| *** seroyer has quit IRC | 00:09 | |
| *** seroyer has joined #openstack-powervm | 00:22 | |
| *** thorst has joined #openstack-powervm | 00:24 | |
| *** kriskend_ has joined #openstack-powervm | 01:46 | |
| *** kriskend has quit IRC | 01:46 | |
| *** esberglu has joined #openstack-powervm | 01:58 | |
| *** esberglu has quit IRC | 02:08 | |
| openstackgerrit | Drew Thorstensen (thorst) proposed openstack/nova-powervm: WIP: Send full fabric to VIOS slot data for LPM https://review.openstack.org/364621 | 02:31 |
|---|---|---|
| *** thorst has quit IRC | 02:45 | |
| *** thorst has joined #openstack-powervm | 02:46 | |
| *** thorst has quit IRC | 02:54 | |
| *** wangqwsh has joined #openstack-powervm | 03:32 | |
| *** tjakobs has joined #openstack-powervm | 03:42 | |
| *** thorst has joined #openstack-powervm | 03:52 | |
| *** tjakobs has quit IRC | 03:57 | |
| *** thorst has quit IRC | 03:59 | |
| *** kriskend_ has quit IRC | 04:04 | |
| *** wangqwsh has quit IRC | 04:13 | |
| *** tjakobs has joined #openstack-powervm | 04:44 | |
| *** tjakobs has quit IRC | 04:48 | |
| *** thorst has joined #openstack-powervm | 04:56 | |
| *** thorst has quit IRC | 05:04 | |
| -openstackstatus- NOTICE: The logs volume is full causing jobs to fail with POST_FAILURE. This is being worked on, please do not recheck until notified. | 05:10 | |
| *** ChanServ changes topic to "The logs volume is full causing jobs to fail with POST_FAILURE. This is being worked on, please do not recheck until notified." | 05:10 | |
| *** ChanServ changes topic to "This channel is for PowerVM-related development and discussion. For general OpenStack support, please use #openstack." | 05:36 | |
| -openstackstatus- NOTICE: Space has been freed up on the log server. If you have POST_FAILURE results it is now safe to issue a 'recheck' | 05:36 | |
| *** thorst has joined #openstack-powervm | 06:02 | |
| *** kotra03 has joined #openstack-powervm | 06:02 | |
| *** kotra03 has quit IRC | 06:04 | |
| *** thorst has quit IRC | 06:09 | |
| *** kairo has joined #openstack-powervm | 06:11 | |
| *** kotra03 has joined #openstack-powervm | 06:23 | |
| *** thorst has joined #openstack-powervm | 06:36 | |
| *** thorst has quit IRC | 06:44 | |
| *** k0da has joined #openstack-powervm | 07:30 | |
| *** thorst has joined #openstack-powervm | 07:42 | |
| *** thorst has quit IRC | 07:48 | |
| *** thorst has joined #openstack-powervm | 08:46 | |
| *** thorst has quit IRC | 08:54 | |
| *** thorst has joined #openstack-powervm | 09:52 | |
| *** thorst has quit IRC | 09:58 | |
| *** kairo has quit IRC | 10:25 | |
| *** kairo has joined #openstack-powervm | 10:35 | |
| *** thorst has joined #openstack-powervm | 10:56 | |
| *** kairo has quit IRC | 10:58 | |
| *** kairo has joined #openstack-powervm | 11:02 | |
| *** thorst has quit IRC | 11:04 | |
| *** kairo has quit IRC | 11:06 | |
| *** kairo has joined #openstack-powervm | 11:21 | |
| *** thorst has joined #openstack-powervm | 11:57 | |
| *** tblakes has joined #openstack-powervm | 12:01 | |
| *** tblakes_ has joined #openstack-powervm | 12:02 | |
| *** tblakes has quit IRC | 12:06 | |
| *** tblakes_ is now known as tblakes | 12:07 | |
| *** kotra03 has quit IRC | 12:25 | |
| *** kairo has quit IRC | 12:25 | |
| *** kairo has joined #openstack-powervm | 12:25 | |
| *** kairo has joined #openstack-powervm | 12:25 | |
| *** apearson has joined #openstack-powervm | 12:43 | |
| *** dwayne_ has quit IRC | 12:49 | |
| efried | thorst, yt? | 13:08 |
| thorst | efried: yeah...just working on hardware requests. | 13:08 |
| efried | Ready to do some steppin'? | 13:08 |
