xenial-pike/proposed - keystone unable to load token provider
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Keystone Charm |
Fix Released
|
High
|
James Page |
Bug Description
I hit this issue whilst testing some updates for xenial-pike today:
2017-10-11 18:16:50.106231 mod_wsgi (pid=10696): Exception occurred processing WSGI script '/usr/bin/
2017-10-11 18:16:50.106249 Traceback (most recent call last):
2017-10-11 18:16:50.106261 File "/usr/bin/
2017-10-11 18:16:50.106275 application = initialize_
2017-10-11 18:16:50.106291 File "/usr/lib/
2017-10-11 18:16:50.106301 config_
2017-10-11 18:16:50.106306 File "/usr/lib/
2017-10-11 18:16:50.106313 startup_
2017-10-11 18:16:50.106318 File "/usr/lib/
2017-10-11 18:16:50.106325 drivers = backends.
2017-10-11 18:16:50.106329 File "/usr/lib/
2017-10-11 18:16:50.106336 token_provider_
2017-10-11 18:16:50.106340 File "/usr/lib/
2017-10-11 18:16:50.106347 init(self, *args, **kwargs)
2017-10-11 18:16:50.106351 File "/usr/lib/
2017-10-11 18:16:50.106358 self.__
2017-10-11 18:16:50.106362 File "/usr/lib/
2017-10-11 18:16:50.106519 super(Manager, self)._
2017-10-11 18:16:50.106533 File "/usr/lib/
2017-10-11 18:16:50.106589 self.driver = load_driver(
2017-10-11 18:16:50.106596 File "/usr/lib/
2017-10-11 18:16:50.106609 raise ImportError(msg, {'name': driver_name, 'namespace': namespace})
2017-10-11 18:16:50.106634 ImportError: (u'Unable to find %(name)r driver in %(namespace)r.', {'namespace': 'keystone.
Switching the keystone.conf to use the 'uuid' entry point resolves this issue, but I'm bemused as to why this has started happening (as I already had an environment deployed on pike prior to redeployment)
tags: | added: uosci |
tags: | added: test-gap |
Changed in charm-keystone: | |
status: | Fix Committed → Fix Released |
https:/ /review. openstack. org/#/c/ 511333/