[LU-7798] ll_prep_inode()) ASSERTION( fid_is_sane(&md.body->mbo_fid1) ) failed:

• Type: Bug
• Resolution: Duplicate
• Priority: Critical
• Labels: soak
• Environment: lola; build: 2.8.50-6-gf9ca359 ;commit f9ca359284357d145819beb08b316e932f7a3060
• Severity: 3
• Assignee: WC Triage
• Reporter: Frank Heckes (Inactive)

Description

Error happened during soak testing of build '20160218' (see: https://wiki.hpdd.intel.com/display/Releases/Soak+Testing+on+Lola#SoakTestingonLola-20160218). DNE is enabled.
MDT's have been formated using ldiskfs, OSTs using zfs.

Event history:

• 2016-02-18 16:24:27,115:fsmgmt.fsmgmt:INFO reseting MDS node lola-10
• 2016-02-18-16:34:04] Lustre client lola-26 crash with LBUG
• 2016-02-18-16:34:06 Lustre client lola-29 crash with LBUG

Error message and stack trace is same for both events:

<6>Lustre: soaked-MDT0004-mdc-ffff880852d1a000: Connection restored to 192.168.1.110@o2ib10 (at 192.168.1.110@o2ib10)
<0>LustreError: 23705:0:(llite_lib.c:2295:ll_prep_inode()) ASSERTION( fid_is_sane(&md.body->mbo_fid1) ) failed:
<0>LustreError: 23705:0:(llite_lib.c:2295:ll_prep_inode()) LBUG
<4>Pid: 23705, comm: pct
<4>
<4>Call Trace:
<4> [<ffffffffa050b875>] libcfs_debug_dumpstack+0x55/0x80 [libcfs]
<4> [<ffffffffa050be77>] lbug_with_loc+0x47/0xb0 [libcfs]
<4> [<ffffffffa0aa0192>] ll_prep_inode+0x752/0xc40 [lustre]
<4> [<ffffffffa07f6d60>] ? lustre_swab_mdt_body+0x0/0x130 [ptlrpc]
<4> [<ffffffffa0ab42b2>] ll_new_node+0x682/0x7f0 [lustre]
<4> [<ffffffffa0ab6b04>] ll_mkdir+0x104/0x220 [lustre]
<4> [<ffffffff8122ec0f>] ? security_inode_permission+0x1f/0x30
<4> [<ffffffff8119d759>] vfs_mkdir+0xd9/0x140
<4> [<ffffffff811a04e7>] sys_mkdirat+0xc7/0x1b0
<4> [<ffffffff8100c6f5>] ? math_state_restore+0x45/0x60
<4> [<ffffffff811a05e8>] sys_mkdir+0x18/0x20
<4> [<ffffffff8100b0d2>] system_call_fastpath+0x16/0x1b
<4>
<0>Kernel panic - not syncing: LBUG
<4>Pid: 23705, comm: pct Not tainted 2.6.32-504.30.3.el6.x86_64 #1
<4>Call Trace:
<4> [<ffffffff815293fc>] ? panic+0xa7/0x16f
<4> [<ffffffffa050becb>] ? lbug_with_loc+0x9b/0xb0 [libcfs]
<4> [<ffffffffa0aa0192>] ? ll_prep_inode+0x752/0xc40 [lustre]
<4> [<ffffffffa07f6d60>] ? lustre_swab_mdt_body+0x0/0x130 [ptlrpc]
<4> [<ffffffffa0ab42b2>] ? ll_new_node+0x682/0x7f0 [lustre]
<4> [<ffffffffa0ab6b04>] ? ll_mkdir+0x104/0x220 [lustre]
<4> [<ffffffff8122ec0f>] ? security_inode_permission+0x1f/0x30
<4> [<ffffffff8119d759>] ? vfs_mkdir+0xd9/0x140
<4> [<ffffffff811a04e7>] ? sys_mkdirat+0xc7/0x1b0
<4> [<ffffffff8100c6f5>] ? math_state_restore+0x45/0x60
<4> [<ffffffff811a05e8>] ? sys_mkdir+0x18/0x20
<4> [<ffffffff8100b0d2>] ? system_call_fastpath+0x16/0x1b

• No errors on Lustre server nodes ca be be correlated to the events.

Attached console, messages and vmcore-dmesg.txt from both clients.
Crash files are available

Issue Links

• duplicates: LU-7422 incorrect ENOENT handling in mdt_intent_reint causes panic on client side
• is related to: LU-7797 Can't mount zpools after OSS restart