| thorst | SR-IOV ones actually...the last ones | 13:08 |
| thorst | steppin? | 13:08 |
| *** tjakobs has joined #openstack-powervm | 13:08 | |
| efried | network_infos and crt_network_infos | 13:08 |
| thorst | can I finish the SR-IOV hardware request? I'm in the zone... | 13:08 |
| thorst | before I get mad about a different thing :-) | 13:09 |
| efried | Sho. But I found some very exciting and interesting stuff out. | 13:09 |
| efried | You won't get mad. This'll be a fun one to solve. | 13:09 |
| efried | Carry on. | 13:09 |
| thorst | fun eh? | 13:09 |
| thorst | see 3894 and https://review.openstack.org/#/c/364621/ | 13:11 |
| thorst | that was / is a real mind bender. | 13:11 |
| thorst | and I hope I can just abandon them TBH | 13:11 |
| *** kriskend_ has joined #openstack-powervm | 13:12 | |
| *** tjakobs has quit IRC | 13:13 | |
| *** seroyer has quit IRC | 13:25 | |
| *** tjakobs has joined #openstack-powervm | 13:33 | |
| *** tjakobs has quit IRC | 13:39 | |
| *** kotra03 has joined #openstack-powervm | 13:39 | |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm: HttpError=>VirtualInterface(Unp|P)lugException https://review.openstack.org/364923 | 13:40 |
| *** seroyer has joined #openstack-powervm | 13:40 | |
| efried | thorst - whom should I add to ^^ to assess pvc impacts? | 13:40 |
| thorst | esha seth, eric larese, myself, svenkat | 13:41 |
| *** kriskend_ has quit IRC | 13:45 | |
| thorst | efried: I need to know something from you for hw | 13:54 |
| thorst | I have 7 SR-IOV cards left. | 13:54 |
| efried | tell me | 13:54 |
| thorst | should I do 4 in one system (super weird config...but...yeah), and then 1 card in 3 systems | 13:54 |
| thorst | or | 13:54 |
| thorst | 2 cards in 3 systems and 1 in 1 system | 13:54 |
| thorst | or | 13:54 |
| thorst | 2 cards in 2 systems and 3 cards in 1 system | 13:55 |
| thorst | or | 13:55 |
| thorst | 1 card in 5 systems and 2 cards in 1 system <--- I'm leaning towards that | 13:55 |
| efried | I would say 2/2/1/1/1 | 13:57 |
| *** dwayne_ has joined #openstack-powervm | 13:57 | |
| efried | or 3/2/1/1 | 13:57 |
| thorst | OK...next q | 13:57 |
| thorst | I'm going to go 2/2/1/1/1 (most systems - better for CI later) | 13:57 |
| efried | Because we want to be able to do LPM permutations that involve asymmetrical card setups | 13:58 |
| thorst | other q....how many ports plugged in per card? Both 10 gig ports? | 13:58 |
| *** kriskend_ has joined #openstack-powervm | 13:58 | |
| efried | uhh. | 13:59 |
| efried | All of them? | 13:59 |
| thorst | OK...that's kinda a pain | 13:59 |
| efried | How many ports on each card, two? | 13:59 |
| thorst | two I think | 13:59 |
| thorst | should be fine...if they were 4 port I'd be grumpier | 13:59 |
| efried | thorst, can we set 'em up one way to start, then move some around? | 14:01 |
| thorst | efried: yeah...but its a PITA | 14:01 |
| thorst | lab team justifiably gets frustrated when we do that | 14:01 |
| thorst | the servers are heavy, the wiring is complex, etc... | 14:01 |
| efried | See, cause I think we should get some testing in with the following config: | 14:01 |
| efried | 4 ports across 2-3 cards and 2 VIOSes. Migrate back and forth between the 2- and 3-card system. That oughtta work. | 14:01 |
| thorst | they like one and done | 14:01 |
| thorst | hmm....I'd hope the PVC SVT is doing that. | 14:02 |
| thorst | we can do that with 2 and 1 card systems? | 14:02 |
| thorst | cause each pport can go to one vios? | 14:02 |
| efried | Yeah, kinda, but that'd only be 2-way redundancy. Different scenario. Less depth exercising nvcastet's placement algo. | 14:03 |
