Add retry to cinder API calls related to volume detach
When shutting down an instance for which volume needs to be
deleted, if cinder RPC timeout expires before cinder volume
driver terminates connection, then an unknown cinder exception
is received and the volume is not removed.
This fix adds a retry mechanism directly in cinder API calls
attachment_delete, terminate_connection, and detach.
Reviewed: https:/ /review. opendev. org/669674 /git.openstack. org/cgit/ openstack/ nova/commit/ ?id=01c334cbdd8 59f4e486ac2c369 a4bdb3ec7709cc
Committed: https:/
Submitter: Zuul
Branch: master
commit 01c334cbdd859f4 e486ac2c369a4bd b3ec7709cc
Author: Francois Palin <email address hidden>
Date: Mon Jul 8 10:12:25 2019 -0400
Add retry to cinder API calls related to volume detach
When shutting down an instance for which volume needs to be
deleted, if cinder RPC timeout expires before cinder volume
driver terminates connection, then an unknown cinder exception
is received and the volume is not removed.
This fix adds a retry mechanism directly in cinder API calls delete, terminate_ connection, and detach.
attachment_
Change-Id: I3c9ae47d0ceb64 fa3082a01cb7df2 7faa4f5a00d
Closes-Bug: #1834659