commit 52947efccd8541831728575fe873b84273aa97f3
Author: Le Tian Ren <email address hidden>
Date: Wed Aug 20 15:05:35 2014 +0800
Timing issue in remote-keystone deployment model
This is actually timing issue between openstack service endpoints becoming
active and powervc-driver's client initialization of those.
1. get_client() call with version_filter might runs into the 'None' client
error mentioned in the bug, which means *currently* only glance sync is
affected. When no version filtered, we do a rediscover for the svc_type.
2. get_client() call without version_filter or new_client() call just choose
the latest avaiable version service, so no 'None' client error, although the
chosen version might not be the real *latest* but the default hardcoded 'v1'
version, due to the timing issue mentioned in the bug. For the latter case,
unless there is an observation/notification mechanism or poll mechanism for
service versions update, based on current design, there is no perfect solution
as far as I can imagine.
Reviewed: https:/ /review. openstack. org/115519 /git.openstack. org/cgit/ stackforge/ powervc- driver/ commit/ ?id=52947efccd8 541831728575fe8 73b84273aa97f3
Committed: https:/
Submitter: Jenkins
Branch: master
commit 52947efccd85418 31728575fe873b8 4273aa97f3
Author: Le Tian Ren <email address hidden>
Date: Wed Aug 20 15:05:35 2014 +0800
Timing issue in remote-keystone deployment model
This is actually timing issue between openstack service endpoints becoming
active and powervc-driver's client initialization of those.
1. get_client() call with version_filter might runs into the 'None' client
error mentioned in the bug, which means *currently* only glance sync is
affected. When no version filtered, we do a rediscover for the svc_type.
2. get_client() call without version_filter or new_client() call just choose notification mechanism or poll mechanism for
the latest avaiable version service, so no 'None' client error, although the
chosen version might not be the real *latest* but the default hardcoded 'v1'
version, due to the timing issue mentioned in the bug. For the latter case,
unless there is an observation/
service versions update, based on current design, there is no perfect solution
as far as I can imagine.
Change-Id: I9ca0f906c8954e cd621be22445c12 fd6a55a1ae7
Closes-Bug: 1358215