| *** thorst_ has joined #openstack-powervm | 00:44 | |
| *** seroyer has joined #openstack-powervm | 00:45 | |
| *** viclarson has quit IRC | 00:47 | |
| *** apearson has joined #openstack-powervm | 01:05 | |
| *** dwayne has quit IRC | 01:07 | |
| *** dwayne has joined #openstack-powervm | 01:08 | |
| *** thorst_ has quit IRC | 02:25 | |
| *** thorst_ has joined #openstack-powervm | 02:26 | |
| *** thorst_ has quit IRC | 02:35 | |
| *** kylek3h has quit IRC | 02:37 | |
| *** kylek3h has joined #openstack-powervm | 02:39 | |
| *** thorst_ has joined #openstack-powervm | 02:47 | |
| *** thorst_ has quit IRC | 02:47 | |
| *** thorst_ has joined #openstack-powervm | 02:48 | |
| *** seroyer has quit IRC | 02:49 | |
| *** thorst_ has quit IRC | 02:53 | |
| *** tjakobs has quit IRC | 03:20 | |
| *** tjakobs has joined #openstack-powervm | 03:36 | |
| *** tjakobs has quit IRC | 03:41 | |
| *** thorst_ has joined #openstack-powervm | 03:51 | |
| *** thorst_ has quit IRC | 03:58 | |
| *** apearson has quit IRC | 04:01 | |
| *** shyama has joined #openstack-powervm | 04:36 | |
| *** thorst_ has joined #openstack-powervm | 04:55 | |
| *** thorst_ has quit IRC | 05:02 | |
| *** thorst_ has joined #openstack-powervm | 06:00 | |
| *** thorst_ has quit IRC | 06:07 | |
| *** k0da has joined #openstack-powervm | 06:07 | |
| *** adi___ has quit IRC | 06:19 | |
| *** adi___ has joined #openstack-powervm | 06:21 | |
| *** k0da has quit IRC | 06:59 | |
| *** thorst_ has joined #openstack-powervm | 07:05 | |
| *** thorst_ has quit IRC | 07:12 | |
| *** thorst_ has joined #openstack-powervm | 08:10 | |
| *** thorst_ has quit IRC | 08:17 | |
| *** k0da has joined #openstack-powervm | 08:38 | |
| *** thorst_ has joined #openstack-powervm | 09:14 | |
| *** thorst_ has quit IRC | 09:22 | |
| *** thorst_ has joined #openstack-powervm | 09:50 | |
| *** thorst_ has quit IRC | 09:58 | |
| *** thorst_ has joined #openstack-powervm | 10:54 | |
| *** thorst_ has quit IRC | 11:02 | |
| *** miltonm has quit IRC | 11:03 | |
| *** seroyer has joined #openstack-powervm | 11:21 | |
| *** thorst_ has joined #openstack-powervm | 11:45 | |
| *** miltonm has joined #openstack-powervm | 11:46 | |
| efried | thorst_, see jichenjc's comment on the blueprint? | 11:50 |
|---|---|---|
| efried | You going to respond to that or want me to? | 11:50 |
| thorst_ | efried: just got in...haven't seen any comments. Struggling with IBM's WiFi certificate expiration code... | 11:51 |
| thorst_ | so I'll look in a bit and then respond. | 11:51 |
| efried | Ugh. I gave up on IBM wifi months ago. | 11:51 |
| thorst_ | well, I'm on a mac without an ethernet port. | 11:52 |
| efried | 0_0 | 11:54 |
| *** thorst__ has joined #openstack-powervm | 12:00 | |
| *** thorst_ has quit IRC | 12:04 | |
| *** thorst__ has quit IRC | 12:05 | |
| *** thorst_ has joined #openstack-powervm | 12:06 | |
| *** thorst__ has joined #openstack-powervm | 12:07 | |
| *** thorst_ has quit IRC | 12:11 | |
| *** thorst__ has quit IRC | 12:12 | |
| *** thorst_ has joined #openstack-powervm | 12:13 | |
| *** thorst__ has joined #openstack-powervm | 12:14 | |
| *** thorst_ has quit IRC | 12:17 | |
| *** tblakes has joined #openstack-powervm | 12:19 | |
| *** thorst__ has quit IRC | 12:19 | |
| *** seroyer has quit IRC | 12:24 | |
| *** edmondsw has joined #openstack-powervm | 12:26 | |
| *** esberglu has joined #openstack-powervm | 12:29 | |
| *** dwayne has quit IRC | 12:52 | |
| *** burgerk has joined #openstack-powervm | 12:57 | |
| *** apearson has joined #openstack-powervm | 12:58 | |
| *** seroyer has joined #openstack-powervm | 13:00 | |
| *** thorst_ has joined #openstack-powervm | 13:02 | |
| *** mdrabe has joined #openstack-powervm | 13:04 | |
| *** apearson has quit IRC | 13:14 | |
| *** seroyer has quit IRC | 13:14 | |
| *** apearson has joined #openstack-powervm | 13:14 | |