| efried | bbiab, mtg | 14:03 |
| *** kairo has quit IRC | 14:05 | |
| *** tjakobs has joined #openstack-powervm | 14:10 | |
| *** apearson has quit IRC | 14:28 | |
| *** apearson has joined #openstack-powervm | 14:30 | |
| adreznec | thorst: We're seeing "cloud-init-nonet[12.52]: waiting 120 seconds for network device" on deploys... that was related to the virt/interfaces.template ipv6 issue, right? | 14:30 |
| thorst | adreznec: yeppers | 14:32 |
| adreznec | *sigh* | 14:32 |
| adreznec | How did we fix that in the CI environment again? | 14:32 |
| thorst | we have a parameter that we set to not do the RMC setting | 14:32 |
| thorst | so we lost RMC but got fast speed up | 14:33 |
| *** esberglu has joined #openstack-powervm | 14:35 | |
| adreznec | thorst: Parameter... | 14:36 |
| thorst | heh, sec | 14:37 |
| adreznec | So... if we just don't create the MGMTVSWITCH | 14:37 |
| thorst | use_rmc_ipv6_scheme = False | 14:37 |
| adreznec | We wouldn't hit this, right? | 14:37 |
| adreznec | Or that | 14:37 |
| adreznec | We need to get this fixed | 14:37 |
| thorst | set the parameter... | 14:37 |
| adreznec | k | 14:38 |
| thorst | yeah, we do. But how can we fix RHEL's cloud-init? | 14:38 |
| thorst | this would be fine if the world just accepted IPv6 :-) | 14:38 |
| thorst | I wonder if you could put an Ipv4 floating IP on an IPv6 network. | 14:38 |
| adreznec | Not sure... | 14:38 |
| thorst | the other fix is to just make your network IPv6. | 14:38 |
| thorst | at first I was kidding about that, but for what you're doing... | 14:39 |
| thorst | that may actually be...desired? | 14:39 |
| adreznec | Yeah... | 14:41 |
| *** apearson_ has joined #openstack-powervm | 14:51 | |
| *** apearson has quit IRC | 14:53 | |
| *** apearson__ has joined #openstack-powervm | 14:56 | |
| *** apearson__ has quit IRC | 14:58 | |
| *** apearson_ has quit IRC | 14:58 | |
| *** apearson__ has joined #openstack-powervm | 15:00 | |
| *** apearson__ has quit IRC | 15:01 | |
| *** apearson__ has joined #openstack-powervm | 15:03 | |
| -openstackstatus- NOTICE: The Gerrit service on review.openstack.org is restarting quickly to relieve resource pressure and restore normal performance | 15:22 | |
| *** k0da has quit IRC | 15:29 | |
| thorst | efried: did you want to chat in here about the vif issue? | 16:12 |
| efried | thorst, yes please. | 16:13 |
| thorst | so what'd you find out | 16:13 |
| efried | We have a CNA *and* a vNIC. | 16:13 |
| efried | Are we supposed to have a CNA? | 16:13 |
| efried | It has a different MAC than the vNIC. | 16:13 |
| efried | And that's the guy that's getting identified by PlugVifs. | 16:13 |
| thorst | ahhhh ha! | 16:13 |
| thorst | OK - I got it | 16:13 |
| thorst | I understand the issue | 16:14 |
| efried | So the actual VIF shows up as "new" because... yeah. | 16:14 |
| thorst | open a bug. | 16:14 |
| thorst | I'll work on it | 16:14 |
| efried | I can nail it if you want. | 16:14 |
| efried | I at least want to talk through the solution. | 16:14 |
| thorst | either way | 16:14 |
| thorst | I'll let you | 16:14 |
| efried | if vif['vnic_type'] == 'direct', get vNICs instead of CNAs? | 16:14 |
| efried | Is that the right answer? | 16:14 |
| thorst | sec...pulling up code | 16:14 |
| efried | Not sure what the return value is used for. | 16:14 |
| thorst | efried: so we always need get_cnas | 16:16 |
| thorst | because the gmt vifs | 16:16 |
| thorst | I think tasks/network.py line 123 (start of execute_impl) | 16:16 |
| thorst | we could loop through the network_infos (which is just a list of vif objects) | 16:17 |
| thorst | and if one of those is type vNic, get the vnic_w_list... | 16:17 |
| thorst | to build the crt/update network_infos | 16:17 |
| efried | okay. | 16:18 |
| efried | thorst, 3900! | 16:21 |
| efried | "pretty" please | 16:21 |
| thorst | I assume there is some pun behind that | 16:22 |
| *** esberglu has quit IRC | 16:31 | |
| efried | thorst, can we assume the PlugVifs Task is single-use? Can I cache the CNAs/vNICs as instance vars? | 16:38 |
| efried | Seems so, cause we're doing it for the update/crt network infos | 16:38 |
| efried | just want to double chec. | 16:38 |
| efried | k | 16:38 |
| *** apearson__ has quit IRC | 16:57 | |
| efried | thorst, also, the comment at the end of execute_impl implies that we're returning the CNAs we *created*. In fact, we're only returning those that already existed before. Is that correct? | 16:57 |
| *** apearson__ has joined #openstack-powervm | 17:00 | |
| efried | thorst, it's used by PlugMgmtVif, which in fact probably needs *all* of the CNAs. This looks like a regression to me. Unless we're going to refactor all the plug methods to return the CNAs they're creating, the easiest thing to do would be to take vm_cnas out of PlugVif.provides and PlugMgmtVif.requires; and just have the latter do a full CNA.get | 17:04 |
| thorst | efried: Yes, single use. | 17:10 |
| thorst | efried: we should return all, and change wording | 17:10 |
| efried | rgr | 17:10 |
| efried | Separate bug? | 17:10 |
| thorst | I'm OK with that, or just doing in same | 17:11 |
| efried | thorst, Was planning to do it in same change set, since it's nigh inextricable; question was whether we need a separate launchpad bug (two Closes-Bug tags in same change set). | 17:13 |
| thorst | nah, just wrap it in | 17:13 |
| thorst | I'm ok with that | 17:13 |
| efried | thorst, will this code path ever run where self.instance is of type VIOS? | 17:17 |
| efried | (ye gods, I hope not) | 17:17 |
| thorst | efried: no | 17:22 |
| *** apearson__ has quit IRC | 17:26 | |
| *** tblakes has quit IRC | 17:28 | |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm: WIP: Create/update vNICs properly on Plug https://review.openstack.org/365078 | 17:37 |
| efried | thorst ^^ -- pvc impacts? | 17:38 |
| -openstackstatus- NOTICE: The infrastructure team is taking Gerrit offline for maintenance, beginning shortly after 18:00 UTC for a potentially 4 hour maintenance window. | 17:38 | |
| thorst | nahhh | 17:58 |
| thorst | efried: ^^ | 17:58 |
| efried | ight. Code is much cleaner now, IMO. | 17:59 |
| thorst | yeah... | 17:59 |
| efried | Well, "much" may be a stretch. But cleaner. | 17:59 |
| thorst | which reminds me...as I might not get to it | 17:59 |
| thorst | we should remove all the import logging statements and switch over to oslo_log | 17:59 |
| thorst | that may be a good one for kairo | 17:59 |
| thorst | I know he's been looking for some. | 17:59 |
| efried | Gosh, I thought we did that ages ago. | 18:00 |
| thorst | missed a few places in nova_powervm | 18:00 |
| thorst | probably specifically *I* missed a few places | 18:00 |
| efried | only four hits, yeah. | 18:00 |
| thorst | or reintroduced the bad bits | 18:00 |
| *** apearson__ has joined #openstack-powervm | 18:02 | |
| -openstackstatus- NOTICE: Gerrit is now going offline for maintenance, reserving a maintenance window through 22:00 UTC. | 18:05 | |
| *** ChanServ changes topic to "Gerrit is now going offline for maintenance, reserving a maintenance window through 22:00 UTC." | 18:05 | |
