Bug #10919: TestClsRbd.object_map_update hangs - rbd - Ceph

Actions

Copy link

Bug #10919

closed

TestClsRbd.object_map_update hangs

Added by Sage Weil about 9 years ago. Updated about 9 years ago.

Status:

Resolved

Priority:

Urgent

Assignee:

Target version:

% Done:

Source:

Q/A

Tags:

Backport:

Regression:

Severity:

3 - minor

Reviewed:

Affected Versions:

ceph-qa-suite:

Pull request ID:

Crash signature (v1):

Crash signature (v2):

Description

ubuntu@teuthology:/a/sage-2015-02-17_22:30:59-rados-wip-sage-testing-distro-basic-multi/764184
ubuntu@teuthology:/a/sage-2015-02-17_22:30:59-rados-wip-sage-testing-distro-basic-multi/763938

2015-02-18T22:24:52.656 INFO:tasks.workunit.client.0.burnupi15.stdout:[ RUN      ] TestClsRbd.object_map_resize
2015-02-18T22:24:52.942 INFO:tasks.workunit.client.0.burnupi15.stdout:[       OK ] TestClsRbd.object_map_resize (285 ms)
2015-02-18T22:24:52.942 INFO:tasks.workunit.client.0.burnupi15.stdout:[ RUN      ] TestClsRbd.object_map_update
2015-02-19T01:24:39.309 INFO:tasks.workunit:Stopping ['cls'] on client.0...

Actions

Copy link

Updated by Jason Dillaman about 9 years ago

OSDs spin with the following after a object_map_update:

2015-02-20 09:45:19.061412 7fe1e8aea700  0 bad crc in data 1776605727 != exp 3001002531
2015-02-20 09:45:19.061709 7fe1e8aea700  0 -- 127.0.0.1:6801/19153 >> 127.0.0.1:6812/1019346 pipe(0x4f56000 sd=97 :6801 s=2 pgs=574030 cs=540711 l=0 c=0x41f0000).fault with nothing to send, going to standby
2015-02-20 09:45:19.061939 7fe1e8ded700  1 -- 127.0.0.1:6801/19153 >> :/0 pipe(0x4cd6000 sd=60 :6801 s=0 pgs=0 cs=0 l=0 c=0x41f2d60).accept sd=60 127.0.0.1:39399/0
2015-02-20 09:45:19.061981 7fe1e8fef700  0 -- 127.0.0.1:6801/19153 >> 127.0.0.1:6809/19560 pipe(0x434f000 sd=55 :51571 s=2 pgs=492480 cs=454035 l=0 c=0x41f1ce0).fault, initiating reconnect

The cached CRC for the object map's bit vector is invalid and needs to be recomputed after its update.

Actions

Copy link