I’ve seen, of all things, an IDE periodically scanning homedirs. If that happens every five minutes ..
On Apr 10, 2026, at 3:16 PM, Paul Mezzanini via ceph-users <ceph-users@ceph.io> wrote:
Is there something in cron (or some other system monitoring) that tickles mounted drives on a 5 minute interval? We leverage autofs heavily here and I've never seen that except when netdata decided it wanted to monitor EVERY MOUNT to make sure it wasn't out of disk space.
--
Paul Mezzanini Platform Engineer III Research Computing Rochester Institute of Technology
________________________________________ From: Greg Ballantine via ceph-users <ceph-users@ceph.io> Sent: Friday, April 10, 2026 12:00 PM To: ceph-users@ceph.io Subject: [ceph-users] AutoFS Timeouts with CephFS
Hi all,
We are running into a strange issue with autofs and CephFS at UMBC. We are attempting to manage CephFS volumes via autofs with an idle timeout, however my volumes do not appear to be unmounting when idle. After some digging, it appears when the timeout is set to five minutes the volumes stay mounted indefinitely, but when the autofs timeout is set to 4.5 minutes, the volumes unmount as expected.
Our environment is:
-
Ceph cluster version: 18.2.7 -
Ceph client version: 18.2.1 kernel client -
OS: Red Hat 9.7 -
auto.master entry: "/umbc/rs /cm/shared/etc/autofs/mounts.rs --timeout=300" -
mounts.rs entry: "archive -fstype=ceph,name=archive,fs=rrstor ceph-mon.example.com:/archive"
We have checked for open file descriptors using lsof, but the mount appears idle, only showing automount processes as holding open file descriptors in the mounts. Has anyone encountered similar behavior before and can comment on what they think might be going on?
Thank you, Greg
-- Gregory Ballantine HPC Specialist, Division of IT University of Maryland, Baltimore County _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io