Merge lp:~clark-laughlin/charms/trusty/nova-compute/arm64-patch-1 into lp:~openstack-charmers-archive/charms/trusty/nova-compute/next
| Status: | Merged |
|---|---|
| Merged at revision: | 110 |
| Proposed branch: | lp:~clark-laughlin/charms/trusty/nova-compute/arm64-patch-1 |
| Merge into: | lp:~openstack-charmers-archive/charms/trusty/nova-compute/next |
| Diff against target: |
113 lines (+30/-3) 5 files modified
hooks/nova_compute_context.py (+4/-1) templates/havana/nova.conf (+7/-0) templates/juno/nova.conf (+7/-0) templates/kilo/nova.conf (+5/-0) unit_tests/test_nova_compute_contexts.py (+7/-2) |
| To merge this branch: | bzr merge lp:~clark-laughlin/charms/trusty/nova-compute/arm64-patch-1 |
| Related bugs: |
| Reviewer | Review Type | Date Requested | Status |
|---|---|---|---|
| Liam Young | 2015-03-12 | Approve on 2015-03-30 | |
| Ryan Beisner | 2015-03-12 | Pending | |
|
Review via email:
|
|||
This proposal supersedes a proposal from 2015-03-12.
Description of the Change
This change allows the nova-compute charm to deploy on arm64 machines without requiring a different configuration to be used. To deploy nova-compute on arm64 requires 2 settings in nova.conf, which are not required for amd64:
libvirt_
libvirt_disk_prefix = vd
Since juju configuration is specified at the service level, there is not a way to provide a separate configuration when deploying an amd64 unit and an arm64 unit -- preventing deployment of openstack with a mixture of arm64 and amd64 nova-compute nodes.
This change simply detects the machine architecture, and adds the two configuration settings to nova.conf when running on an arm64 (aarch64) machine. Behavior when running on other architectures is unchanged.
charm_unit_test #2399 nova-compute-next for clark-laughlin mp252662
UNIT FAIL: unit-test failed
UNIT Results (max last 2 lines):
FAILED (SKIP=5, failures=3)
make: *** [unit_test] Error 1
Full unit test output: http://
Build: http://
| Ryan Beisner (1chb1n) wrote : | # |
It looks like the unit test will also need some updating.
| Clark Laughlin (clark-laughlin) wrote : | # |
Hmm - is it possible that the test results include all of the other changes
on that branch, or is this just for my change? I'll have to figure out how
the test works and what's failing – – curious, because my changes should
not alter the behavior of anything unless it is run on an ARM host
On Wednesday, March 11, 2015, Ryan Beisner <email address hidden>
wrote:
> Review: Needs Fixing
>
> It looks like the unit test will also need some updating.
> --
>
> https:/
> You are the owner of
> lp:~clark-laughlin/charms/trusty/nova-compute/arm64-patch-1.
>
charm_amulet_test #2434 nova-compute-next for clark-laughlin mp252662
AMULET OK: passed
Build: http://
| Ryan Beisner (1chb1n) wrote : | # |
See inline comment.
| Ryan Beisner (1chb1n) wrote : | # |
If you can apply this patch or edit like so, that should do the trick:
http://
To re-confirm manually before committing:
make unit_test
| Clark Laughlin (clark-laughlin) wrote : | # |
Thank you! I'll give it a try later tonight and resubmit.
On Wednesday, March 11, 2015, Ryan Beisner <email address hidden>
wrote:
> If you can apply this patch or edit like so, that should do the trick:
> http://
>
> To re-confirm manually before committing:
> make unit_test
>
> --
>
> https:/
> You are the owner of
> lp:~clark-laughlin/charms/trusty/nova-compute/arm64-patch-1.
>
charm_lint_check #2610 nova-compute-next for clark-laughlin mp252662
LINT FAIL: lint-test failed
LINT Results (max last 2 lines):
unit_
make: *** [lint] Error 1
Full lint test output: http://
Build: http://
charm_unit_test #2401 nova-compute-next for clark-laughlin mp252662
UNIT OK: passed
charm_amulet_test #2435 nova-compute-next for clark-laughlin mp252662
AMULET FAIL: amulet-test failed
AMULET Results (max last 2 lines):
ERROR subprocess encountered error code 1
make: *** [test] Error 1
Full amulet test output: http://
Build: http://
charm_lint_check #2611 nova-compute-next for clark-laughlin mp252682
LINT OK: passed
charm_unit_test #2402 nova-compute-next for clark-laughlin mp252682
UNIT OK: passed
charm_amulet_test #2436 nova-compute-next for clark-laughlin mp252682
AMULET FAIL: amulet-test failed
AMULET Results (max last 2 lines):
ERROR subprocess encountered error code 1
make: *** [test] Error 1
Full amulet test output: http://
Build: http://
| Ryan Beisner (1chb1n) wrote : | # |
Amulet test failed due to environment timeout issue, unrelated to the proposed changes. I kicked off a re-run of that test.
charm_amulet_test #2437 nova-compute-next for clark-laughlin mp252682
AMULET OK: passed
Build: http://
| Liam Young (gnuoy) wrote : | # |
Thank you for the contribution, it is very much appreciated. Approved

charm_lint_check #2609 nova-compute-next for clark-laughlin mp252662
LINT OK: passed
Build: http:// 10.245. 162.77: 8080/job/ charm_lint_ check/2609/