| *** rajinir has quit IRC | 00:15 | |
| *** rama_y has quit IRC | 00:16 | |
| *** rama_y has joined #openstack-ironic | 00:16 | |
| *** hoangcx has joined #openstack-ironic | 00:32 | |
| *** wajdi has joined #openstack-ironic | 00:33 | |
| *** anush_ has quit IRC | 00:34 | |
| *** vishwanathj has joined #openstack-ironic | 00:34 | |
| openstackgerrit | Merged openstack/ironic: DevStack: No need to change the ramdisk filesystem type https://review.openstack.org/355390 | 00:48 |
|---|---|---|
| openstackgerrit | Merged openstack/ironic: Allow changing lock purpose on lock upgrade https://review.openstack.org/356064 | 00:48 |
| *** rloo has quit IRC | 01:08 | |
| *** baoli has joined #openstack-ironic | 01:14 | |
| *** chlong has joined #openstack-ironic | 01:24 | |
| openstackgerrit | Merged openstack/ironic: Fix typo: interations -> iterations. https://review.openstack.org/357437 | 01:26 |
| *** zhangjl has joined #openstack-ironic | 01:27 | |
| *** moshele has joined #openstack-ironic | 01:30 | |
| openstackgerrit | Tang Chen proposed openstack/ironic: Split v1 API root unit and functional tests https://review.openstack.org/348965 | 01:35 |
| openstackgerrit | Tang Chen proposed openstack/ironic: Move root controller tests to functional dir https://review.openstack.org/356188 | 01:35 |
| openstackgerrit | Tang Chen proposed openstack/ironic: Move api driver tests to functional dir https://review.openstack.org/350365 | 01:35 |
| openstackgerrit | Tang Chen proposed openstack/ironic: Move port api tests to functional dir https://review.openstack.org/356287 | 01:35 |
| openstackgerrit | Tang Chen proposed openstack/ironic: Prepare for functional tests https://review.openstack.org/349421 | 01:35 |
| *** rbudden has quit IRC | 01:38 | |
| *** mtanino has quit IRC | 01:42 | |
| openstackgerrit | Naohiro Tamura proposed openstack/python-ironicclient: Add soft reboot/poweroff power states. https://review.openstack.org/247904 | 01:48 |
| *** syed_ has quit IRC | 01:52 | |
| *** causten has quit IRC | 02:02 | |
| *** sabeen has quit IRC | 02:03 | |
| *** sabeen has joined #openstack-ironic | 02:05 | |
| *** aNupoisc has quit IRC | 02:10 | |
| *** chlong has quit IRC | 02:29 | |
| *** adreznec has left #openstack-ironic | 02:39 | |
| *** chlong has joined #openstack-ironic | 02:46 | |
| *** yuanying has quit IRC | 02:56 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: DNM: testing config-drive default d-s-g patch https://review.openstack.org/357582 | 02:58 |
| *** yuanying has joined #openstack-ironic | 02:58 | |
| *** Sukhdev has quit IRC | 03:03 | |
| *** sabeen has quit IRC | 03:05 | |
| *** adreznec has joined #openstack-ironic | 03:05 | |
| *** sabeen has joined #openstack-ironic | 03:07 | |
| *** yuanying has quit IRC | 03:22 | |
| *** baoli has quit IRC | 03:35 | |
| *** zackf has joined #openstack-ironic | 03:37 | |
| *** chlong has quit IRC | 03:39 | |
| *** wajdi has quit IRC | 03:46 | |
| *** wajdi has joined #openstack-ironic | 03:46 | |
| *** wajdi_ has joined #openstack-ironic | 03:47 | |
| *** wajdi has quit IRC | 03:51 | |
| *** chlong has joined #openstack-ironic | 03:52 | |
| *** yuanying has joined #openstack-ironic | 03:54 | |
| zackf | Is there a good way to have nova handle both baremetal and VM requests? | 03:55 |
| zackf | In mitaka* | 03:56 |
| *** fragatina has quit IRC | 03:56 | |
| *** wajdi has joined #openstack-ironic | 03:57 | |
| *** wajdi_ has quit IRC | 04:01 | |
| *** wajdi has quit IRC | 04:02 | |
| openstackgerrit | Merged openstack/ironic: Always set DEFAULT/host in devstack. https://review.openstack.org/357154 | 04:03 |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: DO NOT REVIEW https://review.openstack.org/296432 | 04:06 |
| *** chlong has quit IRC | 04:10 | |
| openstackgerrit | Nguyen Hung Phuong proposed openstack/ironic: Merge imports in code https://review.openstack.org/357597 | 04:14 |
| *** links has joined #openstack-ironic | 04:15 | |
| *** chlong has joined #openstack-ironic | 04:23 | |
| *** tiendc has joined #openstack-ironic | 04:28 | |
| openstackgerrit | Merged openstack/ironic: Centralize config options - [pxe] https://review.openstack.org/309612 | 04:31 |
| openstackgerrit | Merged openstack/ironic: Fix key word argument interface_type -> interface https://review.openstack.org/355537 | 04:31 |
