[LU-12218] ASSERTION( mbits == (req->rq_mbits & PTLRPC_BULK_OPS_MASK) ) failes if OBD_CONNECT_BULK_MBITS isn't used. Created: 24/Apr/19  Updated: 04/May/19  Resolved: 04/May/19

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: Lustre 2.13.0

Type: Bug Priority: Minor
Reporter: Andriy Skulysh Assignee: Andriy Skulysh
Resolution: Fixed Votes: 0
Labels: None

Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

> 2019-03-15T05:57:34.982434-05:00 c0-0c2s11n1 memfill3 invoked oom-killer: gfp_mask=0x14280ca(GFP_HIGHUSER_MOVABLE|__GFP_ZERO), nodemask=(null), order=0, oom_score_adj=0
> 2019-03-15T05:57:34.982440-05:00 c0-0c2s11n1 LNetError: 8294:0:(lib-lnet.h:236:lnet_md_alloc()) LNET: 295539320 total bytes allocated by lnet
> 2019-03-15T05:57:34.982480-05:00 c0-0c2s11n1 memfill3 cpuset=1824551 mems_allowed=0-1
> 2019-03-15T05:57:34.982492-05:00 c0-0c2s11n1 LustreError: 8294:0:(niobuf.c:398:ptlrpc_register_bulk()) snx11242-OST0000-osc-ffff880f8a142800: LNetMDAttach failed x1628044403952257/1: rc = -12
> 2019-03-15T05:57:34.982497-05:00 c0-0c2s11n1 Out of memory: Kill process 53231 (atpBackend) score 1000 or sacrifice child
....
> 2019-03-15T05:57:34.982801-05:00 c0-0c2s11n1 LustreError: 8294:0:(niobuf.c:358:ptlrpc_register_bulk()) ASSERTION( mbits == (req->rq_mbits & PTLRPC_BULK_OPS_MASK) ) failed: first mbits = x1628044403952259, last mbits = x1628044403952262
> 2019-03-15T05:57:34.982812-05:00 c0-0c2s11n1 LustreError: 8294:0:(niobuf.c:358:ptlrpc_register_bulk()) LBUG
> 2019-03-15T05:57:34.982824-05:00 c0-0c2s11n1 CPU: 79 PID: 8294 Comm: ptlrpcd_01_23 Tainted: P W O 4.12.14-25.22_5.0.69-cray_ari_c #1 SLE15 (unreleased)
> 2019-03-15T05:57:34.982832-05:00 c0-0c2s11n1 Hardware name: Cray Inc. Cascade/Cascade, BIOS 5.11 03/22/2018
> 2019-03-15T05:57:34.982839-05:00 c0-0c2s11n1 Call Trace:
> 2019-03-15T05:57:34.982848-05:00 c0-0c2s11n1 dump_stack+0x70/0xa1
> 2019-03-15T05:57:34.982855-05:00 c0-0c2s11n1 libcfs_call_trace+0x6d/0x70 [libcfs]
> 2019-03-15T05:57:34.982860-05:00 c0-0c2s11n1 lbug_with_loc+0x43/0xa0 [libcfs]
> 2019-03-15T05:57:34.982867-05:00 c0-0c2s11n1 ptlrpc_register_bulk+0x8cf/0xa30 [ptlrpc]



 Comments   
Comment by Gerrit Updater [ 24/Apr/19 ]

Andriy Skulysh (c17819@cray.com) uploaded a new patch: https://review.whamcloud.com/34753
Subject: LU-12218 ptlrpc: Bulk assertion fails on -ENOMEM
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 89721ee419ed2a0f6849740e788c7f3b7b83a96f

Comment by Patrick Farrell (Inactive) [ 24/Apr/19 ]

Is this a version interop issue?  If so, which versions?  If not, why is BULK_MBITS not being used in this case?

Comment by Gerrit Updater [ 04/May/19 ]

Oleg Drokin (green@whamcloud.com) merged in patch https://review.whamcloud.com/34753/
Subject: LU-12218 ptlrpc: Bulk assertion fails on -ENOMEM
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: e63a49fa69205cd6730206eaf3a7e39824586238

Comment by Peter Jones [ 04/May/19 ]

Landed for 2.13

Generated at Sat Feb 10 02:50:40 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.