[Check_mk (english)] Windows CPU usage -- What's going on?

Archive checkmk-en
1

Hi guys.
Please, check the attached picture.

VMware and Windows (task manager), showed during that time, 100% cpu usage.

Check_mk showed only 25%. ( The average was 100%, which i do not know what that is ). Why is this?? According to Check_MK, everything was OK !!

I’ve read this:

https://mathias-kettner.de/checkmk_check_winperf_processor.util.html

… and tried to understand… but i didn’t.

Can you give me a hint?

I really have to enable this rule ? --> Maximum number of check attempts for service

?? Why… and how should i use it?

That… is the check attempts per minute?

per second? “per check”?

Will it change the outcome? will it show that the server was with 100% cpu?

Thank you!!

Check_mk-Windows-CPU_2.jpg

2

Looking at the graphs it looks like the monitoring server was not able
to get the data from the Windows agent on your host. This was caused by
the high CPU load. So this is a situation where no data was available
and therefor the CPU load check could not alarm.

Have a look at the "Check_MK" service check of this host during the time
of the incident. It should have reported a non-OK state.

Regards

···

Am 18.12.2016 um 16:03 schrieb Gerardo Ferreyra:

*VMware and Windows* (task manager), showed during that time, 100% cpu
usage.
*Check_mk *showed only *25%*.

--
Robert Sander
Heinlein Support GmbH
Schwedter Str. 8/9b, 10119 Berlin

Tel: 030 / 405051-43
Fax: 030 / 405051-19

Zwangsangaben lt. §35a GmbHG:
HRB 93818 B / Amtsgericht Berlin-Charlottenburg,
Geschäftsführer: Peer Heinlein -- Sitz: Berlin

3

But, check_mk was OK. I checked it.
Please, see the attached capture.

What’s that line, showing the CPU usage peak? (“Average”)

Check_mk-Windows-CPU_3.jpg

···

2016-12-18 14:17 GMT-03:00 Robert Sander r.sander@heinlein-support.de:

Am 18.12.2016 um 16:03 schrieb Gerardo Ferreyra:

VMware and Windows (task manager), showed during that time, 100% cpu

usage.

*Check_mk *showed only 25%.

Looking at the graphs it looks like the monitoring server was not able

to get the data from the Windows agent on your host. This was caused by

the high CPU load. So this is a situation where no data was available

and therefor the CPU load check could not alarm.

Have a look at the “Check_MK” service check of this host during the time

of the incident. It should have reported a non-OK state.

Regards

–

Robert Sander

Heinlein Support GmbH

Schwedter Str. 8/9b, 10119 Berlin

http://www.heinlein-support.de

Tel: 030 / 405051-43

Fax: 030 / 405051-19

Zwangsangaben lt. §35a GmbHG:

HRB 93818 B / Amtsgericht Berlin-Charlottenburg,

Geschäftsführer: Peer Heinlein – Sitz: Berlin

checkmk-en mailing list

checkmk-en@lists.mathias-kettner.de

http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en

4

What time frame in the graph are you referring to?

It seemed to spike up to 90% around 1345, then is flat till around

(Could be that the flat part is due to no new data being collected

but
I don’t know if it stores ‘0’, None, or the last value in those
cases…)

···

On 12/20/2016 2:02 PM, Gerardo Ferreyra
wrote:

But, check_mk was OK. I checked it.
Please, see the attached capture.�

What’s that line, showing the CPU usage peak? (“Average”)

      2016-12-18 14:17 GMT-03:00 Robert

Sander r.sander@heinlein-support.de:

        Am

18.12.2016 um 16:03 schrieb Gerardo Ferreyra:

        > *VMware and Windows* (task manager), showed during that

time, 100% cpu

        > usage.

        > *Check_mk *showed only *25%*.



        Looking at the graphs it looks like the monitoring server

was not able

        to get the data from the Windows agent on your host. This

was caused by

        the high CPU load. So this is a situation where no data was

available

        and therefor the CPU load check could not alarm.



        Have a look at the "Check_MK" service check of this host

during the time

        of the incident. It should have reported a non-OK state.



        Regards

        --

        Robert Sander

        Heinlein Support GmbH

        Schwedter Str. 8/9b, 10119 Berlin



        [http://www.heinlein-support.de](http://www.heinlein-support.de)



        Tel: 030 / 405051-43

        Fax: 030 / 405051-19



        Zwangsangaben lt. �35a GmbHG:

        HRB 93818 B / Amtsgericht Berlin-Charlottenburg,

        Gesch�ftsf�hrer: Peer Heinlein -- Sitz: Berlin





        _______________________________________________

        checkmk-en mailing list

        checkmk-en@lists.mathias-kettner.de

        [http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en](http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en)

_______________________________________________
checkmk-en mailing list

checkmk-en@lists.mathias-kettner.dehttp://lists.mathias-kettner.de/mailman/listinfo/checkmk-en

5

Robert nailed it. You can tell because the stats were perfectly flat in the time-frame you showed…which doesn’t happen at that load level.

-chris

···

On Tue, Dec 20, 2016 at 1:30 PM, Jam Mulch spammagnet10@gmail.com wrote:

What time frame in the graph are you referring to?

It seemed to spike up to 90% around 1345, then is flat till around

(Could be that the flat part is due to no new data being collected

but

I don't know if it stores '0', None, or the last value in those

cases…)

  On 12/20/2016 2:02 PM, Gerardo Ferreyra

wrote:

But, check_mk was OK. I checked it.
Please, see the attached capture.

What’s that line, showing the CPU usage peak? (“Average”)

_______________________________________________
checkmk-en mailing list
checkmk-en@lists.mathias-kettner.de
[http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en](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

        Am

18.12.2016 um 16:03 schrieb Gerardo Ferreyra:

        > *VMware and Windows* (task manager), showed during that

time, 100% cpu

        > usage.

        > *Check_mk *showed only *25%*.



        Looking at the graphs it looks like the monitoring server

was not able

        to get the data from the Windows agent on your host. This

was caused by

        the high CPU load. So this is a situation where no data was

available

        and therefor the CPU load check could not alarm.



        Have a look at the "Check_MK" service check of this host

during the time

        of the incident. It should have reported a non-OK state.



        Regards

        --

        Robert Sander

        Heinlein Support GmbH

        Schwedter Str. 8/9b, 10119 Berlin



        [http://www.heinlein-support.de](http://www.heinlein-support.de)



        Tel: 030 / 405051-43

        Fax: 030 / 405051-19



        Zwangsangaben lt. §35a GmbHG:

        HRB 93818 B / Amtsgericht Berlin-Charlottenburg,

        Geschäftsführer: Peer Heinlein -- Sitz: Berlin





        _______________________________________________

        checkmk-en mailing list

        checkmk-en@lists.mathias-kettner.de

        [http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en](http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en)

      2016-12-18 14:17 GMT-03:00 Robert

Sander r.sander@heinlein-support.de: