Fix cgroup helper init in add_processor_metadata and add_docker_metadata processors#41108
Conversation
|
Pinging @elastic/sec-linux-platform (Team:Security-Linux Platform) |
|
This pull request does not have a backport label.
To fixup this pull request, you need to add the backport labels for the needed
|
|
|
|
Pinging @elastic/elastic-agent-data-plane (Team:Elastic-Agent-Data-Plane) |
belimawr
left a comment
There was a problem hiding this comment.
Not my area of expertise, but LGTM.
I've seen the same CI error transaction_log_integration_test.go:22: timeout waiting for services to be healthy in one of my PRs, so I believe it's related to this one.
|
@fearful-symmetry Thank you so much! |
|
This PR seems to break MAC os development of beats if any of the metadata processors are enabled. In my setup, I get |
…tadata` processors (#41108) (#41185) * only initialize cgroup reader once * docs * remove old struct fields --------- Co-authored-by: Pierre HILBERT <pierre.hilbert@elastic.co> (cherry picked from commit 13d71a6) Co-authored-by: Alex K. <8418476+fearful-symmetry@users.noreply.github.com>
Proposed commit message
Closes #41065
This refactors
add_processor_metadataandadd_docker_metadataso we only callNewReaderonce when we create a new processor, and not every time the processor runs. This should considerably cut down on CPU cycles for certain workloads, sinceNewReaderhas gotten much more complex over the years.Checklist
CHANGELOG.next.asciidocorCHANGELOG-developer.next.asciidoc.