| *** dims has quit IRC | 04:38 | |
| *** dims has joined #openstack-ironic | 04:43 | |
| *** Sukhdev has joined #openstack-ironic | 04:47 | |
| *** bks has joined #openstack-ironic | 05:06 | |
| *** tangchen has quit IRC | 05:08 | |
| *** bks has quit IRC | 05:09 | |
| *** bks has joined #openstack-ironic | 05:10 | |
| *** sabeen has quit IRC | 05:10 | |
| *** hparekh has left #openstack-ironic | 05:11 | |
| *** tiendc has quit IRC | 05:11 | |
| *** chlong has quit IRC | 05:12 | |
| *** tangchen has joined #openstack-ironic | 05:22 | |
| *** sdake_ has quit IRC | 05:24 | |
| *** chlong has joined #openstack-ironic | 05:26 | |
| *** tiendc has joined #openstack-ironic | 05:30 | |
| *** moshele has quit IRC | 05:38 | |
| *** moshele has joined #openstack-ironic | 05:39 | |
| *** zackf has quit IRC | 05:44 | |
| *** moshele has quit IRC | 05:52 | |
| *** chlong has quit IRC | 05:52 | |
| *** e0ne has joined #openstack-ironic | 05:59 | |
| *** sdake has joined #openstack-ironic | 05:59 | |
| *** rama_y has quit IRC | 06:01 | |
| *** bks has quit IRC | 06:04 | |
| *** kumarb has joined #openstack-ironic | 06:04 | |
| *** kumarb has quit IRC | 06:05 | |
| *** kumarb has joined #openstack-ironic | 06:05 | |
| *** zhangjl1 has joined #openstack-ironic | 06:06 | |
| *** e0ne has quit IRC | 06:07 | |
| *** e0ne has joined #openstack-ironic | 06:07 | |
| *** zhangjl has quit IRC | 06:08 | |
| *** sdake has quit IRC | 06:08 | |
| *** e0ne_ has joined #openstack-ironic | 06:14 | |
| *** bks has joined #openstack-ironic | 06:15 | |
| *** e0ne has quit IRC | 06:15 | |
| *** fragatina has joined #openstack-ironic | 06:15 | |
| *** bks has quit IRC | 06:16 | |
| *** zhangjl has joined #openstack-ironic | 06:16 | |
| *** zhangjl1 has quit IRC | 06:16 | |
| *** fragatina has quit IRC | 06:16 | |
| *** bks has joined #openstack-ironic | 06:17 | |
| *** kumarb has quit IRC | 06:17 | |
| *** fragatina has joined #openstack-ironic | 06:17 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: DO NOT REVIEW https://review.openstack.org/296432 | 06:19 |
| *** milan has joined #openstack-ironic | 06:20 | |
| *** bks has quit IRC | 06:21 | |
| *** chlong has joined #openstack-ironic | 06:21 | |
| milan | morning Ironic! :) | 06:22 |
| *** e0ne_ has quit IRC | 06:24 | |
| openstackgerrit | Naohiro Tamura proposed openstack/python-ironicclient: Support soft reboot and soft power off with timeout for OSC https://review.openstack.org/357627 | 06:29 |
| *** _vsaienko has joined #openstack-ironic | 06:32 | |
| *** _vsaienko has quit IRC | 06:40 | |
| *** rcernin has joined #openstack-ironic | 06:41 | |
| *** _vsaienko has joined #openstack-ironic | 06:45 | |
| *** ccamacho has joined #openstack-ironic | 06:46 | |
| openstackgerrit | Tang Chen proposed openstack/python-ironicclient: Support i18n for osc plugin help msg https://review.openstack.org/357639 | 06:49 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: DO NOT REVIEW https://review.openstack.org/296432 | 06:51 |
| *** tiendc has quit IRC | 06:53 | |
| *** _vsaienko has quit IRC | 06:54 | |
| *** _vsaienko has joined #openstack-ironic | 06:54 | |
| *** e0ne has joined #openstack-ironic | 06:55 | |
| *** _vsaienko has quit IRC | 06:55 | |
| *** tesseract- has joined #openstack-ironic | 06:56 | |
| *** wajdi has joined #openstack-ironic | 07:04 | |
| *** Sukhdev has quit IRC | 07:05 | |
| *** tiendc has joined #openstack-ironic | 07:05 | |
| *** hoangcx has quit IRC | 07:08 | |
| *** jpich has joined #openstack-ironic | 07:25 | |
| *** foexle has joined #openstack-ironic | 07:47 | |
| *** vishwanathj has quit IRC | 07:51 | |
| *** dtantsur|afk is now known as dtantsur | 07:52 | |
| dtantsur | Morning Ironic | 07:52 |
| *** amotoki has joined #openstack-ironic | 07:53 | |
| *** hoangcx has joined #openstack-ironic | 07:56 | |
| *** amotoki_ has joined #openstack-ironic | 07:57 | |
| tiendc | morning dtantsur | 07:59 |
| *** amotoki has quit IRC | 07:59 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-ironic | 08:00 | |
| tiendc | dtantsur: I'm waiting for you | 08:00 |
| tiendc | dtantsur: can you help review the nova patch https://review.openstack.org/#/c/328157/ | 08:00 |
| dtantsur | tiendc, I'll try to get it to it today. note, however, that I don't have +2 rights on nova | 08:01 |
| tiendc | dtantsur: It's no problem, with your +1 and Ruby's +1, I will contact nova core reviewers then | 08:02 |
| tiendc | dtantsur: thanks | 08:02 |
| *** openstackgerrit has quit IRC | 08:03 | |
| *** openstackgerrit has joined #openstack-ironic | 08:04 | |
| *** tiendc has quit IRC | 08:09 | |
| *** ohamada has joined #openstack-ironic | 08:11 | |
| *** vishwanathj has joined #openstack-ironic | 08:14 | |
| *** e0ne has quit IRC | 08:14 | |
| milan | morning dtantsur , happy Friday! How about a review of the state patch? O:-) | 08:15 |
| dtantsur | hey, lemme go through email backlog first :) | 08:15 |
| milan | sure :) | 08:16 |
| *** lucas-dinner is now known as lucasagomes | 08:18 | |
| openstackgerrit | Yuiko Takada Mori proposed openstack/ironic: Add node serial console documentation https://review.openstack.org/293872 | 08:22 |
| openstackgerrit | Naohiro Tamura proposed openstack/python-ironicclient: Add a new OSC command for Inject NMI https://review.openstack.org/357661 | 08:22 |
| *** athomas has joined #openstack-ironic | 08:29 | |
| *** moshele has joined #openstack-ironic | 08:29 | |
| *** Goneri has joined #openstack-ironic | 08:33 | |
| *** derekh has joined #openstack-ironic | 08:34 | |
| lucasagomes | morning all | 08:36 |
| milan | morning lucasagomes! :) | 08:39 |
| dtantsur | morning lucasagomes | 08:39 |
| dtantsur | folks, do we allow setting cleaning network UUID via node driver_info or whatever? | 08:40 |
| dtantsur | jroll, lucasagomes ^^^ | 08:41 |
| lucasagomes | dtantsur, afaict, no | 08:42 |
| lucasagomes | only in the config file | 08:42 |
| tangchen | morning. :) | 08:42 |
| dtantsur | hmm, I guess I have to fix it... | 08:42 |
| lucasagomes | dtantsur, what's up? | 08:42 |
| milan | morning tangchen! :) | 08:42 |
| dtantsur | lucasagomes, well, in tripleo we install and configure services much before any objects (networks, etc) are created | 08:42 |
| tangchen | morning milan. :) | 08:43 |
| dtantsur | I'd really love to avoid users having to call stack-update just to set one configuration option... | 08:43 |
| lucasagomes | dtantsur, right | 08:43 |
| lucasagomes | dtantsur, is that possible to just change the order of the things? | 08:43 |
| *** hoonetorg has quit IRC | 08:43 | |
| dtantsur | and it's not consistent with the fact that we force deploy_kernel/deploy_ramdisk on the node | 08:43 |
| dtantsur | lucasagomes, install ironic and neutron after creating a network? :) | 08:44 |
| dtantsur | I don't think it's possible | 08:44 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic: Allow using TempURLs for deploy images https://review.openstack.org/344160 | 08:44 |
| *** hparekh has joined #openstack-ironic | 08:44 | |
| *** hparekh_ has joined #openstack-ironic | 08:44 | |
| lucasagomes | dtantsur, the kernel/initrd is because nodes could have different architectures | 08:45 |
| lucasagomes | so it needs to be different per node | 08:45 |
| *** ohamada has quit IRC | 08:45 | |
| dtantsur | lucasagomes, yeah, but that's a corner case; I can just as well imagine people wanting different cleaning networks | 08:45 |
| *** hparekh has quit IRC | 08:45 | |
| *** hparekh_ has quit IRC | 08:45 | |
| *** ohamada has joined #openstack-ironic | 08:46 | |
| lucasagomes | yeah, I'm sure there are cases for it as well :D | 08:46 |
| * milan relocates | 08:48 | |
| *** milan has quit IRC | 08:48 | |
| *** jtomasek has joined #openstack-ironic | 08:48 | |
| openstackgerrit | Cao ShuFeng proposed openstack/ironic: Log node uuid rather than id when acquiring node lock https://review.openstack.org/355732 | 08:49 |
| dtantsur | lucasagomes, here I summarized my thoughts: https://bugs.launchpad.net/ironic/+bug/1614876 do you think it makes sense? | 08:49 |
| openstack | Launchpad bug 1614876 in Ironic "It is impossible to set cleaning_network_uuid per node" [Undecided,New] | 08:49 |
| * dtantsur is also not sure if it counts as RFE or not | 08:49 | |
| *** hoonetorg has joined #openstack-ironic | 08:53 | |
| tangchen | milan, lucasagomes, dtantsur: Hi, about this oneview cleanup patch, Thiago has approved it. So please help to approve it when you have time. :) | 08:54 |
| tangchen | https://review.openstack.org/#/c/347127/ | 08:54 |
| lucasagomes | tangchen, will take a look | 08:57 |
| lucasagomes | dtantsur, yeah I probably need more investigation to understand it better | 08:58 |
| tangchen | lucasagomes: Thanks a lot. :) | 08:58 |
