You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Rupesh pointed out that the crawler is unable to crawl cpu / memory for containers created using k8 (works for containers created directly using docker).
Turns out cgroupfs location for containers is different in both cases.
While crawler expects them to be at /sys/fs/cgroup/memory/docker/container-id, k8 puts them in /sys/fs/cgroup/memory/kubepods/burstable/pod-id/container-id or /sys/fs/cgroup/memory/kubepods/besteffort/pod-id/container-id.
Fix is to update the search path for the crawler for the containers' cgroupfs, but this goes back into the bigger discussion of making crawler k8-aware.
Description
Rupesh pointed out that the crawler is unable to crawl cpu / memory for containers created using k8 (works for containers created directly using docker).
Turns out cgroupfs location for containers is different in both cases.
While crawler expects them to be at
/sys/fs/cgroup/memory/docker/container-id
, k8 puts them in/sys/fs/cgroup/memory/kubepods/burstable/pod-id/container-id
or/sys/fs/cgroup/memory/kubepods/besteffort/pod-id/container-id
.Fix is to update the search path for the crawler for the containers' cgroupfs, but this goes back into the bigger discussion of making crawler k8-aware.
@nadgowdas @tatsuhirochiba
How to Reproduce
Log Output
Debugging Commands Output
Output of
docker version
:Output of
docker info
:Output of
python --version
:Output of
pip freeze
:The text was updated successfully, but these errors were encountered: