[LU-10601] lov_stripe_lock()) ASSERTION( md->lsm_lock_owner == 0 ) failed

• Type: Bug
• Resolution: Unresolved
• Priority: Minor
• Affects Version/s: Lustre 2.11.0
• Severity: 3
• Assignee: WC Triage
• Reporter: Oleg Drokin

Description

Hit this on current master-next, some sort of use after free it appears.

[119243.473752] Lustre: DEBUG MARKER: == racer test 1: racer on clients: centos-90.localnet DURATION=2700 ================================== 09:38:17 (1517668697)
[119292.154075] Lustre: lfs: using old ioctl(LL_IOC_LOV_GETSTRIPE) on [0x200000401:0xfa:0x0], use llapi_layout_get_by_path()
[119298.778891] Lustre: DEBUG MARKER: racer test_1: @@@@@@ FAIL: generate lss conf (mds1)
[119363.272972] LustreError: 2970:0:(mdt_lvb.c:402:mdt_lvbo_fill()) lustre-MDT0000: small buffer size 448 for EA 472 (max_mdsize 472): rc = -34
[121710.001204] LustreError: 6319:0:(lov_obd.c:1308:lov_stripe_lock()) ASSERTION( md->lsm_lock_owner == 0 ) failed: 
[121710.002273] Slab corruption (Tainted: P           OE  ------------  ): kmalloc-192 start=ffff8800b7853140, len=192
[121710.002278] 000: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6d 6b 6b 6b 6b 6b  kkkkkkkkkkmkkkkk
[121710.002280] 010: 0e 00 00 00 6b 6b 6b 6b c0 04 ef 9e 02 88 ff ff  ....kkkk........
[121710.020795] LustreError: 6319:0:(lov_obd.c:1308:lov_stripe_lock()) LBUG
[121710.021476] Pid: 6319, comm: cat
[121710.022363] 
Call Trace:
[121710.023546]  [<ffffffffa020e7ce>] libcfs_call_trace+0x4e/0x60 [libcfs]
[121710.024252]  [<ffffffffa020e85c>] lbug_with_loc+0x4c/0xb0 [libcfs]
[121710.024971]  [<ffffffffa0870c79>] lov_stripe_lock+0xa9/0xb0 [lov]
[121710.025674]  [<ffffffffa0871283>] lov_attr_get_raid0+0x103/0x1f0 [lov]
[121710.026383]  [<ffffffffa1478c04>] ? ll_inode_size_lock+0x24/0x60 [lustre]
[121710.027106]  [<ffffffffa087429e>] lov_attr_get_composite+0x15e/0x3d0 [lov]
[121710.027838]  [<ffffffffa087103e>] lov_attr_get+0x1e/0x60 [lov]
[121710.028705]  [<ffffffffa0354a28>] cl_object_attr_get+0x78/0x150 [obdclass]
[121710.029720]  [<ffffffffa1463c7f>] ll_merge_attr+0x15f/0x420 [lustre]
[121710.030566]  [<ffffffffa14654e2>] ll_file_write_iter+0x4b2/0x540 [lustre]
[121710.031423]  [<ffffffffa146564b>] ll_file_aio_write+0xdb/0x180 [lustre]
[121710.032751]  [<ffffffffa1465794>] ll_file_write+0xa4/0x170 [lustre]
[121710.033601]  [<ffffffff811ed36d>] vfs_write+0xbd/0x1e0
[121710.034398]  [<ffffffff811ede34>] SyS_write+0x84/0xf0
[121710.035191]  [<ffffffff8170fc49>] system_call_fastpath+0x16/0x1b
[121710.036001] 
[121710.036584] Kernel panic - not syncing: LBUG

Crashdump on onyx-68 in 192.168.123.190-2018-02-03-10:1*