| *** tjakobs has joined #openstack-powervm | 13:21 | |
| thorst_ | efried: I think your actual change set is what made your run a little faster than the new pypowervm patch. | 13:24 |
| thorst_ | though I talked to esberglu today, and he noted that during low periods, the time can get under an hour | 13:24 |
| efried | Heh. Wohoo! | 13:25 |
| efried | This was 5pm last night; wouldn't have thought that was an especially quiet time. | 13:25 |
| efried | But cool cool. | 13:25 |
| thorst_ | so I do think you want to look at those errors | 13:25 |
| thorst_ | see if they're related. | 13:25 |
| efried | So esberglu hasn't integrated the local2remote change yet? | 13:25 |
| thorst_ | efried: not in the production CI | 13:25 |
| efried | Yeah, I'm going to do that. | 13:25 |
| efried | okay. | 13:25 |
| *** dwayne has joined #openstack-powervm | 13:26 | |
| esberglu | efried: That patch worked fine in staging CI. Didn’t see a noticable time decrease though | 13:27 |
| efried | Bah, okay. It's possible everything's using just the one Session. | 13:28 |
| thorst_ | well, each driver process uses one session, maybe 2 for events | 13:28 |
| thorst_ | but doesn't get new sessions often. | 13:28 |
| efried | Also, the bootstrap would go quicker if the first address it tried was the one that worked. | 13:29 |
| efried | On the system you gave me yesterday, the first one was a local IP, so it had to ping-fail before it tried the second one. | 13:29 |
| efried | So the bootstrap process took like 10s. | 13:29 |
| efried | Oh well; it's an improvement regardless. It'll sure make your pvmctl commands quicker. | 13:30 |
| efried | Cause each one of those gets a fresh Session. | 13:30 |
| *** seroyer has joined #openstack-powervm | 13:33 | |
| thorst_ | :-) | 13:33 |
| *** k0da has quit IRC | 13:34 | |
| *** burgerk has quit IRC | 13:45 | |
| *** thorst__ has joined #openstack-powervm | 13:46 | |
| *** thorst_ has quit IRC | 13:49 | |
| *** k0da has joined #openstack-powervm | 13:53 | |
| *** viclarson_ has joined #openstack-powervm | 13:54 | |
| viclarson_ | has anyone tutorial "how to prepare image for openstack and powervm"? | 13:55 |
| thorst__ | burgerk is supposed to have one | 13:55 |
| thorst__ | let me track him down | 13:55 |
| *** thorst__ is now known as thorst_ | 13:59 | |
| thorst_ | nova IRC meeting now in #openstack-meeting - FYI | 14:00 |
| *** apearson has quit IRC | 14:01 | |
| *** kriskend has joined #openstack-powervm | 14:12 | |
| *** burgerk has joined #openstack-powervm | 14:13 | |
| *** _milky_ has joined #openstack-powervm | 14:18 | |
| *** _milky_ has quit IRC | 14:19 | |
| thorst_ | efried edmondsw esberglu: updated the BP again...needed a launchpad link | 14:28 |
| edmondsw | do you want another +1, or does it matter? | 14:28 |
| efried | I did it. | 14:30 |
| *** shyama has quit IRC | 14:43 | |
| *** AndyWojo has quit IRC | 14:55 | |
| *** mdrabe has quit IRC | 15:24 | |
| *** dwayne has quit IRC | 15:30 | |
| *** mdrabe has joined #openstack-powervm | 15:31 | |
| *** dwayne has joined #openstack-powervm | 15:34 | |
| thorst_ | efried: had a few nits on 378000. Looks good to me overall once we verify the CI issue is not related to the change set. | 15:38 |
| thorst_ | and given what I heard from btang, I'm strongly considering it for backport if we attach a bug to it | 15:38 |
| thorst_ | due to high cpu load otherwise. | 15:38 |
| efried | okay | 15:39 |
| thorst_ | which is a scary thought to backport something that big...but I'm now sold on the advantages of it | 15:40 |
| *** k0da has quit IRC | 15:40 | |
| *** AndyWojo has joined #openstack-powervm | 15:43 | |
| *** apearson has joined #openstack-powervm | 15:47 | |
| *** dwayne has quit IRC | 15:56 | |
| openstackgerrit | Eric Fried proposed openstack/networking-powervm: Refactor, consolidate, and clean up agents https://review.openstack.org/378000 | 15:57 |
