| *** svenkat has quit IRC | 00:23 | |
| *** thorst has joined #openstack-powervm | 00:31 | |
| *** apearson has joined #openstack-powervm | 00:33 | |
| *** thorst has quit IRC | 00:36 | |
| *** thorst has joined #openstack-powervm | 01:03 | |
| *** thorst has quit IRC | 01:03 | |
| *** thorst has joined #openstack-powervm | 03:04 | |
| *** thorst has quit IRC | 03:10 | |
| *** chhavi has joined #openstack-powervm | 04:07 | |
| *** apearson has quit IRC | 04:43 | |
| *** apearson has joined #openstack-powervm | 04:44 | |
| *** apearson has quit IRC | 04:45 | |
| *** apearson has joined #openstack-powervm | 04:45 | |
| *** apearson has joined #openstack-powervm | 04:46 | |
| *** apearson has joined #openstack-powervm | 04:46 | |
| *** apearson has quit IRC | 04:47 | |
| *** apearson has joined #openstack-powervm | 04:47 | |
| *** apearson has quit IRC | 04:48 | |
| *** apearson has joined #openstack-powervm | 04:48 | |
| *** apearson has quit IRC | 04:49 | |
| *** apearson has joined #openstack-powervm | 04:49 | |
| *** apearson has quit IRC | 04:49 | |
| *** apearson has joined #openstack-powervm | 04:50 | |
| *** apearson has quit IRC | 04:50 | |
| *** apearson has joined #openstack-powervm | 04:50 | |
| *** apearson has quit IRC | 04:51 | |
| *** apearson has joined #openstack-powervm | 04:51 | |
| *** apearson has quit IRC | 04:52 | |
| *** apearson has joined #openstack-powervm | 04:52 | |
| *** apearson has quit IRC | 04:53 | |
| *** apearson has joined #openstack-powervm | 04:53 | |
| *** thorst has joined #openstack-powervm | 05:05 | |
| *** thorst has quit IRC | 05:10 | |
| *** thorst has joined #openstack-powervm | 06:27 | |
| *** thorst has quit IRC | 06:31 | |
| *** thorst has joined #openstack-powervm | 08:27 | |
| *** thorst has quit IRC | 08:32 | |
| *** thorst has joined #openstack-powervm | 10:29 | |
| *** thorst has quit IRC | 10:33 | |
| *** smatzek has joined #openstack-powervm | 10:55 | |
| *** jwcroppe has quit IRC | 11:29 | |
| *** svenkat has joined #openstack-powervm | 11:31 | |
| *** svenkat_ has joined #openstack-powervm | 11:39 | |
| *** svenkat has quit IRC | 11:40 | |
| *** svenkat_ is now known as svenkat | 11:40 | |
| *** jwcroppe has joined #openstack-powervm | 11:43 | |
| *** thorst has joined #openstack-powervm | 11:54 | |
| *** thorst_ has joined #openstack-powervm | 11:56 | |
| *** thorst_ has quit IRC | 11:56 | |
| *** thorst_ has joined #openstack-powervm | 11:57 | |
| *** thorst has quit IRC | 11:58 | |
| *** jwcroppe has quit IRC | 12:37 | |
| *** dwayne has quit IRC | 12:45 | |
| *** edmondsw has joined #openstack-powervm | 12:59 | |
| *** apearson has joined #openstack-powervm | 13:02 | |
| *** kylek3h has joined #openstack-powervm | 13:03 | |
| *** esberglu has joined #openstack-powervm | 13:07 | |
| *** efried_zzz is now known as efried | 13:29 | |
| *** dwayne has joined #openstack-powervm | 13:41 | |
| *** jwcroppe has joined #openstack-powervm | 13:54 | |
| *** dwayne has quit IRC | 13:59 | |
| *** apearson has quit IRC | 14:01 | |
| *** tjakobs has joined #openstack-powervm | 14:21 | |
| *** apearson has joined #openstack-powervm | 14:23 | |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm master: Adopt new pypowervm power_off APIs https://review.openstack.org/476274 | 15:04 |
|---|---|---|
| openstackgerrit | Eric Fried proposed openstack/nova-powervm master: Adopt new pypowervm power_off APIs https://review.openstack.org/476274 | 15:25 |
| openstackgerrit | Merged openstack/nova-powervm master: Updated from global requirements https://review.openstack.org/484068 | 15:49 |
| *** dwayne has joined #openstack-powervm | 15:55 | |
| *** apearson has quit IRC | 16:00 | |
| efried | esberglu Here's one we don't see often: https://review.openstack.org/#/c/476945/ <== ran out of procs. | 16:54 |
| *** apearson has joined #openstack-powervm | 17:12 | |
| openstackgerrit | Drew Thorstensen (thorst) proposed openstack/nova-powervm master: WIP: Use OVS Attributes within pypowervm https://review.openstack.org/486702 | 17:21 |
