Hi Ed,

We found this bug:

https://www.findbugzero.com/operational-defect-database/vendors/rh/defects/RHEL-33796

https://issues.redhat.com/browse/RHEL-33796

After updating the packages autofs and sssd to versions: autofs-5.1.7-58.el9.x86_64 and sssd-2.9.4-6.el9_4.x86_64 respectively, it does not happen again for us. We are still monitoring the behaviour but we have at least one month without seeing the error. 

Cheers

Carles

Missatge de Ed Rolison <ed.rolison@gmail.com> del dia dc., 11 de set. 2024 a les 3:54:
We've just hit this issue on some of our alma 9.3 hosts The triage was to look at the number of file descriptors that automount had open.

ls -1 /proc/$(pgrep automount)/fd | wc -l

Most clients are well below 100, but a few are in the thousands, and when they hit the default soft limit of 20480 (cat /proc/($pgrep automount)/limits) autofs breaks - starting with too many open files.

systemctl restart autofs resolves it, and seems to 'clear' the problem on any clients straying up past 1000, but it seems there's got to be some sort of file descriptor 'leaking' occurring here.
_______________________________________________
AlmaLinux Users mailing list -- users@lists.almalinux.org
To unsubscribe send an email to users-leave@lists.almalinux.org