-
Notifications
You must be signed in to change notification settings - Fork 9.1k
YARN-11813. Fix the fallback ordering between cgroup v2 and v1. #7631
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
...he/hadoop/yarn/server/nodemanager/containermanager/linux/resources/CGroupsV2HandlerImpl.java
Outdated
Show resolved
Hide resolved
Thanks @brumi1024 ! For finding and fixing this case. LGTM! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@brumi1024 Thanks for investigating and fixing this issue! LGTM
💔 -1 overall
This message was automatically generated. |
💔 -1 overall
This message was automatically generated. |
💔 -1 overall
This message was automatically generated. |
Description of PR
The fallback ordering between cgroup v2 and v1 handlers are now changed, so that now if cgroup v2 is enabled it'll be used first, and if the controller is not mounted in v2 YARN will try to search for it in a v1 hierarchy. This needed some adjustments in the init procedure of the v2 handlers when preconfigured mount paths are used, but the end result should be similar.
How was this patch tested?
Deployed to a cluster, and tested cgroup v1/v2 functionality, by specifically recreating the error scenario described in the ticket.
For code changes:
LICENSE
,LICENSE-binary
,NOTICE-binary
files?