Project

General

Profile

Bug #1665

osd: last_peering_reset incorrect on stray?

Added by Sage Weil almost 10 years ago. Updated almost 10 years ago.

Status:
Resolved
Priority:
High
Assignee:
-
Category:
OSD
Target version:
% Done:

0%

Source:
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

on alexandria,

2011-10-31 10:36:02.650050 7fe305758700 -- [2607:f298:1:3237::6358]:6823/25353 <== osd.18 [2607:f298:1:3237::6358]:6817/25043 1279 ==== PGq v1 ==== 2472+0+0 (60464745 0 0) 0x30d31c0 con 0x25fea00
2011-10-31 10:36:02.650060 7fe305758700 osd.21 162 _dispatch 0x30d31c0 PGq v1
2011-10-31 10:36:02.650068 7fe305758700 osd.21 162 handle_pg_query from osd.18 epoch 150
2011-10-31 10:36:02.650077 7fe305758700 osd.21 162 require_same_or_newer_map 150 (i am 162) 0x30d31c0
2011-10-31 10:36:02.650087 7fe305758700 osd.21 162 ignoring old peering message PGq v1
2011-10-31 10:36:02.650107 7fe305758700 osd.21 162 ignoring old peering message PGq v1

when the pg hasn't changed with epoch 150.

History

#1 Updated by Sage Weil almost 10 years ago

this is why. the interval starts at 150,a nd that is when teh query is sent. on the stray, we hit it in 151:

2011-10-31 10:35:49.794331 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 140/140/93) [18,21] r=1 lcod 0'0 inactive] enter Reset
2011-10-31 10:35:49.794350 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 140/140/93) [18,21] r=1 lcod 0'0 inactive] state<Reset>: Reset advmap
2011-10-31 10:35:49.794381 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 140/140/93) [18,21] r=1 lcod 0'0 inactive] state<Reset>: up or acting affected, calling start_peering_interval again
2011-10-31 10:35:49.794405 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive]  noting past interval(140-149 [18,21]/[18,21] maybe_went_rw)
2011-10-31 10:35:49.794426 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive]  up [18,21] -> [18,21,4], acting [18,21] -> [18,21,4], role 1 -> 1
2011-10-31 10:35:49.794441 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] cancel_generate_backlog
2011-10-31 10:35:49.794457 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] on_change
2011-10-31 10:35:49.794472 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] remove_watchers
2011-10-31 10:35:49.794491 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] exit NotTrimming
2011-10-31 10:35:49.794510 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] enter NotTrimming
2011-10-31 10:35:49.794529 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] cancel_recovery
2011-10-31 10:35:49.794546 7fe305f59700 osd.21 150 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] clear_recovery_state
2011-10-31 10:35:49.921000 7fe305f59700 osd.21 151 Scanning pg pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive]
2011-10-31 10:35:49.921022 7fe305f59700 osd.21 151 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] handle_advance_map [18,21,4]/[18,21,4]
2011-10-31 10:35:49.921044 7fe305f59700 osd.21 151 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] state<Reset>: Reset advmap
2011-10-31 10:35:50.247961 7fe305f59700 osd.21 151 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] handle_activate_map 
2011-10-31 10:35:50.247981 7fe305f59700 osd.21 151 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] exit Reset 0.453649 2 0.000000
2011-10-31 10:35:50.248000 7fe305f59700 osd.21 151 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] enter Started
2011-10-31 10:35:50.248019 7fe305f59700 osd.21 151 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] enter Start
2011-10-31 10:35:50.248056 7fe305f59700 osd.21 151 pg[7.5( v 87'98 (87'96,87'98] n=34 ec=41 les/c 111/113 150/150/93) [18,21,4] r=1 lcod 0'0 inactive] state<Start>: transitioning to Stray

#2 Updated by Samuel Just almost 10 years ago

  • Status changed from New to Resolved

b9a0b2b7a4d3b5a7db1f942af0158712199377a8 reverted 6d123067ce1ba99522281d5c72623bd5ba3e0fc8

Also available in: Atom PDF