topic Re: Panel that displays pods with failure rate more than 50% in Dashboarding

Panel that displays pods with failure rate more than 50%

srikanthd — Wed, 24 Jan 2024 11:32:49 GMT

- i am trying to make a panel to observe the pods which are having failure rate high.

i can able to filter on pod_status but not on the pods failure rate.

Can anyone suggest me what should be used as filter condition to achieve this.

Re: Panel that displays pods with failure rate more than 50%

ChadTurner — Wed, 27 Mar 2024 18:35:37 GMT

I would do something like this:

You can further refine it as needed

Re: Panel that displays pods with failure rate more than 50%

PierreGutierrez — Tue, 30 Apr 2024 03:55:14 GMT

Hi @srikanthd

I think that exist a confused. The pods don't have a metric like a failure rate.
The metric of "failure rate" it's in the service. Pods and Services have a relation, but both are not the same.
In One Pod, you can find different process, and that process can give you a traffic of services.

Pods have a information like : Status, Age, IP address, containers, Total container restarts.
Pods have a metrics like : CPU, Memory .
Containers in Pod have metrics like: CPU (usage, shares, % of limit), Memory (cache, physical total, % of limit), etc.
Services have metrics like : response time, failure rate, CPU, throutput

I Recommend you identify the services in your process group, and then use the metric : builtin:service.errors.server.rate , for the failure rate .

I hope it's helpfull 💪