You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
**Describe something that is unclear or not documented and needs a documentation update
Due to research needs, I now want to use Prometheus to collect fission's p99 tail delay data. I have correctly deployed fission (K8s helm mode) and configured Prometheus and grafana. However, currently I do not know how to use PromQL query. I want to know how the tail latency indicator is defined. Is it the 99th percentile of the process from gateway request to execution completion, or some other value, and how do we count it?
can you give me some advice?
I checked the fission documentation, but it’s not very clear about fission_function_overhead_seconds and fission_function_running_seconds.
The text was updated successfully, but these errors were encountered:
**Describe something that is unclear or not documented and needs a documentation update
Due to research needs, I now want to use Prometheus to collect fission's p99 tail delay data. I have correctly deployed fission (K8s helm mode) and configured Prometheus and grafana. However, currently I do not know how to use PromQL query. I want to know how the tail latency indicator is defined. Is it the 99th percentile of the process from gateway request to execution completion, or some other value, and how do we count it?
can you give me some advice?
I checked the fission documentation, but it’s not very clear about fission_function_overhead_seconds and fission_function_running_seconds.
The text was updated successfully, but these errors were encountered: