[LU-4272] lu_device_fini()) ASSERTION( cfs_atomic_read(&d->ld_ref) == 0 ) failed from lovsub_device_free

• Type: Bug
• Resolution: Duplicate
• Priority: Critical
• Fix Version/s: Lustre 2.8.0
• Affects Version/s: Lustre 2.6.0
• Labels: nfs
• Severity: 3
• Assignee: Niu Yawei (Inactive)
• Reporter: Oleg Drokin

Description

Apparently we have a problem with inode cleaning on unmount exposed by at least nfs.

The easiest reproducer I have right now is this:

sh llmount.sh
echo 0 >/proc/sys/lnet/panic_on_lbug
service nfs start
mount localhost:/mnt/lustre /mnt/nfs -t nfs
touch /mnt/nfs/file
ls -l /mnt/lustre
cp -f /etc/passwd /mnt/nfs/file

umount /mnt/nfs
service nfs stop
sh llmountcleanup.sh

Reading the logs, It appears that after the write file dentry is deleted, but because we have locks on inode - it stays in place (unlike when we do not use nfs where it also goes away after sync).
Then on unmount file inode is found, but is considered dirty and not cleaned, meaning the layout lock and objects remain in place.
Now we get to kill_super and that blows up trying to clean up lov.

Issue Links

• is related to: LU-6794 memory leak in Lustre NFS support code, LASSERT() at unmount
• is related to: LU-2613 opening and closing file can generate 'unreclaimable slab' space

Comments

[Oleg Drokin]

it was fixed by LU-6794 which is essentially the same patch.

[Joseph Gmitter (Inactive)]

Oleg is testing patch from LU-6794 to see if it resolves this issue. He will close this as a duplicate if it is resolved.

[Andreas Dilger]

Oleg, if you are no longer able to reproduce this, please close it with "Cannot Reproduce" to get it off the tracking list.

[Jodi Levi (Inactive)]

Oleg,
Is this problem still occurring? If so, would you be able to provide the logs for Niu?
Thank you!

[Niu Yawei (Inactive)]

It's strange, I tried again, but it still can't be reproduced in my test environment. Oleg, could you post the log here for analysis? Thank you.

[Oleg Drokin]

The problem still exists, I just tried all the steps on current master and it still fails in the exact same way

[Jodi Levi (Inactive)]

Has this been reproduced? Does the problem still exist or should this ticket be closed?

[Niu Yawei (Inactive)]

I can't reproduce it with current master. Oleg, did you test it with clean master?

[Andreas Dilger]

This might be a side effect of the LU-2613 patch landing.

[Peter Jones]

Niu

Is this related to your patch for LU-2613?

Peter