Skip to content
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

"data" size is unreliable, start using RES, as it seems reliable and relevant #10161

Merged
merged 1 commit into from
Mar 10, 2021

Conversation

omoerbeek
Copy link
Member

According to https://www.kernel.org/doc/html/latest/filesystems/proc.html "data" aka "drs" is broken.
In practice most of the time it shows a number somewhere between RES and VIRT, but it seems to be completely off in multi-threaded processed once in a while.

So move to the same number top(1) displays as "RES". This number is not marked broken and top(1) and the metrics agreed all the time in my tests.

Should fix #7591

Short description

Checklist

I have:

  • read the CONTRIBUTING.md document
  • compiled this code
  • tested this code
  • included documentation (including possible behaviour changes)
  • documented the code
  • added or modified regression test(s)
  • added or modified unit test(s)
  • checked that this code was merged to master

@mnordhoff
Copy link
Contributor

The color of the bikeshed concerns me! RES fluctuates based on swap usage, right?

That's potentially good (excluding bits of memory that are never used), but also potentially bad ("oh good, my memory usage is going down" vs. "the server is overloaded and PowerDNS is being swapped to death").

@omoerbeek
Copy link
Member Author

The color of the bikeshed concerns me! RES fluctuates based on swap usage, right?

That's potentially good (excluding bits of memory that are never used), but also potentially bad ("oh good, my memory usage is going down" vs. "the server is overloaded and PowerDNS is being swapped to death").

Yes, RES is only one dimension of a complex equation. Making any evaluation and/or decision based on only such a one-dimension measure is unwise.

@omoerbeek omoerbeek merged commit 6f73d1e into PowerDNS:master Mar 10, 2021
@omoerbeek omoerbeek deleted the real-real-mem-usage branch March 10, 2021 15:06
@omoerbeek omoerbeek changed the title "data" size is unreliable, start using RES, as it seem reliable and relavant "data" size is unreliable, start using RES, as it seems reliable and relevant Mar 10, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

The memory statistics in 4.2.0 beta1 are wrong
3 participants