[bug?] can't unmount filesystem after running "exportfs -u" -- dangling refcount?

[Chris Friesen]

I'm seeing some odd behaviour on an NFS server.

We're running 3.4.82 and have a drbd-replicated filesystem 
(active/standby) that is exported via NFS (explicitly v3, and 
/proc/fs/nfsd is mounted).

On a server switchover we take down the NFS server IP address, run 
"exportfs -u" on the exports, unmount the drbd filesystem, and then 
demote drbd from primary to secondary.

The problem we're seeing is that sometimes "exportfs -u" completes 
successfully but "umount" fails with an error that the filesystem is 
busy.  When this happens, "fuser" doesn't see any userspace users, which 
points at the kernel thinking it has a reference open.

When it's in the stuck state, manually running "exportfs -f" seems to 
resolve the issue so that "unmount" can run.

Are there any known issue in this area?

I'm not currently subscribed to the list, so please CC me on replies.

Thanks,
Chris
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/