| efried | thorst, PlugVifs.revert - shouldn't this only unplug the VIFs we plugged? | 18:23 |
| thorst | yeah. | 18:23 |
| thorst | this is sounding more and more like a bug | 18:23 |
| adreznec | Gerrit's down, time to go home | 18:25 |
| efried | thorst, https://bugs.launchpad.net/nova-powervm/+bug/1619759 | 18:31 |
| openstack | Launchpad bug 1619759 in nova-powervm "PlugVifs revert should only unplug what it plugged" [Undecided,New] | 18:31 |
| efried | May be getting a bit above my pay grade, but willing to tackle if you want to walk me through that second question. | 18:31 |
| thorst | efried: isn't that just changing the unplug to only iterate through the 'crt_vifs' list? | 18:32 |
| efried | Look at the bug report. | 18:32 |
| thorst | o no...what're we getting at. | 18:33 |
| thorst | o nah...I don't think so. The retreament only occurs on nova-compute start up | 18:34 |
| *** k0da has joined #openstack-powervm | 18:37 | |
| *** kotra03 has quit IRC | 18:38 | |
| efried | thorst, my mistake - revert is doing the right thing. | 18:43 |
| thorst | phew | 18:43 |
| *** tblakes has joined #openstack-powervm | 18:44 | |
| *** ChanServ changes topic to "This channel is for PowerVM-related development and discussion. For general OpenStack support, please use #openstack." | 19:14 | |
| -openstackstatus- NOTICE: Gerrit is completing an online re-index, you may encounter slowness until it is complete | 19:14 | |
| *** miltonm has joined #openstack-powervm | 19:26 | |
| *** thorst has quit IRC | 19:36 | |
| *** thorst has joined #openstack-powervm | 19:37 | |
| *** thorst_ has joined #openstack-powervm | 19:39 | |
| *** thorst has quit IRC | 19:41 | |
| *** thorst_ has quit IRC | 19:43 | |
| *** kriskend_ has quit IRC | 19:55 | |
| *** seroyer has quit IRC | 19:57 | |
| *** seroyer has joined #openstack-powervm | 19:58 | |
| *** seroyer has quit IRC | 19:59 | |
| *** openstackgerrit has quit IRC | 20:04 | |
| *** openstackgerrit has joined #openstack-powervm | 20:04 | |
| *** tjakobs has quit IRC | 20:05 | |
| *** seroyer has joined #openstack-powervm | 20:14 | |
| *** tblakes has quit IRC | 20:32 | |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm: Create/update vNICs properly on Plug https://review.openstack.org/365078 | 20:39 |
| *** thorst has joined #openstack-powervm | 20:46 | |
| *** dwayne_ has quit IRC | 20:52 | |
| efried | thorst, yrt? | 20:53 |
| thorst | kinda | 20:53 |
| thorst | what's up | 20:53 |
| efried | possible for self.network_infos to be None? | 20:53 |
| thorst | yep | 20:53 |
| efried | okey. | 20:53 |
| thorst | well, maybe not none, but empty | 20:53 |
| efried | k, adding harmless safety. | 20:59 |
| efried | thorst, what's your availability going to be over the next few business days wrt code reviews? | 21:00 |
| thorst | unclear | 21:01 |
| thorst | I'll be on a plane for most of the next few days | 21:01 |
| efried | k. https://review.openstack.org/365078 is relatively urgent; the other not so much. | 21:02 |
| efried | didn't we implement something in power-off that wouldn't dump the huge glut of pypowervm cruft if the partition doesn't exist? | 21:03 |
| thorst | efried: I think it was something that esberglu was proposing to nova-powervm | 21:04 |
| *** tjakobs has joined #openstack-powervm | 21:06 | |
| efried | thorst, fyi, on .149, vnic create is taking 16-19s; delete 10-11s. | 21:08 |
| thorst | that's not great....but not the worst thing in the world | 21:08 |
| *** tjakobs has quit IRC | 21:13 | |
| efried | thorst, finally got done with those two reviews you gave me this morning (on the migration slot mappings for NPIV) | 21:32 |
