High CPU alarm

newer
not getting cpu cores and network...

Daniel Johansson

12 Dec 2019 12 Dec '19

2:14 p.m.

I have a synology nas with 1CPU and 4 cores in it. Whenever 1 core gets a shitload to do I get a CPU high alarm although it's not the combined cpu load that is high.

Is there something I can do in the alarm checker to combine the cpu loads of all cores and trigger an alarm when all 4 combined is over for example 80% load?

Regards Daniel

Attachments:

attachment.html (text/html — 633 bytes)
aoifaillncgjdkje.png (image/png — 13.5 KB)

Show replies by date

Adam Armstrong

12 Dec 12 Dec

3:15 p.m.

Does this device not generate the "average" meta-entity?

We generate this on some device types for exactly this reason.

What is the OS-type?

adam. On 2019-12-12 13:14:24, Daniel Johansson via observium observium@observium.org wrote: Hi

I have a synology nas with 1CPU and 4 cores in it. Whenever 1 core gets a shitload to do I get a CPU high alarm although it's not the combined cpu load that is high.

Is there something I can do in the alarm checker to combine the cpu loads of all cores and trigger an alarm when all 4 combined is over for example 80% load?

Regards Daniel

Daniel Johansson

3:34 p.m.

It's a Synology nas and their OS is Synology DSM 6.2-24922. It's a linux based OS.

In the poller script it says: o OS dsm o OS Group unix

Regards Daniel

On 2019-12-12 15:15, Adam Armstrong via observium wrote:

...

Does this device not generate the "average" meta-entity?

We generate this on some device types for exactly this reason.

What is the OS-type?

adam.

...
On 2019-12-12 13:14:24, Daniel Johansson via observium observium@observium.org wrote:

Hi

I have a synology nas with 1CPU and 4 cores in it. Whenever 1 core gets a shitload to do I get a CPU high alarm although it's not the combined cpu load that is high.

Is there something I can do in the alarm checker to combine the cpu loads of all cores and trigger an alarm when all 4 combined is over for example 80% load?

Regards Daniel

observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

adama＠observium.org

5:41 p.m.

This should have an “average” meta entity that you can alert on, and ignore the other ones.

It’s marked as unix group, and has HOST-RESOURCES-MIB, which generates the average entity.

Adam.

From: observium observium-bounces@observium.org On Behalf Of Daniel Johansson via observium Sent: 12 December 2019 14:34 To: Adam Armstrong via observium observium@observium.org Cc: Daniel Johansson daz@voodoo-people.com Subject: Re: [Observium] High CPU alarm

It's a Synology nas and their OS is Synology DSM 6.2-24922. It's a linux based OS.

In the poller script it says: o OS dsm o OS Group unix

Regards Daniel

On 2019-12-12 15:15, Adam Armstrong via observium wrote:

Does this device not generate the "average" meta-entity?

We generate this on some device types for exactly this reason.

What is the OS-type?

adam.

On 2019-12-12 13:14:24, Daniel Johansson via observium mailto:observium@observium.org observium@observium.org wrote:

I have a synology nas with 1CPU and 4 cores in it. Whenever 1 core gets a shitload to do I get a CPU high alarm although it's not the combined cpu load that is high.

Is there something I can do in the alarm checker to combine the cpu loads of all cores and trigger an alarm when all 4 combined is over for example 80% load?

Regards Daniel

_______________________________________________ observium mailing list observium@observium.org mailto:observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

Daniel Johansson

5:58 p.m.

So what meta entry should I create the alert on?

Regards Daniel

On 2019-12-12 17:41, adama--- via observium wrote:

...

This should have an “average” meta entity that you can alert on, and ignore the other ones.

It’s marked as unix group, and has HOST-RESOURCES-MIB, which generates the average entity.

Adam.

*From:*observium observium-bounces@observium.org *On Behalf Of *Daniel Johansson via observium *Sent:* 12 December 2019 14:34 *To:* Adam Armstrong via observium observium@observium.org *Cc:* Daniel Johansson daz@voodoo-people.com *Subject:* Re: [Observium] High CPU alarm

It's a Synology nas and their OS is Synology DSM 6.2-24922. It's a linux based OS.

In the poller script it says: o OS dsm o OS Group unix

Regards Daniel

On 2019-12-12 15:15, Adam Armstrong via observium wrote:
Does this device not generate the "average" meta-entity?

We generate this on some device types for exactly this reason.

What is the OS-type?

adam.

    On 2019-12-12 13:14:24, Daniel Johansson via observium
    <observium@observium.org> <mailto:observium@observium.org> wrote:

    Hi

    I have a synology nas with 1CPU and 4 cores in it. Whenever 1
    core gets a shitload to do I get a CPU high alarm although
    it's not the combined cpu load that is high.



    Is there something I can do in the alarm checker to combine
    the cpu loads of all cores and trigger an alarm when all 4
    combined is over for example 80% load?

    Regards
    Daniel



_______________________________________________

observium mailing list

observium@observium.org  <mailto:observium@observium.org>

http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

Adam Armstrong

6 p.m.

For all "unix" and windows systems you match the processor_descr "Average".

It's a good idea to set a delay on processor alerts too.

adam. On 2019-12-12 16:58:44, Daniel Johansson via observium observium@observium.org wrote: Hi

So what meta entry should I create the alert on?

Regards Daniel

On 2019-12-12 17:41, adama--- via observium wrote:

This should have an “average” meta entity that you can alert on, and ignore the other ones. It’s marked as unix group, and has HOST-RESOURCES-MIB, which generates the average entity. Adam. From: observium observium-bounces@observium.org [mailto:observium-bounces@observium.org] On Behalf Of Daniel Johansson via observium Sent: 12 December 2019 14:34 To: Adam Armstrong via observium observium@observium.org [mailto:observium@observium.org] Cc: Daniel Johansson daz@voodoo-people.com [mailto:daz@voodoo-people.com] Subject: Re: [Observium] High CPU alarm It's a Synology nas and their OS is Synology DSM 6.2-24922. It's a linux based OS.

In the poller script it says: o OS dsm o OS Group unix

Regards Daniel

On 2019-12-12 15:15, Adam Armstrong via observium wrote: Does this device not generate the "average" meta-entity? We generate this on some device types for exactly this reason. What is the OS-type? adam. On 2019-12-12 13:14:24, Daniel Johansson via observium observium@observium.org [mailto:observium@observium.org] wrote: Hi

I have a synology nas with 1CPU and 4 cores in it. Whenever 1 core gets a shitload to do I get a CPU high alarm although it's not the combined cpu load that is high.

Is there something I can do in the alarm checker to combine the cpu loads of all cores and trigger an alarm when all 4 combined is over for example 80% load?

Regards Daniel

_______________________________________________ observium mailing list observium@observium.org [mailto:observium@observium.org] http://postman.memetic.org/cgi-bin/mailman/listinfo/observium [http://postman.memetic.org/cgi-bin/mailman/listinfo/observium]