[Bugs] [Bug 1732961] shard file with different gfid in different subvolume causing VM to pause on stale file

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1732961

billycole at mail.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |billycole at mail.com



--- Comment #27 from billycole at mail.com ---
Hello,

We're also seeing this issue with some frequency.  I've been unable to exactly
reproduce the steps, but can attest that it seems to be more prevalent when the
volume is closer to full.  Seemingly anywhere about 80% or 90% and the issue
appears.

Heres a log of our instance:

[2019-12-03 18:16:14.623264] E [MSGID: 109040]
[dht-helper.c:1365:dht_migration_complete_check_task] 0-scratch-dht:
c9b3d9cd-ffb2-4ea3-af8a-dff45151f86d: failed to lookup the file on scratch-dht
[Stale file handle]
[2019-12-03 18:16:14.626687] W [fuse-bridge.c:2830:fuse_readv_cbk]
0-glusterfs-fuse: 189341727: READ => -1
gfid=4068dcae-a9d1-4429-aa86-621f9a242fd1 fd=0x7fe552460b88 (Stale file handle)
[2019-12-03 18:16:14.639266] W [fuse-bridge.c:2830:fuse_readv_cbk]
0-glusterfs-fuse: 189341728: READ => -1
gfid=4068dcae-a9d1-4429-aa86-621f9a242fd1 fd=0x7fe552460b88 (Stale file handle)
[2019-12-03 18:16:14.639405] W [fuse-bridge.c:2830:fuse_readv_cbk]
0-glusterfs-fuse: 189341729: READ => -1
gfid=4068dcae-a9d1-4429-aa86-621f9a242fd1 fd=0x7fe552460b88 (Stale file handle)
[2019-12-03 18:16:14.640918] W [fuse-bridge.c:2830:fuse_readv_cbk]
0-glusterfs-fuse: 189341741: READ => -1
gfid=4068dcae-a9d1-4429-aa86-621f9a242fd1 fd=0x7fe552460b88 (Stale file handle)
[2019-12-03 18:16:14.642831] W [fuse-bridge.c:2830:fuse_readv_cbk]
0-glusterfs-fuse: 189341730: READ => -1
gfid=4068dcae-a9d1-4429-aa86-621f9a242fd1 fd=0x7fe552460b88 (Stale file handle)
[2019-12-03 18:16:18.023463] W [fuse-bridge.c:2830:fuse_readv_cbk]
0-glusterfs-fuse: 189342328: READ => -1
gfid=4068dcae-a9d1-4429-aa86-621f9a242fd1 fd=0x7fe552460b88 (Stale file handle)

We're running Gluster 6.5 on Amazon Linux 2 which is a CentOS 7 variant.

Our shard size is set to 512MB if that matters, and we have a 3x2 setup for
this volume, totaling 14TB.

Thanks!
Tim

-- 
You are receiving this mail because:
You are on the CC list for the bug.