HNAS iSCSI Driver errors in concurrent scenarios
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Cinder |
Fix Released
|
Undecided
|
Erlon R. Cruz |
Bug Description
It might happen that multiple cinder nodes are trying to add volumes to the same target. In this situation, the driver can get an error trying to add a volume to a hlun just taken by other instance. In this case we backoff and retry ISCSI_ATTACH_
How to reproduce:
1 - Create 10 or more instances at the same time (boot from image, creates a new volume)
The following error will be shown:
=======
1b57d45-
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
2015-07-15 13:56:11.223 TRACE cinder.
Changed in cinder: | |
milestone: | none → liberty-3 |
status: | Fix Committed → Fix Released |
Changed in cinder: | |
milestone: | liberty-3 → 7.0.0 |
Fix proposed to branch: master /review. openstack. org/203780
Review: https:/