-
Notifications
You must be signed in to change notification settings - Fork 2.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
settle on container metrics naming #26663
Comments
This sounds related to the Prometheus mapping cc @open-telemetry/wg-prometheus |
We are really sorry about the disruption. We have no naming changes planned for the future (this is the last set to be compliant with the specification. The initial change to metric names was made prematurely, and had to be reverted, hence the unfortunate back-and-forth. Here are the changelog entries:
Feel free to reopen if you have further questions. |
Thanks a lot @dashpole and sorry I missed this in the change log. |
Describe the bug
In the last few weeks container metrics got renamed twice first to container_cpu_usage_total then back to container_cpu_usage_nanoseconds_total
Same for memory which lost then got back a _bytes suffix.
Steps to reproduce
Start a container with otel/opentelemetry-collector-contrib image and enable docker-stats in it's config
Start a watchtower container to keep the image up to date on a daily basis
Observe your grafana graphs being broken regularly as metrics seem to disappear as they get renamed
What did you expect to see?
A warning in the release notes?
What did you see instead?
Empty graphs on grafana
What version did you use?
Latest from docker hub
What config did you use?
Enable docker-stats
Environment
OS: Ubuntu 22.04
It works just fine with minimal performance impact, so very good but I believe stability in naming would be great. I'm all for removing bytes and nanoseconds which seem redondant or can be explained in the help section
The text was updated successfully, but these errors were encountered: