vllm-project / vllm-project/aibrix
engine prefix should be provided by user and we should not amend `vllm:`
Open
@Jeffwan is already working on this.
Since Sep 30, 2025.
area/autoscaling
- Dominant language
- Go
- Stars
- 5.1k
- Forks
- 697
- Avg merge
- 1d 19h
- Merged PRs (30d)
- 104
Description
🐛 Describe the bug
W0930 18:13:41.031286 1 fetcher.go:99] Failed to fetch metric vllm:gpu_cache_usage_perc from pod default/mock-llama2-7b-7cc98b7f5f-764t4: metric vllm:gpu_cache_usage_perc not found in central registry. Returning zero value.
Steps to Reproduce
metricsSources:
- metricSourceType: pod
protocolType: http
port: "8000"
path: metrics
targetMetric: "avg_prompt_throughput_toks_per_s" # change it to `vllm:avg_prompt_throughput_toks_per_s`
targetValue: "60"
scalingStrategy: "KPA"
Expected behavior
it should work
Environment
nightly
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.