[LU-13988] mdt_lock_handle_fini()) ASSERTION( !lustre_handle_is_used(&lh->mlh_reg_lh) ) failed Created: 25/Sep/20 Updated: 13/Jul/22 Resolved: 19/Nov/20 |
|
| Status: | Resolved |
| Project: | Lustre |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | Lustre 2.14.0 |
| Type: | Bug | Priority: | Minor |
| Reporter: | Andriy Skulysh | Assignee: | Andriy Skulysh |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | None | ||
| Issue Links: |
|
||||
| Severity: | 3 | ||||
| Rank (Obsolete): | 9223372036854775807 | ||||
| Description |
00000100:00100000:1.0:1586690876.387325:0:13960:0:(service.c:2228:ptlrpc_server_handle_request()) Handling RPC req@ffff886d97baf180 pname:cluuid+ref:pid:xid:nid:opc:job mdt00_008:fd032a03-ac98-3bd0-9cda-ed98f0173f88+12716:20834:x1663761626790848:12345-192.168.103.15@tcp:101:truncate.0 00010000:00010000:1.0:1586690876.387331:0:13960:0:(ldlm_lockd.c:1186:ldlm_handle_enqueue0()) ### server-side enqueue handler START 00010000:00010000:1.0:1586690876.387333:0:13960:0:(ldlm_lockd.c:1229:ldlm_handle_enqueue0()) @@@ found existing lock cookie 0x9f8ceb1ed5a613fb req@ffff886d97baf180 x1663761626790848/t0(0) o101->fd032a03-ac98-3bd0-9cda-ed98f0173f88@192.168.103.15@tcp:254/0 lens 480/0 e 0 to 0 dl 1586690889 ref 1 fl Interpret:/2/ffffffff rc 0/-1 job:'truncate.0' 00000004:00010000:1.0:1586690876.387353:0:13960:0:(mdt_handler.c:3879:mdt_intent_fixup_resent()) ### Restoring lock cookie ns: mdt-lustre-MDT0000_UUID lock: ffff886d74800240/0x9f8ceb1ed5a613fb lrc: 3/0,0 mode: PW/PW res: [0x200000404:0x16ae6:0x0].0x0 bits 0x40/0x0 rrc: 6 type: IBT flags: 0x40200000000000 nid: 192.168.103.15@tcp remote: 0x806313bfb96a7bd expref: 12716 pid: 13961 timeout: 0 lvb_type: 0 00000004:00010000:1.0:1586690876.387357:0:13960:0:(mdt_handler.c:3881:mdt_intent_fixup_resent()) @@@ restoring lock cookie 0x9f8ceb1ed5a613fb req@ffff886d97baf180 x1663761626790848/t0(0) o101->fd032a03-ac98-3bd0-9cda-ed98f0173f88@192.168.103.15@tcp:254/0 lens 480/0 e 0 to 0 dl 1586690889 ref 1 fl Interpret:/2/ffffffff rc 0/-1 job:'truncate.0' 00000004:00040000:1.0:1586690876.387385:0:13960:0:(mdt_handler.c:3650:mdt_lock_handle_fini()) ASSERTION( !lustre_handle_is_used(&lh->mlh_reg_lh) ) failed: 00000004:00040000:1.0:1586690876.387395:0:13960:0:(mdt_handler.c:3650:mdt_lock_handle_fini()) LBUG #5 [ffff886db2d3bb10] mdt_thread_info_fini at ffffffffc1443206 [mdt]
ffff886db2d3bb18: [ffff886e2ab5d000:kmalloc-4096] 00000000fffffffe
ffff886db2d3bb28: ffff886e352ac510 ffff886db2d3bb70
ffff886db2d3bb38: mdt_intent_policy+218
crash> p (int)0xfffffffe $7 = -2 // ENOENT |
| Comments |
| Comment by Gerrit Updater [ 25/Sep/20 ] |
|
Andriy Skulysh (c17819@cray.com) uploaded a new patch: https://review.whamcloud.com/40045 |
| Comment by Gerrit Updater [ 19/Nov/20 ] |
|
Oleg Drokin (green@whamcloud.com) merged in patch https://review.whamcloud.com/40045/ |
| Comment by Peter Jones [ 19/Nov/20 ] |
|
Landed for 2.14 |
| Comment by Gerrit Updater [ 01/Jul/22 ] |
|
"DELBARY Gael <gael.delbary@cea.fr>" uploaded a new patch: https://review.whamcloud.com/47854 |
| Comment by DELBARY Gael [ 01/Jul/22 ] |
|
We hit this LBUG in production. Root cause analysis (user source code) is in progress. |
| Comment by DELBARY Gael [ 04/Jul/22 ] |
|
The workload is (for one user), under a directory (where the ldlm_lock is held during the crash):
It is not hopefully a valid workload for us... |
| Comment by DELBARY Gael [ 13/Jul/22 ] |
|
Finally on our side, we are not in DoM path but we hit due to this patch https://review.whamcloud.com/47487. Fix now. Anyway pretty useful in DoM workload.
|