Closed RanHuang closed 1 year ago
I'm curious how/when badness is not accurate.
In your screenshots it seems fine:
I'm curious how/when badness is not accurate.
In your screenshots it seems fine:
- gunicorn ... badness 1004, VmRSS 129 MiB
- java ........... badness 1040, VmRSS 3323 MiB
e.g: i want the process java... badness 73, VmRSS 2359MiB to be killed before gunicorn ... badness 1004, VmRSS 129 MiB
Problem: --avoid / --prefer don't work with --sort-by-rss
Is there any issue or concern regarding this request?
No, code is ok but I need to find time to test it manually
On Wed, 11 Oct 2023, 11:32 Ran Huang, @.***> wrote:
Is there any issue or concern regarding this request?
— Reply to this email directly, view it on GitHub https://github.com/rfjakob/earlyoom/pull/300#issuecomment-1757261285, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACGA7YKK2ZTAHEAD3I242DX6ZRTLANCNFSM6AAAAAA45U3WB4 . You are receiving this because you commented.Message ID: @.***>
Merged, thanks!
By default, earlyoom identifies and terminates the process with the highest oom_score. However, in my Linux system, the oom_score is not an accurate indicator. Instead, I would like to identify and terminate the process with the highest memory usage.
find process with the largest oom_score
find process with the largest rss