| lucasagomes | tangchen, +2, I saw you submit a new patch-set based on a nit in the release note | 09:00 |
| lucasagomes | thanks for that, but sometimes a nit is just to point out something | 09:00 |
| lucasagomes | you don't need to resubmit it for it, specially when there's a +2 already :D | 09:00 |
| dtantsur | jroll, please let me know what you think about https://bugs.launchpad.net/ironic/+bug/1614876 when you're here | 09:00 |
| openstack | Launchpad bug 1614876 in Ironic "[RFE] Allow setting cleaning_network_uuid per node" [Wishlist,Confirmed] | 09:00 |
| * dtantsur converted it to an RFE | 09:00 | |
| tangchen | lucasagomes: Ok, I see. :) thanks | 09:01 |
| *** chlong has quit IRC | 09:03 | |
| *** electrofelix has joined #openstack-ironic | 09:04 | |
| tangchen | milan, dtantsur, lucasagomes: I found that in ironicclient, there is no detail doc for each command in doc/source/. | 09:12 |
| tangchen | milan, dtantsur, lucasagomes: In openstackclient, we have doc for each command so that users can refer to it on the website. | 09:13 |
| tangchen | milan, dtantsur, lucasagomes: Do we need to write doc for each command in ironicclient ? | 09:13 |
| lucasagomes | tangchen, that wouldn't be bad, but I think that ultimately we just want to use the OSC command for Ironic | 09:14 |
| tangchen | milan, dtantsur, lucasagomes: It is very simple. Just copy the help messages from each command, and write them in rst files. | 09:14 |
| *** vishwanathj is now known as vishwanathj_zzz | 09:15 | |
| tangchen | lucasagomes: Yes. ironic client acts as a osc plugin. But there is no doc for the "openstack baremetal node xxx" commands on the web site. | 09:15 |
| tangchen | lucasagomes: I mean we can do the same as osc. | 09:15 |
| eset | hey :) does any one knows what's the issue when I create image with disk-image-create and I get resousrce busy in /tmp/dib.build... | 09:16 |
| lucasagomes | eset, I don't, but you may want to ask it at #tripleo | 09:16 |
| eset | what's tripelo? | 09:17 |
| eset | tripleo* ? | 09:17 |
| tangchen | eset: Hi, tripleo means openstack on openstack, which is a tool to setup an openstack environment. | 09:18 |
| lucasagomes | tangchen, right, I believe that it's because Ironic is just a plugin for OSC ? | 09:18 |
| lucasagomes | tangchen, so I think we should document the OSC commands in ironic itself | 09:18 |
| tangchen | eset: You may also ask the question in #openstack-infra. | 09:18 |
| lucasagomes | eset, what tangchen said. And diskimage-build (dib) is maintained by that project | 09:19 |
| tangchen | lucasagomes: Yes. So I'll open a bug and start to write the doc. You will see patches soon. :) | 09:19 |
| eset | thx | 09:20 |
| lucasagomes | tangchen, thx | 09:20 |
| eset | I followed your advice | 09:20 |
| *** fumie has quit IRC | 09:21 | |
| openstackgerrit | Markos Chandras proposed openstack/bifrost: Fix /etc/hosts before starting the rabbitmq server https://review.openstack.org/357718 | 09:27 |
| *** e0ne has joined #openstack-ironic | 09:31 | |
| *** moshele has quit IRC | 09:36 | |
| *** jtomasek has quit IRC | 09:46 | |
| *** jtomasek has joined #openstack-ironic | 09:47 | |
| openstackgerrit | Ankit Kumar proposed openstack/proliantutils: Fix Inspection failure in Gen7 https://review.openstack.org/357134 | 09:50 |
| openstackgerrit | Tang Chen proposed openstack/python-ironicclient: Trivial: Fix doc string for class DeleteBaremetalNode https://review.openstack.org/357731 | 09:58 |
| openstackgerrit | Ankit Kumar proposed openstack/proliantutils: Fix Inspection failure in Gen7 https://review.openstack.org/357134 | 10:04 |
| *** jtomasek has quit IRC | 10:08 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 10:10 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 10:10 |
| *** sambetts|afk is now known as sambetts | 10:17 | |
| sambetts | Morning all | 10:17 |
| dtantsur | morning sambetts | 10:19 |
| sambetts | dtantsur: the reason for setting it globally is because in the old world it could only be one network because we couldn't plumb networks, in the new world we can have different networks for things like provisioning bcause we can program switches | 10:19 |
| sambetts | dtantsur: also by change the order i think lucasagomes means install neutron create the network then install and configure Ironic | 10:19 |
