Activity log for bug #1512387

Date Who What changed Old value New value Message
2015-11-02 15:11:59 Xing Yang bug added bug
2015-11-02 15:12:12 Xing Yang tags drivers emc scaleio
2015-11-02 15:12:20 Xing Yang cinder: status New In Progress
2015-11-02 15:12:24 Xing Yang cinder: assignee Xing Yang (xing-yang)
2015-11-10 02:16:46 Xing Yang description ScaleIO driver should implement update_migrated_volume for migrate volume to work properly. ScaleIO driver should implement update_migrated_volume. Without implementing update_migrated_volume, when the migrate finishes, the original volume will be deleted but the name on the new volume will not be changed to the old id even though the Cinder database will get updated. Create volume ScaleIO Volume ID: b511a2bd00000009 ScaleIO Volume Name: TcBhywAxSsKkgovcBnEsyg== Cinder Volume ID: 4dc061cb-0031-4ac2-a482-8bdc06712cca Migrate started New ScaleIO Volume ID: b511a2be00000000 New ScaleIO Volume Name: nExD3oXWR+K15Uyr/M0ucg== New Cinder Volume ID: d1c2f94c-083a-4959-b806-3ead4f1b6b24 Migrate finished ScaleIO Volume ID: b511a2be00000000 (okay. other volume deleted) ScaleIO Volume Name: nExD3oX0031WR+K15Uyr/M0ucg== (bad!) Cinder Volume ID: 4dc061cb-0031-4ac2-a482-8bdc06712cca (original ID) The problem is that the Volume Name is a base64() encoding of the Cinder Volume ID so any OpenStack calls after the migrate fail since the lookup fails.
2015-11-10 02:16:57 Xing Yang cinder: importance Undecided High
2015-11-18 16:49:47 Xing Yang cinder: milestone mitaka-1
2015-11-19 02:02:04 OpenStack Infra cinder: status In Progress Fix Committed
2015-11-27 22:44:09 OpenStack Infra tags drivers emc scaleio drivers emc in-stable-liberty scaleio
2015-12-03 16:43:47 Thierry Carrez cinder: status Fix Committed Fix Released