Bug #42053
e0 failed to get devid for : udev_device_new_from_subsystem_sysname failed on ''
0%
Description
Hello,
Using ceph-ansible stable-4.0 to deploy Ceph.
It seems there is some problem with mon election:
Environment:
OpenStack Zuul CI:
3 nodes (primary, secondary1, secondary2)
Logs:
primary journal: https://zuul.opendev.org/t/openstack/build/61ce7a19650242599f6c20403d4db267/log/primary/logs/system_logs/syslog.txt.gz
secondary1 journal: https://zuul.opendev.org/t/openstack/build/61ce7a19650242599f6c20403d4db267/log/secondary1/logs/system_logs/syslog.txt.gz
secondary2 journal: https://zuul.opendev.org/t/openstack/build/61ce7a19650242599f6c20403d4db267/log/secondary2/logs/system_logs/syslog.txt.gz
ceph-ansible output: https://zuul.opendev.org/t/openstack/build/61ce7a19650242599f6c20403d4db267/log/primary/logs/ansible/deploy-ceph
group_vars:
https://zuul.opendev.org/t/openstack/build/61ce7a19650242599f6c20403d4db267/log/primary/logs/kolla_configs/group_vars/all/ceph-ansible.yml
https://zuul.opendev.org/t/openstack/build/61ce7a19650242599f6c20403d4db267/log/primary/logs/kolla_configs/group_vars/all/ceph-osds.yml
Can you please help me out here?
This is part of an effort to move from kolla/kolla-ansible owned docker images and code to deploy Ceph to use Ceph-container and ceph-ansible
History
#1 Updated by Greg Farnum over 4 years ago
- Project changed from RADOS to ceph-ansible
This sounds like a ceph-ansible issue so I'm moving it to that project, but other channels for talking with them might work better.
#2 Updated by Neha Ojha over 4 years ago
Feel free to move it back, if there is nothing wrong on the ceph-ansible side.
#3 Updated by Matthias Leopold about 4 years ago
I'm seeing the same on a 3 node Ceph 14.2.8 cluster, that was installed with ceph-deploy, not ceph-ansible. I would be happy to see this explained. My environment is CentOS 7.
#4 Updated by Florent Flament almost 4 years ago
Hi,
I deployed a 3 nodes ceph cluster using ceph-ansible stable-4.0 too. The cluster has been working well for a couple a weeks, though I keep seeing a message similar to the one in this bug title in every ceph-mon logs.
For instance:
```
$ sudo journalctl -u ceph-mon@yumy | tail -10
Apr 27 12:07:42 yumy ceph-mon1046: 2020-04-27 12:07:42.752 7fd39ea25700 -1 mon.yumy@1(electing) e13 failed to get devid for : fallback method has serial ''but no model
Apr 27 12:07:42 yumy ceph-mon1046: 2020-04-27 12:07:42.804 7fd39ea25700 -1 mon.yumy@1(electing) e13 failed to get devid for : fallback method has serial ''but no model
Apr 27 12:35:22 yumy ceph-mon1046: 2020-04-27 12:35:22.483 7fd39ea25700 -1 mon.yumy@1(electing) e13 failed to get devid for : fallback method has serial ''but no model
Apr 27 12:35:22 yumy ceph-mon1046: 2020-04-27 12:35:22.527 7fd39ea25700 -1 mon.yumy@1(electing) e13 failed to get devid for : fallback method has serial ''but no model
Apr 27 12:35:31 yumy ceph-mon1046: 2020-04-27 12:35:31.090 7fd39ea25700 -1 mon.yumy@1(electing) e13 failed to get devid for : fallback method has serial ''but no model
Apr 27 12:35:31 yumy ceph-mon1046: 2020-04-27 12:35:31.090 7fd39ea25700 -1 mon.yumy@1(electing) e13 failed to get devid for : fallback method has serial ''but no model
Apr 27 12:37:04 yumy ceph-mon1046: 2020-04-27 12:37:04.213 7fd39ea25700 -1 mon.yumy@1(electing) e13 failed to get devid for : fallback method has serial ''but no model
Apr 27 12:37:04 yumy ceph-mon1046: 2020-04-27 12:37:04.261 7fd39ea25700 -1 mon.yumy@1(electing) e13 failed to get devid for : fallback method has serial ''but no model
Apr 27 13:10:04 yumy ceph-mon1046: 2020-04-27 13:10:04.757 7fd39ea25700 -1 mon.yumy@1(electing) e13 failed to get devid for : fallback method has serial ''but no model
Apr 27 13:10:04 yumy ceph-mon1046: 2020-04-27 13:10:04.757 7fd39ea25700 -1 mon.yumy@1(electing) e13 failed to get devid for : fallback method has serial ''but no model
```
The 3 nodes are running Ubuntu bionic.