Bug #42369
msg/async: nonexistent auth users leads to auth timeout, not fast failure
% Done:
0%
Source:
Tags:
Backport:
nautilus
Regression:
No
Severity:
3 - minor
Reviewed:
Description
2019-10-18T07:49:03.759 INFO:tasks.workunit.client.0.mira118.stderr:++ ceph crash ls 2019-10-18T07:49:03.759 INFO:tasks.workunit.client.0.mira118.stderr:++ wc -l 2019-10-18T07:49:04.027 INFO:tasks.ceph.mon.a.mira118.stderr:2019-10-18T07:49:04.021+0000 7f7ecadbb700 -1 --2- [v2:172.21.9.120:3300/0,v1:172.21.9.120:6789/0] >> conn(0x55717dd3 4400 0x55717d399700 secure :-1 s=AUTH_ACCEPTING_MORE pgs=0 cs=0 l=1 rx=0 tx=0).run_continuation failed decoding of frame header: buffer::end_of_buffer 2019-10-18T07:49:04.228 INFO:tasks.ceph.mon.a.mira118.stderr:2019-10-18T07:49:04.222+0000 7f7ecfdc5700 -1 --2- [v2:172.21.9.120:3300/0,v1:172.21.9.120:6789/0] >> conn(0x55717d39 fc00 0x55717d398100 secure :-1 s=AUTH_ACCEPTING_MORE pgs=0 cs=0 l=1 rx=0 tx=0).run_continuation failed decoding of frame header: buffer::end_of_buffer 2019-10-18T07:49:04.629 INFO:tasks.ceph.mon.a.mira118.stderr:2019-10-18T07:49:04.623+0000 7f7ecfdc5700 -1 --2- [v2:172.21.9.120:3300/0,v1:172.21.9.120:6789/0] >> conn(0x55717dd3 5400 0x55717d603700 secure :-1 s=AUTH_ACCEPTING_MORE pgs=0 cs=0 l=1 rx=0 tx=0).run_continuation failed decoding of frame header: buffer::end_of_buffer 2019-10-18T07:49:04.964 INFO:tasks.workunit.client.0.mira118.stderr:+ '[' 0 = 4 ']'
/a/kchai-2019-10-18_07:01:56-rados-master-distro-basic-mira/4421143/
Related issues
History
#1 Updated by Kefu Chai over 4 years ago
Dan, could you take a look?
#2 Updated by Dan Mick over 4 years ago
that's not what you call a very useful log. Do you have any idea what the mon is trying to say?
#3 Updated by Sage Weil over 4 years ago
- Status changed from New to In Progress
- Assignee changed from Dan Mick to Sage Weil
- Priority changed from Normal to Urgent
#4 Updated by Sage Weil over 4 years ago
this was broken when i changed ceph-crash to try authenticating using user(s) that don't exist. the root problem though is that 'ceph -n client.doesnotexist ...' will time out instead of erroring out immediately.
#5 Updated by Sage Weil over 4 years ago
- Status changed from In Progress to Fix Under Review
- Pull request ID set to 31052
#6 Updated by Sage Weil over 4 years ago
fixes the problem:
4434495 pass scheduled_sage@teuthology rados/singleton/{all/test-crash.yaml msgr-failures/many.yaml msgr/async-v1only.yaml objectstore/bluestore-stupid.yaml rados.yaml supported-random-distro$/{centos_7.yaml}} 799s
#7 Updated by Sage Weil over 4 years ago
- Status changed from Fix Under Review to Pending Backport
- Backport set to nautilus
#8 Updated by Sage Weil over 4 years ago
- Subject changed from qa/workunits/rados/test_crash.sh fails to msg/async: nonexistent auth users leads to auth timeout, not fast failure
#9 Updated by Sage Weil over 4 years ago
- Project changed from RADOS to Messengers
#10 Updated by Nathan Cutler over 4 years ago
- Copied to Backport #42438: nautilus: msg/async: nonexistent auth users leads to auth timeout, not fast failure added
#11 Updated by Nathan Cutler over 4 years ago
- Status changed from Pending Backport to Resolved
While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are in status "Resolved" or "Rejected".