Requested [myItem]
15725:20081212:122859 Run remote command [/usr/bin/sudo /usr/local/bin/check_myItem >/dev/null 2>&1; echo $?] Result [1] [0]...

[ptader]

Unfortunately it looks to effect other Item checks as well but not if the interval is configured to run with an interval greater than 600 seconds. Those check right on time.

I've increase the number of pollers to 150 but that made no difference at all in the Item check time.

Number of hosts (monitored/not monitored/templates/deleted) 1691(1602/73/16/0)
Number of items (monitored/disabled/not supported)[trapper] 147581(135604/2766/9211)[26492]
Number of triggers (enabled/disabled)[true/unknown/false] 74640(74497/143)[14/25445/49038]
Number of events 2229984
Number of alerts 12801

[MrKen]

Surely this can't be right?

Number of hosts (monitored/not monitored/templates/deleted) 1691(1602/73/16/0)
Number of items (monitored/disabled/not supported)[trapper] 147581(135604/2766/9211)[26492]
Number of triggers (enabled/disabled)[true/unknown/false] 74640(74497/143)[14/25445/49038]
Number of events 2229984
Number of alerts 12801

[ptader]

No, this is correct.

All those trappers are Swatch Items (about 15 per host) that trigger on various regex found in system log files (ie. I/O errors, kernel errors, etc.)

[ptader]

Minor update.

The issue may reside with the server. I reconfigured a node experiencing this trouble to use a test server (same version) and the Update Intervals were updated at the correct intervals.

Is the server not sending the update interval correctly to the client?

[Calimero]

Take a look at Administration > Queue.

[ptader]

Interesting, in the "Queue of Items to be updated" I have a range of checks that are really old. For example, the current time is 10:52am but I see Items that are queued going back almost 3 hours ago (first one 8:04am).

My server is sleeping and maybe under utilized by the zabbix processes.

top - 10:57:34 up 279 days, 22:39, 1 user, load average: 0.11, 0.25, 0.34
Mem: 8161728k total, 7397848k used, 763880k free, 147300k buffers
Swap: 16779884k total, 2368k used, 16777516k free, 3908968k cached

Anything I can do to tune Zabbix to use more resources and shorten this queue?