| sambetts | dtantsur: thats how its done in the devstack process | 10:20 |
| dtantsur | sambetts, well, it's not quite possible in tripleo (nor is it reasonable, I guess) | 10:22 |
| sambetts | dtantsur: you use heat templates right? and you can add dependencies to services if I remember correctly | 10:22 |
| sambetts | dtantsur: I think there are already puppet things to handle the order problem | 10:23 |
| dtantsur | tripleo does not create networks | 10:23 |
| dtantsur | it's up to a user, and I suspect it was a conscious decision | 10:23 |
| dtantsur | I'm not sure people will be fond of using puppet to create networks... | 10:24 |
| sambetts | I meant that the Openstack/Ironic deployment puppet projects already handle it, not that tripleo should use them nessesaryly :-P | 10:25 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironic-inspector-client: Deprecate global functions in favor of ClientV1 methods https://review.openstack.org/357743 | 10:26 |
| dtantsur | sambetts, accepting the cleaning network in driver_info is the easiest thing we can do, I think :) | 10:27 |
| sambetts | dtantsur: we'd need to do the same for provisioning_network too / it depends on the network_interface loaded | 10:28 |
| dtantsur | I'm fine with doing it for provisioning_network too | 10:28 |
| dtantsur | I personally don't find it great to put UUIDs in configuration files | 10:28 |
| dtantsur | it assumes that the person who installs ironic and the person who used has the same rights | 10:29 |
| sambetts | dtantsur: its a pain I agree | 10:29 |
| dtantsur | which was never true, and is even less true now that we have policy support | 10:29 |
| sambetts | dtantsur: would it fall back to the config file option? or would we deprecate it? | 10:29 |
| dtantsur | sambetts, I'm fine with fall back | 10:29 |
| *** milan has joined #openstack-ironic | 10:30 | |
| *** _vsaienko has joined #openstack-ironic | 10:31 | |
| sambetts | dtantsur: right now some of the interfaces fail conductor start if e.g. cleaning_uuid is missing, how would we handle that? | 10:31 |
| sambetts | dtantsur: (actually it might be that we skipped that at the moment with a warning that it will next cycle) | 10:31 |
| dtantsur | good question, I didn't notice that | 10:32 |
| dtantsur | yeah, I think we'll have to stop failing early, and fail e.g. validation instead | 10:33 |
| *** athomas has quit IRC | 10:33 | |
| dtantsur | sambetts, updated https://bugs.launchpad.net/ironic/+bug/1614876 with what we discussed | 10:35 |
| openstack | Launchpad bug 1614876 in Ironic "[RFE] Allow setting {provisioning,cleaning}_network_uuid in node driver_info" [Wishlist,Confirmed] | 10:35 |
| sambetts | dtantsur: awesome :) | 10:35 |
| *** athomas has joined #openstack-ironic | 10:38 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Increase verbosity for functional tests https://review.openstack.org/357753 | 10:44 |
| jroll | dtantsur: I'm not sure I agree with any of that RFE | 10:47 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: DO NOT REVIEW https://review.openstack.org/296432 | 10:47 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: functional: allow passing kwargs to all calls https://review.openstack.org/357763 | 10:48 |
| *** moshele has joined #openstack-ironic | 10:48 | |
| jroll | 1) the provisioning/cleaning networks are part of deployments, not usage. the "user" shouldn't care about them, the deployer should be handling it (because the probably know the network architecture) | 10:48 |
| dtantsur | jroll, well, networks are created by users | 10:49 |
| jroll | they can be created by deployers, too :) | 10:49 |
| dtantsur | no, they are created by users. a deployer can be a user, of course :) | 10:50 |
| jroll | but it isn't like someone is going to later say "oh, I think cleaning for this node should be on a different network" | 10:50 |
| dtantsur | but that does not change the logic separation: neutron networks are created by users, configuration files are edited by deployers | 10:50 |
| dtantsur | I can say the same for deploy_kernel :) | 10:50 |
| dtantsur | anyway, this is not the most important point | 10:51 |
| dtantsur | this separation violation that we have gives me huge pain with bringing ironic support in tripleo | 10:51 |
| dtantsur | because tripleo separates deploying stage very clearly (for good or for bad) | 10:51 |
