[Enhance] Too many log messages for memory stats when run out of memory #159
Labels
type: Bug
Something isn't working
Comments
|
Hey, @gfphoenix78 welcome!🎊 Thanks for taking the time to point this out.🙌 |
|
No reproducible steps provided. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Cloudberry Database version
1.3, perhaps all versions have this issue.
What happened
When a query causes out of memory, it may call
RunawayCleaner_RunawayCleanupDoneForProcess. This function will dump all memory stats from TopMemoryContext to its descendant. The log message may be huge, up to tens giga bytes. It looks dangerous that may cause out of disk for the cluster.What you think should happen instead
No response
How to reproduce
skip.
Operating System
centos
Anything else
No response
Are you willing to submit PR?
Code of Conduct
The text was updated successfully, but these errors were encountered: