| *** esberglu has joined #openstack-powervm | 01:16 | |
| *** esberglu has quit IRC | 01:21 | |
| *** adreznec has joined #openstack-powervm | 04:54 | |
| *** adreznec has quit IRC | 04:55 | |
| *** adreznec has joined #openstack-powervm | 04:57 | |
| -openstackstatus- NOTICE: Due to some unforseen Zuul issues the gate is under very high load and extremely unstable at the moment. This is likely to persist until PST morning | 07:04 | |
| *** ChanServ changes topic to "Due to some unforseen Zuul issues the gate is under very high load and extremely unstable at the moment. This is likely to persist until PST morning" | 07:04 | |
| *** adreznec has quit IRC | 08:32 | |
| *** adreznec has joined #openstack-powervm | 08:32 | |
| *** efried is now known as fried_rice | 12:29 | |
| *** svenkat has joined #openstack-powervm | 13:18 | |
| *** svenkat_ has joined #openstack-powervm | 13:22 | |
| *** svenkat has quit IRC | 13:22 | |
| *** svenkat_ is now known as svenkat | 13:22 | |
| *** ChanServ changes topic to "This channel is for PowerVM-related development and discussion. For general OpenStack support, please use #openstack." | 14:00 | |
| -openstackstatus- NOTICE: The issues have been fixed, Zuul is operating fine again but has a large backlog. You can recheck jobs that failed. | 14:00 | |
| *** esberglu has joined #openstack-powervm | 14:24 | |
| esberglu | fried_rice: Got a couple questions about the vscsi change when you have a sec | 14:36 |
|---|---|---|
| fried_rice | esberglu Okay. I haven't looked at it yet, but shoot. | 14:52 |
| esberglu | https://review.openstack.org/#/c/526094/5/nova/virt/powervm/volume/vscsi.py | 14:55 |
| esberglu | L159 | 14:55 |
| esberglu | What's the conditional execution of the stg_ftsk about? | 14:56 |
| fried_rice | oh, ugh, yeah. | 14:56 |
| fried_rice | Man, I'm having a hard time remembering. mdrabe You around? | 14:56 |
| mdrabe | What's up? | 14:58 |
| mdrabe | Isn't it if there's no difference between the server response and what you're trying to POST? | 15:00 |
| fried_rice | No, different layer. | 15:00 |
| fried_rice | mdrabe I thought the conditional stg_ftsk business had something to do with the way pvc was overriding methods or something. | 15:01 |
| mdrabe | Oh that conditional on L159 | 15:02 |
| mdrabe | What about that docstring? | 15:03 |
| mdrabe | I'd give that as my answer | 15:03 |
| mdrabe | I'm assuming yea maybe that was something for pvc where we wanted to do things with the feed task outside of this volume driver | 15:05 |
| fried_rice | We really need thorst to answer this. | 15:05 |
| mdrabe | I'm probably wrong | 15:06 |
| esberglu | fried_rice: mdrabe: https://github.com/openstack/nova-powervm/commit/9b1319c1 | 15:07 |
| esberglu | Changeset it was introduced in, didn't help me understand further though | 15:08 |
| fried_rice | Yeah, I've been looking through that one. | 15:08 |
| mdrabe | Yup okay that makes sense | 15:08 |
| mdrabe | It's so we don't POST for NPIV and vSCSI separately right? | 15:08 |
| mdrabe | Can do it in one? | 15:08 |
| fried_rice | That's what the ftsk is for, yes. Or more to the point so we can do multiple attach/detach ops in the same POST. | 15:09 |
| fried_rice | But what esberglu is asking is why we have the conditional business, where if stg_ftsk is None we build and run it locally in the method. | 15:09 |
| fried_rice | I'm 90% sure it had something to do with pvc. | 15:09 |
| fried_rice | And we should probably get an answer from thorst as to whether we need to carry that into the in-tree driver. I imagine we do - unless pvc has quit needing it. | 15:10 |
| fried_rice | I'm trying to reach him on slack. We're not going to resolve this without his input. | 15:10 |
| esberglu | fried_rice: Let me know if you hear back, otherwise I'll toss him an email | 15:11 |