| efried | thorst_ ^^ | 15:57 |
| efried | esberglu, see ^^ for CI failures. Have we seen this before? Looks like a VIOS bug to me. | 15:57 |
| efried | Does the CI capture VIOS snaps when failures happen? | 15:58 |
| thorst_ | link to logs? You just reproposed..so which are you looking at? | 15:58 |
| *** mdrabe has quit IRC | 16:01 | |
| *** mdrabe has joined #openstack-powervm | 16:01 | |
| *** dwayne has joined #openstack-powervm | 16:14 | |
| burgerk | viclarson_: you had a question about creating an image ? | 16:16 |
| viclarson_ | yes | 16:18 |
| viclarson_ | how to? | 16:18 |
| viclarson_ | have you tutorial? | 16:18 |
| efried | thorst_, sorry: http://184.172.12.213/00/378000/22/check/networking-powervm-pvm-dsvm-tempest-full/2a5b172/ | 16:32 |
| efried | Looking forward to seeing if latest patch set has similar problems. | 16:32 |
| efried | But I imagine we would see them everywhere. | 16:32 |
| *** shyama has joined #openstack-powervm | 16:45 | |
| thorst_ | efried: I agree with your analysis on those files... | 17:15 |
| thorst_ | but scary that those VIOS issues are poking through | 17:15 |
| efried | thorst_, agree. | 17:26 |
| thorst_ | good news is...events seem to work | 17:26 |
| thorst_ | and that change set...totally shaved 5-10 minutes off the run time. | 17:27 |
| thorst_ | "Task plug_vifs completed in 0 seconds for instance pvm3-tempest-Ima-18de1ca0[00m" | 17:27 |
| thorst_ | color me impressed...that's awesome. arnoldje will be shocked. | 17:28 |
| thorst_ | almost all of them are 0-ish seconds | 17:28 |
| *** tblakes has quit IRC | 17:34 | |
| *** tjakobs has quit IRC | 17:53 | |
| *** tjakobs has joined #openstack-powervm | 18:08 | |
| *** viclarson_ has quit IRC | 18:09 | |
| *** tblakes has joined #openstack-powervm | 18:15 | |
| *** tjakobs has quit IRC | 18:33 | |
| thorst_ | burgerk: is there a way to tell cloud init NOT to do the networking...just use whats in the /etc/network/interfaces file (a DHCP) | 18:41 |
| burgerk | applying network is in the main path, not a separate module, so not really. So you want it to ignore what is in the ConfigDrive ? | 18:43 |
| burgerk | or just the network info in the ConfigDrive? | 18:44 |
| burgerk | thorst_ ^ | 18:45 |
| thorst_ | I guess I just did what viclarson did...I just changed the interfaces.template to be exactly what I wanted. | 18:47 |
| thorst_ | so I wouldn't get stuck...as this problem I'm debugging requires many VMs to be deployed | 18:48 |
| thorst_ | seroyer: I think I have it figured out | 18:57 |
| thorst_ | what a mess... | 18:57 |
| thorst_ | can I give you a call? | 18:57 |
| seroyer | thorst_, in a bit. In a meeting. | 18:58 |
| thorst_ | seroyer: OK - net is, I think your side is fine. | 18:58 |
| seroyer | Ok. Should I remove my test hack then? | 18:59 |
| thorst_ | yeah...net is its the overlay and the routing table in the network node's router | 19:00 |
| thorst_ | so for physical networks...we're actually good. Overlay's this affects every hypervisor. | 19:01 |
| *** k0da has joined #openstack-powervm | 19:04 | |
| *** kriskend_ has joined #openstack-powervm | 19:08 | |
| *** kriskend has quit IRC | 19:09 | |
| esberglu | 3.0.0 newton releases have been tagged for nova-powervm, ceilometer-powervm and networking-powervm | 19:13 |
| thorst_ | woooooooooooooooooooooooooooooo | 19:13 |
| thorst_ | good job team! | 19:13 |
| seroyer | thorst_ Sorry, complete failure to parse your response. | 19:18 |
| thorst_ | the packet gets sent from the migration destination host. The response goes to the network node. Neutron still thinks its on the source (cause the port 'host' is updated on completion of the live migration). Neutron's router sends the ping response to the source host. | 19:19 |
| seroyer | Oh. So will there be any value to our enhancment? | 19:21 |
| thorst_ | yes. | 19:23 |