| thorst | efried: saw...and yeah, they need work | 21:32 |
| thorst | or the core has a change it could make too | 21:33 |
| thorst | both options are..meh | 21:33 |
| efried | Okay. Well, it took me all day, but I think I get what you're doing here. | 21:33 |
| efried | Still some pending questions of design, though I think reviewing the community one answered my boggles in the pypowervm one. | 21:34 |
| efried | Clearly you're expecting the same number of VIOSes at the destination - was that always the expectation? | 21:34 |
| thorst | efried: it is now | 21:34 |
| thorst | not really just that, but that the same slots line up on a target VIOS | 21:36 |
| thorst | efried: 365078... | 21:38 |
| efried | yes | 21:38 |
| thorst | are you assuming that all the network_info are direct or sea? | 21:38 |
| thorst | nevermind...it doesn't look like you are | 21:39 |
| efried | I'm assuming anything that is 'direct' needs to look for VNIC; and anything else needs to look for CNA. | 21:39 |
| thorst | the _vif_exists confused me | 21:39 |
| efried | Did I misname that? | 21:39 |
| thorst | no, you're fine...I just confused myself | 21:39 |
| efried | I was using 'vif' as an umbrella for "CNA or VNIC (or possible future other thingy)" | 21:39 |
| *** k0da has quit IRC | 21:44 | |
| thorst | efried: just one thing I don't like | 21:44 |
| efried | Just one? I did well. | 21:44 |
| efried | Tell me. | 21:44 |
| thorst | you made the pipeline slower | 21:45 |
| thorst | I know why you did this | 21:45 |
| thorst | but I disagree it has to be done | 21:45 |
| efried | You mean because I'm doing a fresh CNA feed get in PlugMgmt? | 21:45 |
| thorst | indeed | 21:45 |
| efried | The alternative was to accumulate all the CNAs as I create them. | 21:46 |
| efried | Which would entail changing the spec of alllll the plug methods to return said CNAs. | 21:46 |
| efried | Which I could do. | 21:46 |
| efried | But a CNA feed get is really fast, isn't it? | 21:46 |
| efried | hence not worth the extra code & risk therein entailed | 21:47 |
| efried | That was my logic, anyway. | 21:47 |
| thorst | efried: to a VM...yeah, it can be | 21:49 |
| thorst | but why bother? | 21:49 |
| thorst | its like 3 lines less code | 21:50 |
| thorst | but a feed get otherwise | 21:50 |
| efried | sorry, you lost me. | 21:50 |
| thorst | its like half a second... | 21:50 |
| thorst | so I'd rather just write the three loc | 21:50 |
| thorst | back in a bit | 21:50 |
| *** tjakobs has joined #openstack-powervm | 21:55 | |
| efried | thorst, okay, I see, the leaf plug methods are already returning the vifs - it's just nova_powervm.virt.powervm.vif.plug that would need to return it back up. | 21:56 |
| *** tjakobs has quit IRC | 21:59 | |
| *** apearson__ has quit IRC | 22:14 | |
| *** tjakobs has joined #openstack-powervm | 22:16 | |
| *** tjakobs has quit IRC | 22:21 | |
| *** miltonm has quit IRC | 22:31 | |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm: Create/update vNICs properly on Plug https://review.openstack.org/365078 | 23:10 |
| efried | thorst ^^ (more than 3LOC, btw) | 23:11 |
| efried | tested, works. | 23:16 |
| efried | (though I didn't test it through all possible permutations, by any means) | 23:16 |
| efried | Signing off. Feel free to merge it if you're happy. | 23:17 |
| *** openstackgerrit has quit IRC | 23:18 | |
| *** openstackgerrit has joined #openstack-powervm | 23:19 | |
| thorst | efried: LGTM | 23:28 |
| *** thorst has quit IRC | 23:30 | |
| *** thorst has joined #openstack-powervm | 23:32 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!