raising error during image clone doen't delete cache
Bug #1801595 reported by
Shay Halsband
This bug affects 4 people
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Cinder |
Fix Released
|
Undecided
|
Alan Bishop |
Bug Description
XtremIO, like other arrays, has a snapshot limit per volume group. Volume snapshot is the mechanism used for volume cloning and image cache. hence the driver need to restart the cache from time to time.
The way thing worked till Pike in order to restart the cache, the driver would raise CinderException from create_
I suspect this change I547fb4bcdd4783
description: | updated |
To post a comment you must log in.
When we return at this code:
/usr/lib/ python2. 7/site- packages/ cinder/ volume/ flows/manager/ create_ volume. py
else:
model_ update, cloned = self._create_ from_image_ cache(
The
CinderException:
Cloned = False
(Because it reach our limit per volume)
Now in default value when you enter:
def _create_ from_image_ cache_or_ download( self, context, volume,
image_ location, image_id,
image_ meta, image_service,
update_ cache=False) :
So you get here:
if not cloned and update_cache:
should_ create_ cache_entry = True
With should_ create_ cache_entry = false
And we not fall back to default behavior in the rest of the code that happened when
should_ create_ cache_entry = True