Hi everyone,

right now I have 2 local sites with about 4500 hosts evenly distributed
and running on 2 physical machines.

I was planning clustering the sites with pacemaker and make them
switchable between the 2 physical machines.

Now this thread got me thinking - maybe it's better to split the 2 sites
into multiple smaller ones and make all of them clustered.

I was thinking maybe

1 master site (no services, just frontend)
1 slave site with linux systems
1 slave site with network systems
1 slave site with windows systems
1 slave site with vmware systems
1 slave site as SNMP trap-receiver
Maybe split it up even more as windows/linux systems are quite big chunks

Right now the system load is minimal, just updating the config and restarting
the core is a pain in the neck as it takes ages. Several small sites would 
probably
fix that.

Any thoughts?

Cheers,
Kai

    




Dr. Ing. h.c. F.  Porsche Aktiengesellschaft
Sitz der Gesellschaft: Stuttgart
Registergericht: Amtsgericht Stuttgart HRB-Nr. 730623
Vorsitzender des Aufsichtsrats: Dr. Wolfgang Porsche
Vorstand: Oliver Blume, Vorsitzender
Lutz Meschke, stv. Vorsitzender
Andreas Haffner, Detlev von Platen, Albrecht Reimold, Uwe-Karsten Städter, 
Michael Steiner

Die vorgenannten Angaben werden jeder E-Mail automatisch hinzugefügt. Dies ist 
kein Anerkenntnis,
dass es sich beim Inhalt dieser E-Mail um eine rechtsverbindliche Erklärung der 
Porsche AG handelt.
Erklärungen, die die Porsche AG verpflichten, bedürfen jeweils der Unterschrift 
durch zwei zeichnungs-
berechtigte Personen der AG.





From: checkmk-en <checkmk-en-boun...@lists.mathias-kettner.de> on behalf of 
Andreas Döhler <andreas.doeh...@gmail.com>
Sent: Sunday, March 11, 2018 13:36
To: Marcel Schulte
Cc: <checkmk-en@lists.mathias-kettner.de>
Subject: Re: [Check_mk (english)] Multiple sites per server
  

Hi Marcel,


that is not right the biggest problems on big sites are generation of 
configuration and core restarts.
If you split one very big installation in 3 or 4 smaller sites on the same 
machine the overall systems will run better.
One other point for very big sites is the amount of log information generated, 
this is better handled with smaller sites.


br
Andreas 


Marcel Schulte <schulte.mar...@gmail.com> schrieb am So., 11. März 2018 um 
09:47 Uhr:
  Hi Gerardo,

in terms of performance /never/ split up a big site to multiple sites
on the same system: every  site has its own core, apache, rrdcached
and other processes running, consuming performance.

BR,
Marcel

2018-03-10 20:31 GMT+01:00 Andreas Döhler <andreas.doeh...@gmail.com>:
> Hi Gerardo,
>
> yes i would setup more then one site on such a big system. The size of the
> single sites depends on the usage of this sites.
> One point would be, how often there are changes made on the configuration of
> the monitoring objects.
> If your setup is more static then you can leave all 4k hosts on one site no
> problem.
>
> br
> Andreas
>
> Gerardo Ferreyra <raptor...@gmail.com> schrieb am Sa., 10. März 2018 um
> 19:56 Uhr:
>>
>> Hi guys.
>> Suppose a physical server with enough power (32 cores, 64 gb ram).
>> Do you recommend to create multiple sites on it?
>> It's better in terms of optimization?
>>
>> Or nowadays, one instance/site of CMK EE, can handle the monitoring of
>> 4000 servers without any issue?
>>
>> Thanks!
>> _______________________________________________
>> checkmk-en mailing list
>> checkmk-en@lists.mathias-kettner.de
>>  http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en
>
>
> _______________________________________________
> checkmk-en mailing list
> checkmk-en@lists.mathias-kettner.de
>  http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en
_______________________________________________
checkmk-en mailing list
checkmk-en@lists.mathias-kettner.de
http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en     
_______________________________________________
checkmk-en mailing list
checkmk-en@lists.mathias-kettner.de
http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en

Reply via email to