Request for improved details surrounding Envoy stats and metrics documentation
- Dominant language
- C++
- Stars
- 28.9k
- Forks
- 5.6k
- Avg merge
- 1d 22h
- Merged PRs (30d)
- 430
Description
*Title*: *Request for improved details surrounding stats and metrics documentation*
*Description*:
When customers are debugging issues such as connection resets failures (Contour/Envoy) deployment there have been requests for more in depth descriptions for metrics and statistics that are available in Envoy and documented on the Envoy page.
One example of this is the “upstream_cx_destroy_with_active_rq” metric. It is defined as "Total connections destroyed with 1+ active request". Also found an older issue where a similar request was made, specific to this metric, https://github.com/envoyproxy/envoy/issues/6455
A second example is the "upstream cx_open" stat, but it is not clear how connections are managed, when are they closed when are the reused, etc.
This request is general but provides two examples.
[optional *Relevant Links*:]
https://www.envoyproxy.io/docs/envoy/latest/configuration/upstream/cluster_manager/cluster_stats
https://www.envoyproxy.io/docs/envoy/v1.15.0/configuration/http/http_conn_man/stats#config-http-conn-man-stats
Contributor guide
Assessment
This issue has not been assessed yet.