0

Whenever my host hits 60% usage, OOM killer kicks in and kills one of my processes that has the highest oom score.

However, the memory on the host only hits 60%! Why is it kicking in so early? It is consistent and kills a process at 60% total usage every single time.

I have no swap on the host. Is there some sort of configuration that I need to change?

Rezzy
  • 1
  • Would you fly on this plane: "An aircraft company discovered that it was cheaper to fly its planes with less fuel on board. The planes would be lighter and use less fuel and money was saved. On rare occasions however the amount of fuel was insufficient, and the plane would crash. This problem was solved by the engineers of the company by the development of a special OOF (out-of-fuel) mechanism. In emergency cases a passenger was selected and thrown out of the plane. ..." Put enough fuel on the airplane... – Andrew Henle Oct 31 '23 at 12:15
  • 1
    OoM killer logs a kernel message why it gets triggered. Note Memory usage might spike more quickly than your monitoring can follow and the OoM-killer event occurs without a gradual ramp up of memory consumption. - Adding swap (even in the form of a swap file) might allow your system to deal more gracefully with applications requesting more memory than is physically available – HBruijn Oct 31 '23 at 12:19
  • 2
    Of interest might also be https://serverfault.com/q/606185 and adjust the systctl kernel tunables that manage memory (overcommit) – HBruijn Oct 31 '23 at 12:35
  • 1
    Please edit your question to add the kernel messages from your logs. "invoked oom-killer", "Tasks state", the kernel stack, and all the context around that. Also the contents of /proc/meminfo during a high mark in normal operation. See some of the other questions tagged oom-killer for examples, although most of them aren't providing enough detail either. – John Mahowald Oct 31 '23 at 20:13

0 Answers0