Process monitoring

Hi all,

I need to monitor something like 2000 process spread among several servers.
I would like to get some guidelines one what will be the best and fastest way to do this.
My idea was to a process discovery entry for each server, but since I’ve around 500 servers, and I do not find any bulk import option, I’m getting a bit lost on what would be the best approach to do this.

Btw, initially my idea was to built a shell script for this, and I even developed a simple one, but then I wasn’t able to discover it, when I do a full scan inside WATO on the service host panel.

I hope to get some clarification and ideas on how to better handle this situation.

Cheers

This is my current configuration, where shall I add the process names to monitor?

The ruleset “Process discovery” allows you to specify processes that should be monitored if running.
Create a rule for each process there and let service discovery find them.

Hallo,
do you know the names of the processes?
I had some days ago a similiar problem in a small 10 host network.
I my case the problem was to „catch“ all processes permanent and those running only some minutes per day.
Ralf

Hi r.sander,
I’ve configured like this and I was able to view then this rule on the discovery for this server, meantime, on this rule I’m missing all the process I need to monitor. Where can I specify the process I need to monitor.
I leave here an example of what processes I need to monitor:
4 with the user oand4 (inc1,bwd5,pat,aa_ser)
3 with the user zzet (inc2,tar,pat)

Hi rprengle,
Yes I know their names.
I leave here an example of what processes I need to monitor:
4 with the user oand4 (inc1,bwd5,pat,aa_ser)
3 with the user zzet (inc2,tar,pat)
I’ve added a photo of my current configuration. I think I’m missing some configuration, somewhere.

Activate the parameter “Process Matching”. There you can specify the exact process name or a regular expression matching the command line.

Thanks r.sander.

Ok, I got what you say, but this way I need to create a rule for each single process, since I’ve around 2000 to monitor, I need to create 2000 rules.

Isn’t any other way to do this? or a way of performing a bulk import for this rule set?

Are all 2000 processes different? Then yes.
If you have the same process on many hosts then you need only one rule that matches the process.

You can use the WATO API to edit the ruleset: https://docs.checkmk.com/latest/en/web_api.html

Hallo,
you an use tags to organize hosts and the processes.
Give every host the tags for every process you want to check and than use these tags in the checks to select the target systems.
And yes you can configure hosts with tags using a bulk function.

Ralf

Ok,
Basically I’ll need to create rules aggregated by server and user. Somehow this will reduce some work, but I’ll have a lot to insert. I’m going to explore the WATO API a bit in order to understand if it can help to reduce the time of import.

Meantime, I’m introducing one by one to perform some tests and I’ve found a process that I cannot found when I do a full scan in the server. I’m trying to monitor an oracle listener, but it doesn’t show up, strange is that the rule is well done and if I can from the listener to different process name, it find it.

Is there a way so I can troubleshoot this in a sort of log, in order to understand what is going wrong with this in particular?

Thanks again in advance.

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed. Contact @fayepal if you think this should be re-opened.