[LU-15278] ASSERTION( equi(!strcmp(name, XATTR_LUSTRE_LOV) || !strcmp(name, XATTR_NAME_LOV), !lod_dt_obj(dt)->ldo_comp_cached))) Created: 25/Nov/21  Updated: 04/Nov/22

Status: Open
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Major
Reporter: Zhenyu Xu Assignee: Zhenyu Xu
Resolution: Unresolved Votes: 0
Labels: None

Issue Links:
Related
Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   
[341197.549203] LustreError: 27243:0:(lod_object.c:4483:lod_xattr_set()) ASSERTION( (!!(!strcmp(name, "lustre.""lov") || !strcmp(name, "trusted.lov")) == !!(!lod_dt_obj(dt)->ldo_comp_cached)) ) failed: 
[341197.553942] LustreError: 27243:0:(lod_object.c:4483:lod_xattr_set()) LBUG
[341197.556044] Pid: 27243, comm: mdt_rdpg00_041 3.10.0-1160.31.1.el7_lustre.ddn15.x86_64 #1 SMP Fri Jul 2 13:54:06 PDT 2021
[341197.556047] Call Trace:
[341197.556110] [<0>] libcfs_call_trace+0x90/0xf0 [libcfs]
[341197.556118] [<0>] lbug_with_loc+0x4c/0xa0 [libcfs]
[341197.556157] [<0>] lod_xattr_set+0xcd9/0xd40 [lod]
[341197.556196] [<0>] mdo_xattr_set+0x75/0x190 [mdd]
[341197.556210] [<0>] mdd_xattr_set+0x159f/0x18c0 [mdd]
[341197.556273] [<0>] mo_xattr_set+0x46/0x48 [mdt]
[341197.556293] [<0>] mdt_close_handle_layouts+0x8b0/0xc10 [mdt]
[341197.556309] [<0>] mdt_mfd_close+0x543/0x870 [mdt]
[341197.556325] [<0>] mdt_close_internal+0x121/0x220 [mdt]
[341197.556340] [<0>] mdt_close+0x25b/0x7d0 [mdt]
[341197.556590] [<0>] tgt_request_handle+0xaee/0x15f0 [ptlrpc]
[341197.556643] [<0>] ptlrpc_server_handle_request+0x24b/0xab0 [ptlrpc]
[341197.556696] [<0>] ptlrpc_main+0xb34/0x1470 [ptlrpc]
[341197.556702] [<0>] kthread+0xd1/0xe0
[341197.556707] [<0>] ret_from_fork_nospec_begin+0x7/0x21
[341197.556847] [<0>] 0xfffffffffffffffe
[341197.556857] Kernel panic - not syncing: LBUG
Reading symbols from /root/work/lustre/lustre/mdd/mdd.ko...done.
(gdb) l *(mdd_xattr_set+0x159f)
0x2f3cf is in mdd_xattr_set (/root/work/lustre/lustre/mdd/mdd_object.c:1619).
1614		if (rc != 0)
1615			GOTO(out, rc);
1616	
1617		rc = mdo_xattr_set(env, obj, buf_vic, XATTR_NAME_LOV, LU_XATTR_MERGE,
1618				   handle);
1619		if (rc)
1620			GOTO(out, rc);
1621	
1622		rc = mdo_xattr_del(env, vic, XATTR_NAME_LOV, handle);
1623		if (rc)

It shows that a mirror extend prepared a merged layout stored in lod_object in declare phase, but in the transaction execution phase the in-memory layout has been changed by others.



 Comments   
Comment by Gerrit Updater [ 26/Nov/21 ]

"Bobi Jam <bobijam@hotmail.com>" uploaded a new patch: https://review.whamcloud.com/45663
Subject: LU-15278 lod: protect LOD layout cache in layout change
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: a5810932551d7734c950e38ef4cb462e5c760896

Comment by Gerrit Updater [ 02/Dec/21 ]

This is an important patch, it fixes a issue that could mess up lod_object memory

"Bobi Jam <bobijam@hotmail.com>" uploaded a new patch: https://review.whamcloud.com/45710
Subject: LU-15278 lod: distinguish DIR/REGULAR lod_object members
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 4f4b0dfc1fc0a198e9d00ae099adb1bef0ff5e45

Comment by Gerrit Updater [ 24/Dec/21 ]

"Bobi Jam <bobijam@hotmail.com>" uploaded a new patch: https://review.whamcloud.com/45939
Subject: LU-15278 debug: try to understand the lod device ref leak
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: b1e02ba3ffaa6834cdf52e4a6eaa6e41665f26b0

Comment by Gerrit Updater [ 10/Feb/22 ]

"Bobi Jam <bobijam@hotmail.com>" uploaded a new patch: https://review.whamcloud.com/46491
Subject: LU-15278 mdt: protect layout xattr
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 0c2b866ded8f4e815371cf77a9ddf6eef7bb3998

Comment by Gerrit Updater [ 30/May/22 ]

"Oleg Drokin <green@whamcloud.com>" merged in patch https://review.whamcloud.com/45710/
Subject: LU-15278 lod: distinguish DIR/REGULAR lod_object members
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 7a9c9ccabe93f2d96c80e90f8cbb786faca74835

Generated at Sat Feb 10 03:16:57 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.