[LU-3672] See we have it LU-2943 opening case for tracking

• Status: Resolved
• Project: Lustre
• Affects Version/s: Lustre 2.1.4
• Type: Bug
• Priority: Critical
• Reporter: Mahmoud Hanafi
• Assignee: Hongchao Zhang
• Resolution: Duplicate
• Severity: 2

Description

<0>LustreError: 6211:0:(mdt_open.c:1056:mdt_reconstruct_open()) ASSERTION( (!(rc < 0) || (lustre_msg_get_transno(req->rq_repmsg) == 0)) ) failed: ^M
<0>LustreError: 6211:0:(mdt_open.c:1056:mdt_reconstruct_open()) LBUG^M
<4>Pid: 6211, comm: mdt_18^M
<4>^M
<4>Call Trace:^M
<4> [<ffffffffa0596785>] libcfs_debug_dumpstack+0x55/0x80 [libcfs]^M
<4> [<ffffffffa0596d97>] lbug_with_loc+0x47/0xb0 [libcfs]^M
<4> [<ffffffffa0d5c697>] mdt_reconstruct_open+0x7c7/0xa80 [mdt]^M
<4> [<ffffffffa0d4e1f5>] mdt_reconstruct+0x45/0x120 [mdt]^M
<4> [<ffffffffa0d3b099>] mdt_reint_internal+0x709/0x8e0 [mdt]^M
<4> [<ffffffffa0d3b53d>] mdt_intent_reint+0x1ed/0x500 [mdt]^M
<4> [<ffffffffa0d39c09>] mdt_intent_policy+0x379/0x690 [mdt]^M
<4> [<ffffffffa07e9351>] ldlm_lock_enqueue+0x361/0x8f0 [ptlrpc]^M
<4> [<ffffffffa080f1ad>] ldlm_handle_enqueue0+0x48d/0xf50 [ptlrpc]^M
<4> [<ffffffffa0d3a586>] mdt_enqueue+0x46/0x130 [mdt]^M
<4> [<ffffffffa0d2f772>] mdt_handle_common+0x932/0x1750 [mdt]^M
<4> [<ffffffffa0d30665>] mdt_regular_handle+0x15/0x20 [mdt]^M
<4> [<ffffffffa083db4e>] ptlrpc_main+0xc4e/0x1a40 [ptlrpc]^M
<4> [<ffffffffa083cf00>] ? ptlrpc_main+0x0/0x1a40 [ptlrpc]^M
<4> [<ffffffff8100c0ca>] child_rip+0xa/0x20^M
<4> [<ffffffffa083cf00>] ? ptlrpc_main+0x0/0x1a40 [ptlrpc]^M
<4> [<ffffffffa083cf00>] ? ptlrpc_main+0x0/0x1a40 [ptlrpc]^M
<4> [<ffffffff8100c0c0>] ? child_rip+0x0/0x20^M
<4>^M
<0>Kernel panic - not syncing: LBUG^M

Comments

[Peter Jones]

Hongchao

Bruno is out this week so could you please confirm whether this is indeed a duplicate of LU-2943?

Thanks

Peter

[Jay Lan (Inactive)]

I confirm this is a dup of LU-2943.

Well, I actually have cherry-picked the 2.1 patch in LU-2943 some time ago on top of 2.1.5 release. Two of our lustre file systems are running that version. This case can be closed.

It would be nice Intel completes the review cycle of the 2.1 patch from LU-2943 and lands the patch to b2_1 tree though.

[Hongchao Zhang]

Thanks Jay!
Yes, this should be a duplicate of LU-2943.

[Jay Lan (Inactive)]

We hit this again!

[Peter Jones]

duplicate of LU-2943