Skip to content
This repository was archived by the owner on Sep 2, 2025. It is now read-only.

Commit 97d255d

Browse files
Added a note that cgroup metrics exposed by cAdvisor cannot be collected by the Splunk OTel collector.
1 parent 2b2974a commit 97d255d

File tree

1 file changed

+4
-1
lines changed

1 file changed

+4
-1
lines changed

gdi/monitors-monitoring/cadvisor.rst

Lines changed: 4 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -16,8 +16,11 @@ If you are using Kubernetes, consider the
1616
expose cAdvisor on a network port, even though they are running it
1717
within Kubelet.
1818

19+
.. :note:: The Splunk Distribution of OpenTelemetry Collector cannot collect specific cgroup metrics exposed by cAdvisor, such as ``container_cpu_cfs_*``
20+
in managed environments such as Amazon EKS, and the supported `kubeletstats` receiver does not expose these metrics by default.
21+
1922
If you are running containers with Docker, retrieved metrics might
20-
overlap with ``docker-container-stats``\ '. Consider not enabling the
23+
overlap with ``docker-container-stats``. Consider not enabling the
2124
Docker monitor in a Kubernetes environment, or else use filtering to
2225
allow only certain metrics. This will cause the built-in Docker
2326
dashboards to be blank, but container metrics will be available on the

0 commit comments

Comments
 (0)