[Check_mk (english)] how are calculated snmp interface in errors.

Jim Welch jim.welch at oit.gatech.edu
Tue Feb 17 16:17:33 CET 2015


Are these RHEL 6 hosts? I think RedHat has changed the way that the 
network interface
errors are reported in recent kernels. Try 'cat /proc/net/dev' and see 
if you are consistently
getting data in the 'errs' and 'fifo' columns. If so, it's not check_mk, 
but the kernel that
is giving you the bogus(?) data. I had to set the net-in thresholds to 
keep from getting
notifications on some hosts after I updated them. I think they are 
including some
data as errors where before they were not. It seems that the percentage 
is worse on
systems with little 'normal' traffic. It may also have something to do 
with channel
bonded interfaces, though I haven't tried to verify that.



On 02/17/2015 06:32 AM, Mario Garcia wrote:
> nagios/cmk started reporting in-error of more than 5%. our network 
> teams claims it is not true.. as there are not so many errors on the 
> interfaces.
>
> I would like to understand why then cmk is returning such huge values
>
> thank you
>
> -- 
> Mario Garcia
> ABSI NV | System Engineer
> Humaniteitslaan 224 - Bld de l'Humanité - 1190 Bruxelles
> mariog at absi.be <mailto:mariog at absi.be> / http://www.absi.be / Tel: 
> +32(0)2 333 40 00 Fax: +32(0)2 333 40 60
>
>
> _______________________________________________
> checkmk-en mailing list
> checkmk-en at lists.mathias-kettner.de
> http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mathias-kettner.de/pipermail/checkmk-en/attachments/20150217/5c210952/attachment.html>


More information about the checkmk-en mailing list