Project

General

Profile

Actions
Copy link

Bug #61645

open

MON: Monitor crash after downgrading from 17.2.6 to 17.2.5

Added by Andrea Bolzonella 11 months ago.

Status:
New
Priority:
Normal
Assignee:
-
Category:
Monitor
Target version:
-
% Done:

0%

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

Description

If you donwgrade a cluster from 17.2.6 to 17.2.5 with

ceph orch upgrade start --image=quay.io/ceph/ceph:v17.2.5

the fist monitor to be downgrade crash with

2023-06-05T14:43:57.850+0000 7f8750ede880  7 mon.alpha@-1(???).log v549915 update_from_paxos applying incremental log 549915 2023-06-05T13:33:25.003363+0000 mon.b
ravo (mon.2) 508 : audit [DBG] from='mgr.33514 192.168.56.11:0/2172538025' entity='mgr.bravo.tjdhbl' cmd=[{"prefix": "quorum_status"}]: dispatch
2023-06-05T14:43:57.850+0000 7f8750ede880  7 mon.alpha@-1(???).log v549915 update_from_paxos applying incremental log 549915 2023-06-05T13:33:25.004290+0000 mon.b
ravo (mon.2) 509 : audit [DBG] from='mgr.33514 192.168.56.11:0/2172538025' entity='mgr.bravo.tjdhbl' cmd=[{"prefix": "mon ok-to-stop", "ids": ["alpha"]}]: dispatch
2023-06-05T14:43:57.850+0000 7f8750ede880 10 mon.alpha@-1(???).log v549915  summary.channel_info {audit=172359,182364,cephadm=0,5056,cluster=540067,550068}
2023-06-05T14:43:57.850+0000 7f8750ede880 10 mon.alpha@-1(???).log v549915 check_subs
2023-06-05T14:43:57.850+0000 7f8750ede880  1 mon.alpha@-1(???).paxosservice(auth 251..319) refresh upgraded, format 0 -> 3
2023-06-05T14:43:57.850+0000 7f8750ede880 10 mon.alpha@-1(???).auth v319 update_from_paxos
2023-06-05T14:43:57.850+0000 7f8750ede880 10 mon.alpha@-1(???).auth v319 update_from_paxos version 319 keys ver 0 latest 301
2023-06-05T14:43:57.850+0000 7f8750ede880  7 mon.alpha@-1(???).auth v319 update_from_paxos loading summary e 301
2023-06-05T14:43:57.850+0000 7f8750ede880  7 mon.alpha@-1(???).auth v319 update_from_paxos latest length 3693
terminate called after throwing an instance of 'ceph::buffer::v15_2_0::end_of_buffer'
  what():  End of buffer
*** Caught signal (Aborted) **
 in thread 7f8750ede880 thread_name:ceph-mon
 ceph version 17.2.5 (98318ae89f1a893a6ded3a640405cdbb33e08757) quincy (stable)
 1: /lib64/libpthread.so.0(+0x12cf0) [0x7f874e355cf0]
 2: gsignal()
 3: abort()
 4: /lib64/libstdc++.so.6(+0x9009b) [0x7f874d96b09b]
 5: /lib64/libstdc++.so.6(+0x9653c) [0x7f874d97153c]
 6: /lib64/libstdc++.so.6(+0x96597) [0x7f874d971597]
 7: /lib64/libstdc++.so.6(+0x967f8) [0x7f874d9717f8]
 8: (ceph::buffer::v15_2_0::list::iterator_impl<true>::copy(unsigned int, char*)+0xa5) [0x7f8750777655]
 9: (CryptoKey::decode(ceph::buffer::v15_2_0::list::iterator_impl<true>&)+0x80) [0x7f8750410450]
 10: (KeyServer::decode(ceph::buffer::v15_2_0::list::iterator_impl<true>&)+0x2f8) [0x55af3c202158]
 11: (AuthMonitor::update_from_paxos(bool*)+0x485) [0x55af3c1ebcc5]
 12: (Monitor::refresh_from_paxos(bool*)+0x104) [0x55af3c129ed4]
 13: (Monitor::preinit()+0x95d) [0x55af3c15731d]
 14: main()
 15: __libc_start_main()
 16: _start()

It can be reproduced in a freshly installed cluster with version 17.2.6 or
with a cluster upgraded from version 17.2.5

No data to display

Actions
Copy link

Also available in: Atom PDF