[LU-4249] exception RIP: lqe64_hash_keycmp+12

• Type: Bug
• Resolution: Fixed
• Priority: Major
• Fix Version/s: Lustre 2.5.3
• Affects Version/s: Lustre 2.4.0
• Environment: our source tree is at https://github.com/jlan/lustre-nas; Server running 2.4.0-4nasS
• Severity: 3
• Assignee: Niu Yawei (Inactive)
• Reporter: Mahmoud Hanafi

Description

OSS crashed with Projection Fault at lqe64_hash_keycmp

See attached files.
Looks like we are crashing here

libcfs/libcfs/hash.c
static cfs_hlist_node_t *
cfs_hash_bd_lookup_intent(cfs_hash_t *hs, cfs_hash_bd_t *bd,
.
.
.
cfs_hlist_for_each(ehnode, hhead) { <<<<<<<< CRASH
if (!cfs_hash_keycmp(hs, key, ehnode))
continue;

cfs_hlist_for_each defined as hlist_for_each but hlist_for_each doesn't seem to be defined any where.

See attached backtrace info.

Issue Links

• is duplicated by: LU-4807 repeating DQACQ failed with -37
• is related to: LU-4633 exception RIP: qsd_entry_iter_cb+29

Comments

[Peter Jones]

Landed for 2.7

[Niu Yawei (Inactive)]

b2_4: http://review.whamcloud.com/11019
b2_5: http://review.whamcloud.com/11020

[Niu Yawei (Inactive)]

http://review.whamcloud.com/10988

[Niu Yawei (Inactive)]

Hi, Ihara, the ENOLCK error message problem is addressed in another ticket, please see LU-4920.

This debug patch is to collect information on lqe refcount when system crash on "exception RIP: ...", please keep the debug patch applied until the crash problem reproduced. Thanks.

[Shuichi Ihara (Inactive)]

Niu, we applied patches, and then, same error messages have been showing up. Attahced are recent debug log and syslog messages after applies patches.

[Niu Yawei (Inactive)]

Hi Niu, so, finally, what exactly patches should we apply againt b2_4 if we want enabled debug patches?

fix of LU-3460 http://review.whamcloud.com/#/c/8169/ and debug patch http://review.whamcloud.com/9833

[Shuichi Ihara (Inactive)]

Hi Niu, so, finally, what exactly patches should we apply againt b2_4 if we want enabled debug patches?

[javed shaikh (Inactive)]

thanks niu. deleted from here.

[Niu Yawei (Inactive)]

Hi Javed

Looks your post isn't related to this bug, could you open another ticket to track it? Thanks.

[Jay Lan (Inactive)]

Hi Niu, I cherry-picked LU-3460 and then #9833. It applied clean now. Thanks!

[Niu Yawei (Inactive)]

Sorry, I made the patch against master mistakenly, here is the patch for b2_4: http://review.whamcloud.com/9833

This new patch doesn't address the conflict in lustre/quota/qsd_writeback.c, because there is a quota fix introduced in 2.4.2 (LU-3460 http://review.whamcloud.com/#/c/8169/), I suggest you apply this fix first.