Closed
Description
Name and Version
Current docker version of llamacpp server
Operating systems
Linux
Which llama.cpp modules do you know to be affected?
llama-server
Command line
docker run ghcr.io/ggml-org/llama.cpp:server --metrics
Problem description & steps to reproduce
The metrics initially added at #5708 are using the format
llamacpp:metric_name_goes_here
which is invalid per https://prometheus.io/docs/concepts/data_model/#metric-names-and-labels, in particular:
Note: The colons are reserved for user defined recording rules. They should not be used by exporters or direct instrumentation.
they should use the format
llamacpp_metric_name_goes_here
First Bad Commit
No response