| *** jwcroppe has joined #openstack-powervm | 00:38 | |
| *** thorst has quit IRC | 00:43 | |
| *** edmondsw has joined #openstack-powervm | 02:38 | |
| *** edmondsw has quit IRC | 02:39 | |
| *** thorst has joined #openstack-powervm | 02:44 | |
| *** thorst has quit IRC | 02:49 | |
| *** edmondsw has joined #openstack-powervm | 03:41 | |
| *** edmondsw has quit IRC | 03:46 | |
| *** mdrabe has quit IRC | 03:58 | |
| *** thorst has joined #openstack-powervm | 04:45 | |
| *** thorst has quit IRC | 04:51 | |
| *** jwcroppe has quit IRC | 04:57 | |
| *** jwcroppe has joined #openstack-powervm | 04:57 | |
| *** edmondsw has joined #openstack-powervm | 05:29 | |
| *** edmondsw has quit IRC | 05:33 | |
| *** thorst has joined #openstack-powervm | 06:46 | |
| *** thorst has quit IRC | 06:52 | |
| *** edmondsw has joined #openstack-powervm | 07:17 | |
| *** edmondsw has quit IRC | 07:21 | |
| *** k0da has joined #openstack-powervm | 08:13 | |
| *** k0da has quit IRC | 08:36 | |
| *** thorst has joined #openstack-powervm | 08:48 | |
| *** thorst has quit IRC | 08:53 | |
| *** thorst has joined #openstack-powervm | 09:49 | |
| *** thorst has quit IRC | 09:54 | |
| *** smatzek has joined #openstack-powervm | 10:31 | |
| *** smatzek has quit IRC | 10:32 | |
| *** smatzek has joined #openstack-powervm | 10:32 | |
| *** edmondsw has joined #openstack-powervm | 10:53 | |
| *** edmondsw has quit IRC | 10:58 | |
| *** thorst has joined #openstack-powervm | 11:12 | |
| *** thorst has quit IRC | 11:13 | |
| *** smatzek has quit IRC | 11:15 | |
| *** smatzek has joined #openstack-powervm | 11:15 | |
| *** thorst has joined #openstack-powervm | 11:18 | |
| *** thorst_ has joined #openstack-powervm | 11:19 | |
| *** thorst has quit IRC | 11:22 | |
| *** thorst_ has quit IRC | 11:44 | |
| *** svenkat has joined #openstack-powervm | 11:47 | |
| *** svenkat_ has joined #openstack-powervm | 11:54 | |
| *** svenkat has quit IRC | 11:55 | |
| *** svenkat_ is now known as svenkat | 11:55 | |
| *** edmondsw has joined #openstack-powervm | 12:14 | |
| *** mdrabe has joined #openstack-powervm | 12:42 | |
| *** esberglu has joined #openstack-powervm | 12:58 | |
| *** thorst has joined #openstack-powervm | 13:29 | |
| *** jwcroppe has quit IRC | 13:43 | |
| *** k0da has joined #openstack-powervm | 13:57 | |
| *** jwcroppe has joined #openstack-powervm | 14:01 | |
| mdrabe | efried: https://review.openstack.org/#/c/471926/ | 14:11 |
|---|---|---|
| mdrabe | (the NVRAM performance stuff) Anything you think should tested explicitly with that? | 14:12 |
| efried | mdrabe How about RR? | 14:12 |
| efried | Did you put the caching stuff into that same change? | 14:13 |
| mdrabe | Yea it's in there | 14:13 |
| efried | I haven't looked at it since PS1 | 14:13 |
| efried | okay, then the commit message should change. | 14:13 |
| mdrabe | Should be testing the performance side of things today | 14:13 |
| efried | good | 14:13 |
| mdrabe | I think I did change the commit message | 14:13 |
| mdrabe | I still put it as WIP though | 14:13 |
| efried | Sorry, I meant the commit header. I see the text in the message, but the header should mention cache and/or perf somehow. | 14:14 |
| efried | But yeah, it'll need another edit to remove the WIP/TODOs. | 14:15 |
| mdrabe | I still gotta UT some stuff too | 14:15 |
| efried | Anyway, if it gets through perf testing and pvc RR tests, I'll be satisfied it doesn't have obvious regressions. | 14:15 |
| efried | Course, I'm still reserving final judgment pending actual perf results. | 14:16 |
| mdrabe | Yup yup | 14:16 |
| *** esberglu has quit IRC | 14:22 | |
| mdrabe | efried: Are you okay with hitting the RR test cases after it merges (if it passes performance)? | 14:23 |
| efried | mdrabe Uh, no, the point of the RR tests is to make sure it doesn't regress behavior. And RR is the scenario for which this whole mess was created in the first place. So we should definitely regression test RR for a change this pervasive. | 14:24 |
| *** esberglu has joined #openstack-powervm | 14:25 | |
