[omd-users] OMD 2.60 - Activate Changes taking too long

Andreas Döhler andreas.doehler at gmail.com
Wed Oct 4 14:57:58 CEST 2017


Hi Mahesh,

on my bigger systems this has no impact on the overall performance. It is
better for applying changes as every instance has not so much to do and the
generation of configuration happens in parallel. With such a setup
EventConsole is only running in one instance and all other services must
also run on a single instance. For PNP and RRDCache i think it is better if
the cache files are not so big and the writes to disc happens in smaller
chunks.

br
Andreas

Mahesh Patade <patademahesh at gmail.com> schrieb am Sa., 16. Sep. 2017 um
11:03 Uhr:

> Thanks for your prompt reply andreas.
>
> We are already running two sites for two different DCs. Hosts are almost
> equally divided between them.
>
> IMO creating more sites on single instance would lead to resources crunch.
> Will it? Because the two different sites will run their own nagios core and
> all other services. Correct me if i am wrong.
>
> Thanks,
> Mahesh
>
> On Sep 16, 2017 1:44 PM, "Andreas Döhler" <andreas.doehler at gmail.com>
> wrote:
>
> Hi Mahesh,
>
> One option you have is to split your setup to more then one site. It is
> possible to run some sites (4-5) without problems on the same host and do
> the configuration from only one of these sites.
> I have some setups with hosts between 1000 and 2000 and only around 500
> per site. Now if i configure some changes only for one sub-folder inside
> WATO only the site theses hosts belong to are restarted.
> With such a setup there is only one restriction. It is not possible to
> have parent child relations across sites.
>
> br
> Andreas
>
> Mahesh Patade <patademahesh at gmail.com> schrieb am Sa., 16. Sep. 2017 um
> 09:11 Uhr:
>
>> Hi All,
>>
>> Recently we upgraded our OMD from 1.20 to 2.60. We are monitoring approx.
>> 1500 hosts and 32k services. Everything is working fine except "Activate
>> changes" task. It takes around 10-15 minutes to make changes.
>>
>> When i checked in the background it runs check_mk.py - automation -
>> restart task which iterates over pre-compiled checks for all the hosts
>> monitored from that OMD instance. Which in turns takes so much time because
>> its a synchronous process.
>>
>> Is there any we can run this only for changed services/hosts OR make this
>> process run parallel ?
>>
>> Thanks a lot for such a wonderful gift to OSS community.
>>
>>
>> Best Regards,
>> Mahesh
>>
>> _______________________________________________
>> omd-users mailing list
>> omd-users at lists.mathias-kettner.de
>> http://lists.mathias-kettner.de/mailman/listinfo/omd-users
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mathias-kettner.de/pipermail/omd-users/attachments/20171004/942c3703/attachment.html>


More information about the omd-users mailing list