Devstack- Cinder Volume failed to initialize driver

[maestropandy]

In devstack liberty, while ./rejoin-stack.sh, cinder volume throwing error as Failed to initialize driver and not sending heart beat

" ERROR cinder.service [-] Manager for service cinder-volume maestro@lvmdriver-1 is reporting problems, not sending heartbeat. Service will appear "down".
"

Have to initialize lvmdriver backing file every time to get rid of it, by running

sudo losetup -f /opt/stack/data/stack-volumes-lvmdriver-1-backing-file
sudo losetup -f /opt/stack/data/stack-volumes-default-backing-file

Find c-vol logs http://paste.openstack.org/show/521854/

Should be fixed ?

[Syed Ahsan Shamim Zaidi (ahsanmohsin04)]

I had the same error which I resolved using this simple line of code:

sudo losetup -f --show /opt/stack/data/stack-volumes-lvmdriver-1-backing-file

Kindly let me know if it gets resolved.

[maestropandy (maestropandy)]

Syed, Can you update on this bug which you assigned for yourself

[Michal Dulko (michal-dulko-f)]

rejoin_stack.sh is gone on master and it never was supposed to be used to restart DevStack after reboot. Please use unstack.sh and stack.sh for that purpose. Personally I'm running my DevStack on a VirtualBox VM, which I'm able to suspend and resume after my laptop gets rebooted.

[Josh Fuhs (fuhsjr00)]

I am seeing this problem on both Ubuntu 16.04 and 18.04 with every HEAD revision of devstack pulled in the last couple months. Here's the last commit where I saw the failure: 9689083d74dcab67aeb34113096d4d94f749af39

The lvm backing file initialization commands noted in the description correct the problem, but it would be nice if a devstack node could be restarted and resume its functions without intervention.