Bug 1717727 - [RHOSP 13] Create new image volume cache entry when cloning fails
Summary: [RHOSP 13] Create new image volume cache entry when cloning fails
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat OpenStack
Classification: Red Hat
Component: openstack-cinder
Version: 13.0 (Queens)
Hardware: Unspecified
OS: Unspecified
medium
urgent
Target Milestone: ---
: ---
Assignee: Pablo Caruana
QA Contact: Tzach Shefi
Tana
URL:
Whiteboard:
Depends On: 1719755
Blocks:
TreeView+ depends on / blocked
 
Reported: 2019-06-06 02:42 UTC by Takashi Kajinami
Modified: 2019-09-03 16:54 UTC (History)
2 users (show)

Fixed In Version: openstack-cinder-12.0.7-3.el7ost
Doc Type: Bug Fix
Doc Text:
Previously, in certain cases, when the Block Storage service failed to create a cloned volume from the existing image cache volume, it retrieved an image from the Image service, but did not recreate the image cache. With this update, the Block Storage services recreates the image cache after failing to create a cloned volume.
Clone Of:
: 1719755 (view as bug list)
Environment:
Last Closed: 2019-09-03 16:54:21 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Launchpad 1801595 0 None None None 2019-06-06 02:44:03 UTC
OpenStack gerrit 663518 0 'None' 'MERGED' 'Create new image volume cache entry when cloning fails' 2019-12-04 02:01:51 UTC
Red Hat Product Errata RHBA-2019:2627 0 None None None 2019-09-03 16:54:29 UTC

Description Takashi Kajinami 2019-06-06 02:42:54 UTC
Description of problem:

If we enable image cache feature in cinder, cinder create an image cache volume
in storage and create a volume by cloning that image cache volume.

In current RHOSP13, when cinder fails to create an cloned volume from the existing
image cache volume, it retrieves image from glance but does never recreate the image cache.
This causes a problem when we use XtremIO as cinder backend, as XtremIO has a snapshot limit
per volume group, which prevents that cloning when we have a number of volumes already
created from the same image cache volume.

There is a bug report on launchpad[1], and the fix is already merged in master [2],
so we should backport this to also our downstream version.

[1] https://bugs.launchpad.net/cinder/+bug/1801595
[2] https://review.opendev.org/#/c/630753/


How reproducible:
Always

Steps to Reproduce:
1. Configure XtremIO cinder backend and image cache enabled
2. Create 100(=xtremio_volumes_per_glance_cache) volumes from the same image
3. Again create some volumes from the same image

Actual results:
- Cinder always retrieves an image from glance

Expected results:
- Cinder creates a new image cache volume and create volumes using it

Comment 10 errata-xmlrpc 2019-09-03 16:54:21 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2019:2627


Note You need to log in before you can comment on or make changes to this bug.