]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
server : update help metrics processing/deferred (#11512)
authorDaniel Bevenius <redacted>
Fri, 31 Jan 2025 05:04:53 +0000 (06:04 +0100)
committerGitHub <redacted>
Fri, 31 Jan 2025 05:04:53 +0000 (06:04 +0100)
commita2df2787b32e0846205f7151dfad88ceab592beb
treea7fd1aae3db28725ab2f0979026e620bba40ab02
parent553f1e46e9e864514bbd6bf4009146db66be0541
server : update help metrics processing/deferred (#11512)

This commit updates the help text for the metrics `requests_processing`
and `requests_deferred` to be more grammatically correct.

Currently the returned metrics look like this:
```console
\# HELP llamacpp:requests_processing Number of request processing.
\# TYPE llamacpp:requests_processing gauge
llamacpp:requests_processing 0
\# HELP llamacpp:requests_deferred Number of request deferred.
\# TYPE llamacpp:requests_deferred gauge
llamacpp:requests_deferred 0
```

With this commit, the metrics will look like this:
```console
\# HELP llamacpp:requests_processing Number of requests processing.
\# TYPE llamacpp:requests_processing gauge
llamacpp:requests_processing 0
\# HELP llamacpp:requests_deferred Number of requests deferred.
\# TYPE llamacpp:requests_deferred gauge
llamacpp:requests_deferred 0
```
This is also consistent with the description of the metrics in the
server examples [README.md](https://github.com/ggerganov/llama.cpp/tree/master/examples/server#get-metrics-prometheus-compatible-metrics-exporter).
examples/server/server.cpp