| jroll | I don't think it's a violation, and I don't think tripleo's architecture is a reason to do much in ironic :) | 10:52 |
| jroll | only in tripleo is it difficult to update a config file with puppet | 10:52 |
| dtantsur | jroll, create a network based on user's desire, then update the file | 10:52 |
| dtantsur | we don't have networks before installing openstack :) | 10:52 |
| jroll | a neutron network is just a dependency | 10:53 |
| dtantsur | also yes it is violation. this requires going back and forth from user-accesible API to editing configuration files | 10:53 |
| jroll | anyway, even though I disagree that it's useful, I'm not against it because it falls back | 10:53 |
| jroll | I just think it's silly | 10:53 |
| jroll | and confusing | 10:54 |
| sambetts | jroll: in the new world different nodes could have different networks anyway | 10:54 |
| jroll | and we have too many per-node things | 10:54 |
| dtantsur | I am open to other ideas :) | 10:54 |
| jroll | sambetts: ? | 10:54 |
| dtantsur | I hate putting it (or deploy_kernel) to driver_info | 10:54 |
| sambetts | jroll: we configure the switch on the fly, so different nodes could have different provisioning networks | 10:54 |
| dtantsur | but I also hate creating networks in the middle of OpenStack installation | 10:55 |
| sambetts | jroll: if thats the way a deployer wants to do it | 10:55 |
| jroll | sambetts: oh yeah, it isn't a technical problem (except for flat/cleaning) | 10:55 |
| sambetts | jroll: just thinking it might be an advantage to allow that flexiblity for deployers | 10:55 |
| dtantsur | I wonder if with complex architectures like spine-and-leaf cleaning networks can be different even in flat case... | 10:56 |
| *** _vsaienko has quit IRC | 10:56 | |
| jroll | sambetts: possibly, I just find it dangerous to make things super configurable before someone has a use case | 10:57 |
| jroll | dtantsur: networks can have multiple subnets :) | 10:57 |
| dtantsur | true; I don't know neutron well enough to think more about this case :) | 10:57 |
| jroll | so even if they're different physical nets, because ironic needs connectivity, they can't be 100% separated | 10:57 |
| jroll | heh | 10:58 |
| *** _vsaienko has joined #openstack-ironic | 10:58 | |
| jroll | dtantsur: I also really don't like losing the startup validation | 10:59 |
| *** ppiela has quit IRC | 10:59 | |
| dtantsur | the picture I have in mind is "1. $INSTALLER_NAME installs openstack, 2. an admin users opens Horizon and starts configuring things, 3. she creates a network, 4.. wait but ironic already needs a network??' | 10:59 |
| dtantsur | and while TripleO is based on OpenStack and can maybe solve it, some installers might not, so creating networks might be an issue for them... | 10:59 |
| *** ppiela has joined #openstack-ironic | 11:00 | |
| jroll | I mean, here's the two options, right? | 11:00 |
| jroll | maybe 3 | 11:00 |
| jroll | 1) admin installs, creates network, adds to config file, re-deploys ironic config | 11:01 |
| jroll | 2) admin installs neutron, creates network, puts it in config, installs the rest | 11:01 |
| jroll | 3) admin installs all the things, creates network, runs node-update for *every* node | 11:01 |
| jroll | in what world is (3) more sane? | 11:01 |
| dtantsur | s/node-update/node-create/ | 11:01 |
| jroll | sure | 11:01 |
| dtantsur | in the same world where setting the same deploy_kernel is sane ;) (sorry, I'm really annoyed by it, so bear with me :) | 11:02 |
| *** Sylvester has joined #openstack-ironic | 11:02 | |
| jroll | dtantsur: that was before I worked on ironic, I also think it's terrible | 11:02 |
| jroll | don't use things you hate as a reason to do something :) | 11:02 |
| dtantsur | true, this is not the reason | 11:02 |
| dtantsur | the reason is that I'd like clean separation between steps "install openstack" and "use openstack" | 11:03 |
| * sambetts wishes we could use network names for this, e.g. default network name in Ironic is "ironic-{provisioning, cleaning}" then we can create the networks after Ironic is already running, they just have to have the right name | 11:03 | |
| jroll | so this being node-create makes this less terrible, because you'd add it to your script | 11:03 |
| dtantsur | variants 1 and 2 have a split point in the first step, and assume the same person is doing these two steps | 11:03 |
