-
-
Notifications
You must be signed in to change notification settings - Fork 4.3k
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
[Bug]: In v0.6.0 and above, Some of monitoring metrics are not correct. #8178
Comments
And when I activate --num-scheduler-steps option, vllm:time_per_output_token_seconds metric turns into per event latency, not TPOT. In v0.5.5, it was well represented as TPOT. |
@youkaichao @robertgshaw2-neuralmagic @njhill but when I activate --num-scheduler-steps option, vllm:time_per_output_token_seconds metric turns into per event latency, not TPOT. |
@ashgold - Thanks for reporting. We are still working on har@dening |
In v0.6.1.post1, when --num-scheduler-steps=8, the following metric values become strange. |
@youkaichao @robertgshaw2-neuralmagic @njhill @robertgshaw2-neuralmagic |
Your current environment
The output of `python collect_env.py`
🐛 Describe the bug
If I call /metrics in v0.6.0, only the following metrics exist.
Other important metrics that were visible in v0.5.5 have disappeared, except for the ones shown above.
Before submitting a new issue...
The text was updated successfully, but these errors were encountered: