[slurm-users] Troubles with cgroups

[Hermann Schwärzler]

Hi everybody,

I would like to give you a quick update on this problem (hanging systems 
when swapping due to cgroup memory-limits is happening):

We had opened a case with RedHat's customer support. After some to and 
fro they could reproduce the problem. Last week they told us to upgrade 
to version 9.2 (kernel 5.14.0-284.11.1.el9_2.x86_64) which has been 
released a few days ago.
Doing so *did indeed fix the problem*!

The related bug was/is one in main-line kernels and its fix (from March 
2022) is discussed and described here:
https://lore.kernel.org/all/20220221111749.1928222-1-cgel.zte@gmail.com/

We asked support whether there are plans to backport this to 8.x 
versions of the system. The answer was: a BUG is raised and it is 
expected to be fixed in version 8.9 (release date not yet set).

Regards,
Hermann