| jroll | dtantsur: so, strawman... should we put auth info elsewhere as well? someone needs to talk to keystone to create the users... | 11:04 |
| sambetts | jroll: they can be created after the service is already running though | 11:04 |
| sambetts | the service being Ironic | 11:04 |
| dtantsur | heh, this is a fair question, but I think I feel some difference.. and sambetts is also right | 11:05 |
| *** zhangjl has quit IRC | 11:05 | |
| jroll | dtantsur: also, we need to stop assuming 1) deployers are incapable of using the API, and 2) ironic deployers don't talk to the ironic users configuring nodes | 11:05 |
| jroll | can you choose the tenant id when you create them? | 11:05 |
| jroll | s/the API/an API/ I think | 11:05 |
| dtantsur | ok, I think I understand the difference | 11:06 |
| * jroll gets less okay with this the more we talk about it, heh | 11:06 | |
| dtantsur | jroll, the difference is that service users are not something real users touch or define. the network, on the other hand, is something users define. I don't think it's set in stone, but that seems to be a typical assumption. | 11:06 |
| jroll | erm | 11:07 |
| jroll | dtantsur: right, so I think a "service network" should also be defined by a non-real user, whatever that means | 11:08 |
| dtantsur | note that cleaning network in simplest case will match the tenant network | 11:08 |
| openstackgerrit | Vasyl Saienko proposed openstack/python-ironicclient: Add create external event functionality https://review.openstack.org/354770 | 11:08 |
| jroll | well, it will match the provider network | 11:08 |
| jroll | it's a deployer-owned network, in all cases | 11:08 |
| sambetts | jroll: only in flat case | 11:08 |
| jroll | right, "in simplest case" | 11:08 |
| jroll | dtantsur: in any case, surely someone will soon sell a contract where the customer wants multiple cleaning networks, so we'll need to do it sometime :( | 11:11 |
| jroll | I *really* dislike losing the startup validation | 11:11 |
| jroll | another place where someone can easily misconfigure things that fail later | 11:12 |
| * dtantsur just notices that https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/network/flat.py#L39 if done will break tripleo... | 11:12 | |
| dtantsur | jroll, isn't it what validate() call is for? | 11:12 |
| jroll | dtantsur: should probably have an "if CONF.automated_clean" in there as well, would that unbreak tripleo? | 11:13 |
| dtantsur | yep | 11:13 |
| jroll | dtantsur: it is | 11:13 |
| sambetts | jroll: i was about to say that :-P | 11:13 |
| * jroll is sad that there's deployments that skip cleaning | 11:13 | |
| * dtantsur hates that tripleo disables cleaning | 11:13 | |
| dtantsur | lol +1 | 11:13 |
| jroll | :) | 11:13 |
| * dtantsur -> quick food, continue later | 11:14 | |
| lucasagomes | jroll, dtantsur re deploy_{kernel, initrd} in the node, that's useful. In a multi-arch env that's needed | 11:16 |
| * lucasagomes don't think it's terrible | 11:16 | |
| -openstackstatus- NOTICE: Precise tests on OSIC provider are currently failing, please stop your checks until the issue is resolved. | 11:16 | |
| lucasagomes | perhaps we could have a fallback to a configuration file to make it simpler | 11:17 |
| *** ccamacho is now known as ccamacho|lunch | 11:17 | |
| lucasagomes | but, that's a required node configuration IMO | 11:17 |
| mat128 | good morning | 11:17 |
| lucasagomes | mat128, morning | 11:18 |
| mat128 | reading scrollback about your chicken and egg problem | 11:20 |
| mat128 | how does one define a swift account/container for Ironic to use? | 11:20 |
| dtantsur | lucasagomes, I've heard we don't support multi-arch now... and anyway this quite a corner case too | 11:21 |
| dtantsur | mat128, account as usual, containers do not have to be pre-created | 11:21 |
| lucasagomes | dtantsur, ofc totally do | 11:21 |
| lucasagomes | we totally* | 11:21 |
| mat128 | ah! | 11:21 |
| -openstackstatus- NOTICE: DSVM jobs on OSIC currently failing because of IP collisions, fix is in the gate - https://review.openstack.org/#/c/357764/ - please hold rechecks until merged | 11:21 | |
| lucasagomes | dtantsur, I really don't find it fair to compare deploy_kernel in the node with entwork | 11:22 |
