Bug #43882
osd to mon connection lost, osd stuck down
0%
Description
This is a similar symptom to #43825, but it does not appear to be related to split/merge.
OSD is marked down, but up, but a bit (but not way) behind on OSDMaps. monc is not talking to the mon.
ceph daemon commands hang.
running under valgrind.
/a/sage-2020-01-28_23:42:06-rados-wip-sage-testing-2020-01-28-1413-distro-basic-smithi/4715655
description: rados/verify/{centos_latest.yaml ceph.yaml clusters/{fixed-2.yaml openstack.yaml}
d-thrash/default/{default.yaml thrashosds-health.yaml} msgr-failures/few.yaml msgr/async-v1only.yaml
objectstore/bluestore-bitmap.yaml rados.yaml tasks/rados_api_tests.yaml validater/valgrind.yaml}
ceph-osd.1.log
History
#1 Updated by Sage Weil over 3 years ago
- Status changed from New to Need More Info
adding debug: https://github.com/ceph/ceph/pull/32968
#2 Updated by Sage Weil over 3 years ago
- Subject changed from osd to mon connection lost, osd stuck down to monclient connection to mon failed
- Status changed from Need More Info to In Progress
- Assignee set to Sage Weil
#3 Updated by Sage Weil over 3 years ago
- Subject changed from monclient connection to mon failed to osd to mon connection lost, osd stuck down
- Status changed from In Progress to Need More Info
- Assignee deleted (
Sage Weil)
i thought i reproduced this, but it was a bug in another PR i was testing.
#4 Updated by Neha Ojha over 3 years ago
- Status changed from Need More Info to Can't reproduce