TypeError on unicode token with some openstack setups
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
pyjuju |
Fix Released
|
High
|
Martin Packman | ||
0.5 |
Fix Released
|
High
|
Martin Packman |
Bug Description
Using the lastest trunk (561) I had this error in the bootstrap with the openstack provider:
flocon:~$ juju -v bootstrap
2012-07-30 10:30:19,806 DEBUG Initializing juju bootstrap runtime
2012-07-30 10:30:19,815 DEBUG openstack: using auth-mode 'userpass' with http://
2012-07-30 10:30:20,133 DEBUG openstack: authenticated til u'2012-
2012-07-30 10:30:20,134 DEBUG openstack: GET 'http://
Traceback (most recent call last):
Failure: twisted.
2012-07-30 10:30:20,136 ERROR Traceback (most recent call last):
Failure: twisted.
2012-07-30 10:30:20,136 ERROR [<twisted.
Turn out to be the token returned by keystone that was in UTF-8 and makes the next Request Generation to fail.
So if I force the str type in the token id recuperation in the _handle_v2_auth function the error go away.
But it surely have a better way to solve this.
Related branches
- Clint Byrum (community): Approve
- Diff: 0 lines
- John A Meinel: Approve
- Martin Packman (community): Approve
-
Diff: 126 lines (+72/-9)2 files modifiedjuju/providers/openstack/client.py (+6/-3)
juju/providers/openstack/tests/test_client.py (+66/-6)
Changed in juju: | |
importance: | Undecided → Medium |
status: | New → Confirmed |
summary: |
- Openstack provider do not support UTF-8 token + TypeError on unicode token with some openstack setups |
Changed in juju: | |
importance: | Medium → High |
Changed in juju: | |
assignee: | nobody → Martin Packman (gz) |
status: | Confirmed → In Progress |
milestone: | none → 0.6 |
Changed in juju: | |
status: | In Progress → Fix Committed |
Changed in juju: | |
status: | Fix Committed → Fix Released |
Thanks Patrick. That change is sufficient if the token does not in fact contain any non-ascii characters?