Red Hat Bugzilla – Bug 1160641
[USS]: After snapshot restore , if user is under snap directory which was restored, ls on the snap directory hangs
Last modified: 2016-09-17 09:04:14 EDT
Description of problem: ========================= After restoring volume to a particular snapshot, if user is under .snaps/snap-directory(which was restored), ls under the snap directory hangs, instead it should give appropriate error message Version-Release number of selected component (if applicable): ============================================================= glusterfs 3.6.0.30 How reproducible: ================= 2/2 Steps to Reproduce: =================== 1.Create a 2x2 dist rep volume and start it 2.Fuse and NFS mount the volume, enable USS and create some IO 3.Take a snapshot(snap1) 4. Create some more IO and take few more snapshots(snap2, snap3) 5. cd to .snaps and cd to snap2 and list the files under them [root@dhcp-0-97 vol0_fuse]# cd dir2 [root@dhcp-0-97 dir2]# cd .snaps [root@dhcp-0-97 .snaps]# ls snap1 snap2 snap3 [root@dhcp-0-97 .snaps]# cd snap2 [root@dhcp-0-97 snap2]# ll total 10240 -rw-r--r--. 1 root root 1048576 Nov 5 14:18 fuse11 -rw-r--r--. 1 root root 1048576 Nov 5 14:18 fuse12 -rw-r--r--. 1 root root 1048576 Nov 5 14:18 fuse13 -rw-r--r--. 1 root root 1048576 Nov 5 14:18 fuse14 -rw-r--r--. 1 root root 1048576 Nov 5 14:18 fuse15 -rw-r--r--. 1 root root 1048576 Nov 5 14:18 fuse16 -rw-r--r--. 1 root root 1048576 Nov 5 14:18 fuse17 -rw-r--r--. 1 root root 1048576 Nov 5 14:18 fuse18 -rw-r--r--. 1 root root 1048576 Nov 5 14:18 fuse19 -rw-r--r--. 1 root root 1048576 Nov 5 14:18 fuse20 [root@dhcp-0-97 snap2]# pwd /mnt/vol0_fuse/dir2/.snaps/snap2 Remain in this path. 6.Stop the volume and restore it to snap2 7. Go back to the terminal in step5 and from the same path do a ls, it hangs . Instead it should give appropriate error message Actual results: =============== After snapshot restore , if user is under snap directory which was restored, ls on the snap directory hangs Expected results: ================ Appropriate error message should be displayed. Additional info:
Version :glusterfs 3.6.0.33 ======= Another scenario where accessing .snaps hangs : ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -Create a 2x2 dist-rep volume -Fuse and NFS mount the volume, create some dir and files, -Create few snapshots. -Restore a volume to a deactivated snapshot- Successful. -From fuse and nfs mount, cd to .snaps from the dirs created- it hangs [root@dhcp-0-97 d2_n]# cd .snaps
Version :glusterfs 3.6.0.33 ======= Another scenario where accessing .snaps hangs : ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - Restore volume to a activated snapshot while glusterd is down on one node - it is Successful - Start the volume - Start glusterd on the node which was down - cd to .snaps from fuse and nfs mount- it hangs
I tried reproducing the issue mentioned in description of the bug and also comment-3. But I did not hit the hang issue mentioned. I'll try reproducing using steps mentioned in comment-4.
I tried to reproduce this issue as mentioned in the comments, but I could not hit the issue. Can you please have a look into this.
Patch https://code.engineering.redhat.com/gerrit/#/c/37398/ has fixed this issue
Version : glusterfs 3.6.0.38 ======== Retried the steps as mentioned in Description and Comment 3 , works fine. No hang issue seen Scenario in Comment 4 is tracked by bz 1163209 Marking this bug as 'Verified'
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHBA-2015-0038.html