| efried | mdrabe The performance tests just do a bunch of spawns, right? So the only path we're really testing there is storing the NVRAM data, and the only thing we've verified if it works is that those paths don't raise exceptions. We haven't proven that they actually *work*. | 14:27 |
| efried | You're changing the keying mechanism by which the data is stored & retrieved. So we need to test storage/retrieval in such a way that it actually matters what we're storing and retrieving. | 14:27 |
| mdrabe | efried: Yes... pinging you my perspective | 14:29 |
| efried | Proving things like: We're saving and updating the NVRAM data for the right VM at the right times; we're storing/updating VM1's NVRAM data keyed against VM1, not VM2; when we retrieve the NVRAM data for VM1, we're getting the data for VM1, not VM2; etc. | 14:30 |
| edmondsw | esberglu did we have a workaround we can remove now that https://review.openstack.org/#/c/474987/ has merged? I'm sure we must, but I can't remember what it was... | 14:35 |
| esberglu | edmondsw: Not sure off of a quick glance. I'll take a deeper look later today | 14:44 |
| edmondsw | esberglu tx | 14:45 |
| *** k0da has quit IRC | 15:26 | |
| *** k0da has joined #openstack-powervm | 15:57 | |
| *** burgerk has joined #openstack-powervm | 15:59 | |
| edmondsw | esberglu nevermind about https://review.openstack.org/#/c/474987/ I remember what that was for now, and it wasn't to do with our CI | 17:29 |
| esberglu | edmondsw: ack | 17:30 |
| *** jwcroppe has quit IRC | 17:33 | |
| *** smatzek has quit IRC | 17:44 | |
| *** jwcroppe has joined #openstack-powervm | 17:59 | |
| *** jwcroppe has quit IRC | 18:04 | |
| *** jwcroppe has joined #openstack-powervm | 18:07 | |
| *** burgerk_ has joined #openstack-powervm | 18:18 | |
| *** burgerk has quit IRC | 18:22 | |
| *** burgerk_ has quit IRC | 18:22 | |
| *** openstackgerrit has joined #openstack-powervm | 19:22 | |
| openstackgerrit | Matt Rabe proposed openstack/nova-powervm master: WIP: Change NVRAM manager and Swift APIs to accept UUID https://review.openstack.org/471926 | 19:22 |
| openstackgerrit | Matt Rabe proposed openstack/nova-powervm master: WIP: Change NVRAM manager and Swift APIs to accept UUID https://review.openstack.org/471926 | 19:27 |
| thorst | efried: per our discussion yesterday, see 5533. It doesn't pass py27 yet...but I talked with nvcastet and I like this direction more. | 20:48 |
| thorst | it allows the REST API to take in arbitrary metadata for the OVS | 20:48 |
| *** esberglu has quit IRC | 20:57 | |
| *** esberglu has joined #openstack-powervm | 20:57 | |
| *** svenkat has quit IRC | 21:01 | |
| *** esberglu has quit IRC | 21:02 | |
| efried | thorst ovs_ext_ids seems like a terrible name. | 21:02 |
| efried | They seem more like ext_attrs than ext_ids | 21:02 |
| thorst | well, in OVS they're external_ids | 21:03 |
| thorst | external_id:vm-uuid=thing1 | 21:03 |
| thorst | external_id:vm-status=thing2 | 21:03 |
| thorst | but yeah, maybe we need more iteration there | 21:03 |
| efried | thorst Then they're poorly named there. | 21:03 |
| efried | But okay. | 21:03 |
| thorst | heh, peace with that. | 21:03 |
| efried | Probably not a good idea for us to deviate. | 21:03 |
| efried | Though it would be nice if our docstrings mentioned the correlation there. | 21:04 |
| thorst | well...let me go verify things a bit more | 21:05 |
| edmondsw | I was just commenting on the same thing | 21:13 |
| *** thorst has quit IRC | 21:33 | |
| *** edmondsw has quit IRC | 22:23 | |
| *** edmondsw has joined #openstack-powervm | 22:28 | |
| *** edmondsw has quit IRC | 22:32 | |
| *** mdrabe has quit IRC | 22:40 | |
| *** edmondsw has joined #openstack-powervm | 22:56 | |
| *** edmondsw has quit IRC | 23:00 | |
| *** thorst has joined #openstack-powervm | 23:04 | |
| *** thorst has quit IRC | 23:09 | |
| *** k0da has quit IRC | 23:28 | |
| *** thorst has joined #openstack-powervm | 23:33 | |
| *** thorst has quit IRC | 23:34 | |
| *** efried has quit IRC | 23:41 | |
| -openstackstatus- NOTICE: nb03.openstack.org has been cleaned up and rebooted, and should return to building rotation | 23:44 | |
| *** efried has joined #openstack-powervm | 23:51 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!