| fried_rice | ack | 15:11 |
| esberglu | fried_rice: Next up. L321. I know we are trying to avoid CONF options, wasn't sure what to do with that one | 15:14 |
| fried_rice | esberglu Yeah, get rid of it for now; we can add it later if needed. And collapse that whole thing to just making sure we don't have zero. | 15:16 |
| mdrabe | fried_rice, esberglu I can tell that where we build the volume connector we don't pass in a pre-built feed task | 15:20 |
| mdrabe | But it's not obvious to me why | 15:20 |
| fried_rice | mdrabe "we" being pvc? | 15:21 |
| mdrabe | Ya | 15:21 |
| fried_rice | Okay, so that supports what I said above. Along with the fact that there's nowhere in nova-powervm that we *don't* pass in the stg_ftsk. | 15:21 |
| fried_rice | But we still don't know why. I probably have it in my slack or ST history somewhere, but finding it... | 15:22 |
| fried_rice | esberglu I'm going to make an executive decision here: Take that logic out, make the stg_ftsk mandatory. If pvc needs this, it's gonna be several releases before they're ready to switch to the in-tree driver anyway, and we can add it back in (and justify it somehow?) at that time. | 15:23 |
| fried_rice | I mean, we could try to get away with it now, but I doubt we're going to slip that one by reviewers. | 15:23 |
| esberglu | fried_rice: ack | 15:24 |
| esberglu | fried_rice: Okay last one for now. L118 | 15:30 |
| esberglu | Right now this isn't getting used at all | 15:30 |
| esberglu | I think the only place it gets used OOT is | 15:30 |
| esberglu | https://github.com/openstack/nova-powervm/blob/master/nova_powervm/virt/powervm/driver.py#L1105 | 15:30 |
| esberglu | Which seems to be iscsi specific | 15:30 |
| esberglu | I can rip it out? | 15:30 |
| fried_rice | esberglu get_volume_connector is a ComputeDriver override that I suspect we need to implement for volume support in general. | 15:33 |
| fried_rice | It gets used by the compute manager. | 15:34 |
| esberglu | fried_rice: https://github.com/openstack/nova-powervm/blob/master/nova_powervm/virt/powervm/driver.py#L1111 | 15:40 |
| esberglu | Then just set that to None for now? | 15:40 |
| fried_rice | Prolly to whatever's prior to Ia78d2ed17232cc0f42fed1391786384e4dea28c1 | 15:41 |
| fried_rice | 'cept that guy is referencing iscsi too. And I know we didn't have actual iscsi. So it must be a thing where the name isn't significant. | 15:42 |
| fried_rice | well wtf | 15:43 |
| fried_rice | esberglu How about this version: https://review.openstack.org/#/c/424290/3/nova_powervm/virt/powervm/driver.py | 15:44 |
| fried_rice | (base of that patch) | 15:44 |
| fried_rice | geez, has it been that long since we started working on iscsi? | 15:45 |
| esberglu | fried_rice: https://github.com/openstack/nova-powervm/blob/master/nova_powervm/virt/powervm/volume/__init__.py#L101 | 15:52 |
| esberglu | The reason I was thinking to just set it None | 15:53 |
| fried_rice | esberglu Okay, sure, that wfm. | 16:02 |
| *** AlexeyAbashkin has joined #openstack-powervm | 16:40 | |
| *** AlexeyAbashkin has quit IRC | 16:44 | |
| *** fried_rice is now known as fried_rolls | 17:41 | |
| *** AlexeyAbashkin has joined #openstack-powervm | 19:46 | |
| *** fried_rolls is now known as fried_rice | 19:53 | |
| *** svenkat has quit IRC | 19:56 | |
| *** AlexeyAbashkin has quit IRC | 19:56 | |
| *** ChanServ has quit IRC | 20:17 | |
| *** ChanServ has joined #openstack-powervm | 20:24 | |
| *** barjavel.freenode.net sets mode: +o ChanServ | 20:24 | |
| *** k0da has joined #openstack-powervm | 21:10 | |
| *** fried_rice is now known as efried_cya_jan | 21:12 | |
| *** edmondsw has joined #openstack-powervm | 21:55 | |
| *** edmondsw has quit IRC | 21:56 | |
| *** esberglu has quit IRC | 23:10 | |
| *** k0da has quit IRC | 23:50 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!