Merge lp:~zulcss/nova/nova-iscsi-choice into lp:~hudson-openstack/nova/trunk
Proposed by
Chuck Short
Status: | Work in progress |
---|---|
Proposed branch: | lp:~zulcss/nova/nova-iscsi-choice |
Merge into: | lp:~hudson-openstack/nova/trunk |
Diff against target: |
195 lines (+94/-36) 3 files modified
nova/flags.py (+2/-0) nova/tests/test_volume.py (+2/-2) nova/volume/driver.py (+90/-34) |
To merge this branch: | bzr merge lp:~zulcss/nova/nova-iscsi-choice |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Brian Lamar (community) | Needs Fixing | ||
Review via email: mp+71188@code.launchpad.net |
Description of the change
Allow the user to choose between iscsitarget and tgt for iscsi userland tools.
To post a comment you must log in.
Unmerged revisions
- 1418. By Chuck Short
-
Fix pep8 errors
- 1417. By Chuck Short
-
Fixup thinkos
- 1416. By Chuck Short
-
Allow the user to choose either ietadm or tgtadm
19 + self.volume. driver. _execute( "sudo", "%s" %FLAGS. iscsi_helper, "--op", "show",
This and there are a number of other PEP8 violations to fix.
169 + if FLAGS.iscsi_helper == 'tgtadm': disk/by- path/ip- %s-iscsi- %s-lun- 0" % properties[ 'target_ portal' ], s['target_ iqn'])) disk/by- path/ip- %s-iscsi- %s-lun- 0" % properties[ 'target_ portal' ], s['target_ iqn']))
170 + mount_device = ("/dev/
171 + (iscsi_
172 + iscsi_propertie
173 + else:
174 + mount_device = ("/dev/
175 + (iscsi_
176 + iscsi_propertie
I don't think we need this if statement? Unless I'm missing something it's the same code. If you'd like you can put a note in the code indicating that the two supported iscsi implementations share the same strategy here but I'm not sure even that is necessary.
How are you currently testing this code? Manually? Have you tested both ietadm and tgtadm? I get very uneasy around code that's not only not tested, but isn't really testable.