Andreas,

Thank you for the answer. I took another look and want to clarify what I'm 
seeing:

If one of the hosts that GenericJMX is watching is hung, the plugin will no 
longer gather data from all the hosts. All other plugins (CPU, swap, memory, 
etc) work fine.

Does anyone on the list use GenericJMX to watch multiple hosts and has seen 
this issue?




- josh jacob



On Mar 3, 2012, at 6:00 AM, 
<[email protected]<mailto:[email protected]>> 
<[email protected]<mailto:[email protected]>> wrote:

Hi.

Yes we see this quiet often but it is harmless and caused by jitter
in your system time. Depending on your collectd interval and the system load
this will occur quiet often. We usually see this messages on a VMware
under heavy I/O load.

As said before, this is nothing to worry about.

HTH,

Andreas.

On 03/02/12 15:15, Josh Jacob wrote:
I have a collectd 5.0.1 instance monitoring 6 local Tomcat instances with the 
GenericJMX plugin. One of my Tomcat instances will occasionally run out of heap 
and essentially become unresponsive. When that happens, collectd/GenericJMX 
won't collect data from any of the other Tomcat instances even though the other 
instances are running fine. The collectd log starts to fill with lots of these 
messages:

[2012-03-02 06:39:59] rrdtool plugin: rrd_update_r 
(/opt/collectd/var/lib/collectd/rrd/localhost/GenericJMX-tomcat1-memory_pool-CMS
 Perm Gen/jmx_memory-init.rrd) failed: 
/opt/collectd/var/lib/collectd/rrd/localhost/GenericJMX-tomcat1-memory_pool-CMS 
Perm Gen/jmx_memory-init.rrd: illegal attempt to update using time 1330688269 
when last update time is 1330688269 (minimum one second step)

Has anyone on the list seen or addressed anything like this?

- josh


_______________________________________________
collectd mailing list
[email protected]<mailto:[email protected]>
http://mailman.verplant.org/listinfo/collectd



_______________________________________________
collectd mailing list
[email protected]
http://mailman.verplant.org/listinfo/collectd

Reply via email to