| thorst_ | efried: ^^ My WIP for the OVS stuff... | 17:21 |
| thorst_ | curious if/when you get initial impressions | 17:22 |
| *** thorst_ has quit IRC | 17:38 | |
| *** thorst has joined #openstack-powervm | 17:44 | |
| *** thorst has quit IRC | 17:49 | |
| *** chhavi has quit IRC | 18:10 | |
| *** thorst has joined #openstack-powervm | 18:12 | |
| efried | thorst Well, I looked at it. Can you help me understand what's going on? | 18:16 |
| efried | Before, if new_vif, we were creating a CNA and then running some linux commands. | 18:17 |
| efried | Now it looks like, if new_vif, we're creating a CNA and then creating another CNA. | 18:17 |
| thorst | we were always creating a pair | 18:19 |
| thorst | a trunk and a client | 18:19 |
| thorst | trunk being a CNA on the NL | 18:19 |
| efried | thorst I left some comments in the review | 18:20 |
| efried | But those are more stylistic/procedural than functional. | 18:21 |
| thorst | and an 'autospec' | 18:22 |
| thorst | which means I have to go learn auto spec'ing? | 18:22 |
| thorst | quick reference for me to work off of? | 18:22 |
| efried | thorst https://review.openstack.org/#/c/438119/5..6 | 18:25 |
| efried | If by "reference" you mean "examples". | 18:25 |
| efried | If you mean actual reference, https://docs.python.org/3/library/unittest.mock.html#auto-speccing | 18:27 |
| thorst | thx | 18:27 |
| efried | thorst Did you change the superclass with the intent of doing away with LIO stuff entirely? | 18:27 |
| efried | Cause at this point it looks like the super.plug isn't doing anything. | 18:27 |
| thorst | my eventual goal is to get rid of the LB implementation | 18:28 |
| thorst | I'd need adreznec to sign off on that though | 18:28 |
| thorst | and then make a vif folder | 18:28 |
| thorst | and then put in an 'ovs.py', 'sea.py', 'sriov.py', etc... | 18:28 |
| efried | ight. Then I guess you'll have to pull the PvmLioVifDriver.plug impl into PvmOvsVifDriver somehow. | 18:28 |
| thorst | it already is | 18:29 |
| thorst | and PvmOvsVifDriver no longer extends PvmLioVifDriver....only the LB implementation does | 18:30 |
| thorst | so if I can toss that... | 18:30 |
| thorst | then I can get rid of that whole class chain | 18:30 |
| efried | Oh, I see, that code actually *is* duplicated from the superclass. And the super.plug is what, just for form? | 18:30 |
| efried | Since the superclass method is abstract, prolly should just toss that call. | 18:31 |
| thorst | yeah, it was for form | 18:31 |
| thorst | but k | 18:31 |
| thorst | so you just want me to autospec the mgmt_partition thing? | 18:34 |
| efried | thorst The policy we enacted was basically to autospec anything from pypowervm in test cases that we touch. So all the pypowervm mocks in all the tests you touched here. | 18:36 |
| efried | I'd be especially concerned with the two CNA ones whose APIs changed in 1.1.6. | 18:36 |
| thorst | k...wondering how to auto spec a method instead of a class | 18:36 |
| thorst | investigating | 18:36 |
| efried | Point of autospec is to make sure we're calling those guys with the correct arg names & numbers, not just calling it how our test claims to be expecting it called. | 18:37 |
| thorst | right right. | 18:37 |
| efried | thorst See e.g. https://review.openstack.org/#/c/438119/5..6/nova/tests/unit/virt/powervm/test_driver.py@35 to mock a method. | 18:37 |
| efried | ...to autospec* a method | 18:38 |
| thorst | that seems absurdly simple | 18:38 |
| efried | :) | 18:38 |
| efried | thorst Added another couple comments | 18:42 |
| efried | thorst Oh, and what about 5588? If these fields are totally arbitrary, can they contain e.g. '<' and therefore screw up our XML? | 18:43 |
| thorst | efried: this seems odd " unbound method search() must be called with CNA instance as first argument (got MagicMock instance instead)" | 18:44 |
| thorst | efried: yep, they could. | 18:44 |
| thorst | so we probably need to sanitize.... | 18:44 |
| efried | At the REST level. It needs to be CDATA'd. | 18:44 |