| thorst_ | seroyer: Flat & VLAN networks will now work | 19:23 |
| thorst_ | things that don't require on a network node basically. | 19:24 |
| thorst_ | and we may want to open a bug to neutron on this... | 19:24 |
| thorst_ | and by we...I mean me... | 19:24 |
| thorst_ | I wonder if geneve networks are affected...since the packet has more metadata... | 19:26 |
| seroyer | Clearly we should try it! | 19:26 |
| thorst_ | yeah, its actually easy to test... | 19:26 |
| thorst_ | man...we've been working on that for days... | 19:27 |
| seroyer | Are you ready for me to remove my hack? | 19:29 |
| *** apearson has quit IRC | 19:29 | |
| *** apearson has joined #openstack-powervm | 19:34 | |
| thorst_ | seroyer: your hack being not having the block during live migration? | 19:48 |
| seroyer | Right. | 19:48 |
| *** apearson has quit IRC | 19:48 | |
| thorst_ | yeah, lets get rid of that. | 19:48 |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm: Rebase on set_vnic_backdevs(redundancy) https://review.openstack.org/383428 | 19:53 |
| *** apearson has joined #openstack-powervm | 19:54 | |
| efried | thorst_ ^^ Would like to get that back to newton so we don't have to support min/max forever. | 19:54 |
| efried | We haven't tagged a release of pypowervm lately, have we? | 19:54 |
| thorst_ | agree... | 19:54 |
| thorst_ | needs a bug | 19:54 |
| efried | Sho | 19:55 |
| efried | 1631131 | 19:58 |
| efried | esberglu thorst: latest CI run on 378000 didn't have those cfg_drive VIOS failures - but did have the "multiple networks" failures. | 20:11 |
| thorst_ | efried: yeah, esberglu has been working on that multiple networks failures for a while | 20:14 |
| esberglu | Problem right now is that the staging env. is currently occupied with OSA work so I haven’t had a chance to try it, but talking with thorst_ earlier we think we may have a solution | 20:17 |
| thorst_ | esberglu: remind me what that solution was? Make them both public? | 20:17 |
| esberglu | Yeah that’s what you said. Because the tempest vms couldn’t find the network named private | 20:18 |
| esberglu | and then didn’t know which one to choose from the other networks | 20:19 |
| openstackgerrit | Shyama proposed openstack/nova-powervm: SSP Volume Adapter https://review.openstack.org/372254 | 20:21 |
| thorst_ | right right...then tenancy thing | 20:22 |
| esberglu | thorst_: Another reason switching to creating the networks after the stack will be nice. Easier to do that when creating the network than trying to do it through devstack config options | 20:25 |
| thorst_ | esberglu: yeah, can be ripped from the under cloud's setup | 20:27 |
| thorst_ | as it does that now | 20:27 |
| *** shyama has quit IRC | 20:28 | |
| *** kriskend_ has quit IRC | 20:42 | |
| *** burgerk has quit IRC | 21:08 | |
| *** seroyer has quit IRC | 21:11 | |
| *** thorst_ has quit IRC | 21:27 | |
| *** thorst_ has joined #openstack-powervm | 21:27 | |
| *** seroyer has joined #openstack-powervm | 21:28 | |
| *** thorst__ has joined #openstack-powervm | 21:29 | |
| *** thorst_ has quit IRC | 21:32 | |
| *** thorst__ has quit IRC | 21:34 | |
| *** esberglu has quit IRC | 21:37 | |
| *** tblakes has quit IRC | 21:51 | |
| *** mdrabe has quit IRC | 22:04 | |
| *** apearson has quit IRC | 22:28 | |
| *** seroyer has quit IRC | 22:45 | |
| *** seroyer has joined #openstack-powervm | 22:49 | |
| *** seroyer has quit IRC | 22:49 | |
| *** k0da has quit IRC | 22:59 | |
| *** svenkat has joined #openstack-powervm | 23:09 | |
| *** esberglu has joined #openstack-powervm | 23:24 | |
| *** esberglu has quit IRC | 23:35 | |
| *** svenkat has quit IRC | 23:44 | |
| *** esberglu has joined #openstack-powervm | 23:51 | |
| *** esberglu has quit IRC | 23:51 | |
| *** thorst_ has joined #openstack-powervm | 23:52 | |
| *** thorst_ has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!