Re: [Nagios-devel] Nagios 3 Performance Monitoring
Posted: Sun Oct 28, 2007 12:45 pm
Hi Aaron,
Aaron Devey schrieb:
> Hi Hendrik,
> By any chance, are you using NDO utils? If so, do you see the same
> results with NDO utils disabled?
No NDO Utils at the moment. It's just a testing machine before leaving
Nagios 2
> Does the load/process count, on your machine increase at a similar rate
> to your nagios latency?
Load is relativ stable. Since I normally run 5 parallel instance I had a
load around 15 to 20 (4 CPUs).
I know it sounds like a big thing, but aslong the system doesn't begin
to swap or the I/O wait increases... i think it is not optimal but a
little bit away from beeing bad.
Process count might be interesting for that, I will add some general
process count monitoring tomorrow morning at work.
Thanks for the idea.
> There are also several settings in nagios.cfg that can cause degraded
> nagios performance, depending on how they are tuned. Perhaps you could
> make a copy of your nagios.cfg available for view? (with any
> confidential information edited out, of course)
If you only want to see the nagios.cfg, there isn't any confidential
data in.
Out of my mind, my nagios.cfg diffs from the sample-config in:
- processing performance data (in that form that nagios only writes the
to a file which is rotatet by "mv" every 30 seconds)
- decrease of the different timeouts
- disabling ePN
- enabling large install tweaks
- since today: enabling rescheduling window
- ext cmd checking '-1' (this might be a little stressfactor)
I'll see if I can post my config tomorrow from office.
-
Hendrik
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]
Aaron Devey schrieb:
> Hi Hendrik,
> By any chance, are you using NDO utils? If so, do you see the same
> results with NDO utils disabled?
No NDO Utils at the moment. It's just a testing machine before leaving
Nagios 2
> Does the load/process count, on your machine increase at a similar rate
> to your nagios latency?
Load is relativ stable. Since I normally run 5 parallel instance I had a
load around 15 to 20 (4 CPUs).
I know it sounds like a big thing, but aslong the system doesn't begin
to swap or the I/O wait increases... i think it is not optimal but a
little bit away from beeing bad.
Process count might be interesting for that, I will add some general
process count monitoring tomorrow morning at work.
Thanks for the idea.
> There are also several settings in nagios.cfg that can cause degraded
> nagios performance, depending on how they are tuned. Perhaps you could
> make a copy of your nagios.cfg available for view? (with any
> confidential information edited out, of course)
If you only want to see the nagios.cfg, there isn't any confidential
data in.
Out of my mind, my nagios.cfg diffs from the sample-config in:
- processing performance data (in that form that nagios only writes the
to a file which is rotatet by "mv" every 30 seconds)
- decrease of the different timeouts
- disabling ePN
- enabling large install tweaks
- since today: enabling rescheduling window
- ext cmd checking '-1' (this might be a little stressfactor)
I'll see if I can post my config tomorrow from office.
-
Hendrik
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]