[LU-7053] Do not use osd_object_find in osd_index_ea_insert Created: 27/Aug/15 Updated: 20/Oct/17 Resolved: 14/Dec/15 |
|
| Status: | Resolved |
| Project: | Lustre |
| Component/s: | None |
| Affects Version/s: | Lustre 2.8.0 |
| Fix Version/s: | Lustre 2.8.0 |
| Type: | Bug | Priority: | Major |
| Reporter: | Di Wang | Assignee: | Di Wang |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | None | ||
| Issue Links: |
|
||||||||
| Severity: | 3 | ||||||||
| Rank (Obsolete): | 9223372036854775807 | ||||||||
| Description |
|
Do not use osd_object_find in osd_index_ea_insert() to avoid that notorious object find hung. |
| Comments |
| Comment by Gerrit Updater [ 27/Aug/15 ] |
|
wangdi (di.wang@intel.com) uploaded a new patch: http://review.whamcloud.com/16117 |
| Comment by Andreas Dilger [ 17/Sep/15 ] |
|
Please describe this problem more fully. Is this a crash, LBUG, deadlock, performance issue, code cleanup? How easily is this problem hit, and under what kind of operations (simple usage, racer, racer+migration, never)? |
| Comment by Di Wang [ 17/Sep/15 ] |
|
I found this problem during racer + migration, and it might cause deadlock if we use os_object_find in osd_index_ea_insert(), because of double object find in one service handler(). |
| Comment by Alex Zhuravlev [ 01/Oct/15 ] |
|
would http://review.whamcloud.com/#/c/15924/ help ? |
| Comment by Di Wang [ 01/Oct/15 ] |
|
Hmm, this is different issue, do I miss sth? |
| Comment by Alex Zhuravlev [ 06/Oct/15 ] |
|
well, still the question: who is destroying the object being referenced by a new direntry ? |
| Comment by Gerrit Updater [ 09/Nov/15 ] |
|
Alex Zhuravlev (alexey.zhuravlev@intel.com) uploaded a new patch: http://review.whamcloud.com/17092 |
| Comment by Gerrit Updater [ 14/Dec/15 ] |
|
Oleg Drokin (oleg.drokin@intel.com) merged in patch http://review.whamcloud.com/17092/ |
| Comment by Peter Jones [ 14/Dec/15 ] |
|
Landed for 2.8 |