1509398: Performance issue during volume detachment in ISCSIConnector when iSCSI multipath is used

Bug #1521648 reported by Alexey Stupnikov
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Mirantis OpenStack
Invalid
High
Alexey Stupnikov
7.0.x
Confirmed
High
Alexey Stupnikov

Bug Description

This is a "sub-bug" of https://bugs.launchpad.net/mos/+bug/1509398

When detaching a multipath volume in an environment where there are many attached volumes, excessive multipath calls are
generated and it takes too much time. This issue is due to the fact that a multipath -ll call against many multipath devices takes a few seconds.

When detaching a volume, _disconnect_volume_multipath_iscsi() calls 'multipath -ll <each iscsi device>' against every iscsi devices. This behavior is to identify the IQNs used by multipath device, however the IQNs are extracted from one 'multipath -ll' call without massive 'multipath -ll <each iscsi device>' calls.

Original issue: https://bugs.launchpad.net/os-brick/+bug/1456480
Original code-review: https://review.openstack.org/#/c/232846/

Changed in mos:
milestone: 7.0-mu-2 → 7.0-updates
description: updated
Revision history for this message
Alexey Stupnikov (astupnikov) wrote :

Original patch was pushed to fix the bug with os-brick in Liberty release. Liberty components use os-brick software to initiate all block storage connections. The purposed fix was backported from liberty to kilo, but only cinder drivers utilises brick in kilo, nova has its own iSCSI initiator. So the patch wouldn't fix anything, since nova compute will still work the old way.

Changed in mos:
milestone: 7.0-updates → 8.0
status: Confirmed → New
Changed in mos:
status: New → Invalid
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.