| efried | that is, it needs to be sent to us from the server CDATA'd, and we need to CDATA it when we pass it in. | 18:45 |
| efried | thorst Autospec is making sure you're calling with the right kind of arguments. Whatever MagicMock you're using there needs to be specced itself. | 18:45 |
| thorst | efried: is that something nvcastet needs to do? | 18:46 |
| efried | thorst Yeah, or whoever handles the REST side. But also something we'll have to handle in pypowervm. | 18:46 |
| thorst | don't all fields that have string inputs have this issue? | 18:47 |
| efried | thorst Name me another that can accept binary or '<'. | 18:49 |
| efried | or '&' | 18:50 |
| thorst | lpar name? | 18:50 |
| efried | thorst Even if that's true, which I kinda doubt, nobody's gonna be stupid enough to put those in their LPAR names. | 18:55 |
| efried | thorst And no. See #novalink | 18:56 |
| efried | We have encoded data in some Jobs, and that stuff *is* CDATA-escaped. | 18:57 |
| *** jwcroppe has quit IRC | 18:58 | |
| *** apearson has quit IRC | 18:58 | |
| *** jwcroppe has joined #openstack-powervm | 18:59 | |
| *** apearson has joined #openstack-powervm | 19:01 | |
| *** jwcroppe_ has joined #openstack-powervm | 19:01 | |
| efried | thorst Looks like we get the python side for free. | 19:02 |
| thorst | for lpar_name or both? | 19:02 |
| efried | lpar_name is n/a, can't take ( ) \ < > * $ & ? | [ ] ' " `, ! # % + / | 19:02 |
| thorst | right... | 19:03 |
| efried | Actually, what I'm seeing is it automatically encodes/decodes. | 19:03 |
| efried | It's not CDATAing. | 19:03 |
| efried | lxml is taking care of that. | 19:03 |
| thorst | yeah, that's what I would've expected. | 19:03 |
| thorst | so we really just need REST to enforce | 19:03 |
| efried | So e.g. it converts > to > | 19:03 |
| thorst | ? | 19:03 |
| *** jwcroppe has quit IRC | 19:03 | |
| efried | Well | 19:03 |
| efried | Yeah, it would do the trick if REST was able to send & receive stuff thusly encoded. | 19:04 |
| efried | Which it would make sense if LWP already did transparently. | 19:04 |
| efried | I guess. | 19:04 |
| efried | Maybe. | 19:04 |
| efried | Needs to be tested out. | 19:04 |
| *** thorst is now known as thorst_afk | 19:15 | |
| *** apearson has quit IRC | 19:39 | |
| *** jwcroppe_ has quit IRC | 19:45 | |
| *** jwcroppe has joined #openstack-powervm | 19:45 | |
| *** apearson has joined #openstack-powervm | 19:46 | |
| *** jwcroppe has quit IRC | 19:50 | |
| *** jwcroppe has joined #openstack-powervm | 19:53 | |
| *** jwcroppe has quit IRC | 20:05 | |
| *** mdrabe has quit IRC | 20:05 | |
| *** mdrabe has joined #openstack-powervm | 20:09 | |
| openstackgerrit | Drew Thorstensen (thorst) proposed openstack/nova-powervm master: WIP: Use OVS Attributes within pypowervm https://review.openstack.org/486702 | 20:11 |
| *** jwcroppe has joined #openstack-powervm | 20:24 | |
| *** svenkat has quit IRC | 20:57 | |
| *** edmondsw has quit IRC | 21:04 | |
| *** edmondsw has joined #openstack-powervm | 21:04 | |
| *** edmondsw_ has joined #openstack-powervm | 21:05 | |
| *** edmondsw has quit IRC | 21:09 | |
| *** edmondsw_ has quit IRC | 21:10 | |
| *** smatzek has quit IRC | 21:20 | |
| *** apearson has quit IRC | 21:25 | |
| *** jwcroppe has quit IRC | 21:26 | |
| *** jwcroppe has joined #openstack-powervm | 21:32 | |
| *** edmondsw has joined #openstack-powervm | 21:39 | |
| *** jwcroppe has quit IRC | 21:42 | |
| *** edmondsw has quit IRC | 21:44 | |
| *** jwcroppe has joined #openstack-powervm | 21:45 | |
| *** esberglu has quit IRC | 21:46 | |
| *** dwayne has quit IRC | 22:05 | |
| *** esberglu has joined #openstack-powervm | 22:05 | |
| *** esberglu has quit IRC | 22:10 | |
| *** kylek3h has quit IRC | 22:13 | |
| *** jwcroppe has quit IRC | 22:37 | |
| *** tjakobs has quit IRC | 22:37 | |
| *** jwcroppe has joined #openstack-powervm | 22:44 | |
| *** jwcroppe has quit IRC | 23:06 | |
| *** dwayne has joined #openstack-powervm | 23:14 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!