| sambetts | I did wonder about Ironic creating the provisioning network Neutron | 11:22 |
| dtantsur | lucasagomes, do we provide different pxelinux.0 for different architectures? | 11:22 |
| lucasagomes | network separation is more about security | 11:22 |
| dtantsur | I think that was the issue | 11:22 |
| lucasagomes | kernel/initrd is a technical issue | 11:22 |
| lucasagomes | dtantsur, that's a DHCP configuration | 11:22 |
| dtantsur | yes, but we define it. anyway, I think we have a spec/patch for multi-arch | 11:23 |
| lucasagomes | DHCP servers can look at client-arch and send a different PXE image | 11:23 |
| mat128 | dtantsur: our downstream testing script need to get the swift account UUID where we only have the name because it's dynamically created | 11:23 |
| mat128 | dtantsur: so we essentially do a name lookup | 11:23 |
| lucasagomes | dtantsur, it's out of topic with the networks anyway | 11:23 |
| lucasagomes | I just don't think it's a fair comparison | 11:23 |
| dtantsur | it's fair from logical point. logically, both images and networks are created by admin users, not by deployers (even though it's often the same person) | 11:24 |
| dtantsur | I agree that technically these are different | 11:25 |
| mat128 | dtantsur: if we follow that logic, *_network has to go into the node because that's under the admin's responsibility, not the deployer | 11:25 |
| dtantsur | *_network is what? enabled_network_interfaces is a property of the deployment. | 11:25 |
| dtantsur | however, we do allow setting network_interface on a node | 11:26 |
| sambetts | I think *_network_uuid is what mat128 meant | 11:26 |
| mat128 | sambetts is correct | 11:26 |
| mat128 | my bad | 11:26 |
| dtantsur | aha, cool, thanks | 11:26 |
| * dtantsur is confused who argues against what already :D | 11:27 | |
| mat128 | dtantsur: not arguing :) | 11:28 |
| dtantsur | jroll, see my comment on https://bugs.launchpad.net/ironic/+bug/1614876 I don't get why you think it's specific to tripleo | 11:28 |
| openstack | Launchpad bug 1614876 in Ironic "[RFE] Allow setting {provisioning,cleaning}_network_uuid in node driver_info" [Wishlist,Confirmed] - Assigned to Dmitry Tantsur (divius) | 11:28 |
| jroll | lucasagomes: yep, totally agree with you, this is why I didn't want to discuss deploy_kernel etc | 11:28 |
| *** moshele has quit IRC | 11:29 | |
| mat128 | dtantsur: I could see point #3 happening in my deployment, even without tripleO | 11:29 |
| jroll | dtantsur: this started as "this is a problem for tripleo because updating config is a long stack-update" | 11:30 |
| mat128 | right now our deployment network is just a vlan ID, but I can see a day where there are too many machines on this vlan and we need to have more than one | 11:30 |
| dtantsur | well, yes, but I've provided quite a few tripleo-independent points | 11:30 |
| dtantsur | also tripleo limitation comes from puppet modules (and I suspect their policy), so it must be similar for fuel, for example | 11:30 |
| jroll | dtantsur: right, which is why I didn't say no :) | 11:30 |
| jroll | dtantsur: well, we used puppet when we first deployed ironic... we just created a network before installing, and put it in the config :) | 11:31 |
| jroll | wasn't a problem | 11:31 |
| *** lucasagomes is now known as lucas-hungry | 11:31 | |
| dtantsur | well, this assumes you have a break between installing the remaining openstack and ironic, right? | 11:32 |
| jroll | yes | 11:32 |
| jroll | I also believe openstack shouldn't be a monolith all deployed at once, but that's another story :) | 11:33 |
| dtantsur | "openstack shouldn't be a monolith" is a good call, but yeah :) it's a whole different story | 11:33 |
| dtantsur | btw, I've heard talks that service users might eventually be replaces by certificates | 11:33 |
| jroll | thank $deity! | 11:33 |
| jroll | service auth is pretty bad right now, especially taking actions on user behalf | 11:34 |
| jroll | (like port-update) | 11:34 |
| dtantsur | yep | 11:34 |
| dtantsur | getting back to the topic, I think I understand people wanting to install $PRODUCT_NAME in one shot too | 11:35 |
| jroll | yeah | 11:35 |
| jroll | I just don't think injecting a neutron net-create command is that terrible :) | 11:35 |
| *** dprince has joined #openstack-ironic | 11:35 | |
| mat128 | dtantsur, jroll: I think the main point (for me at least) is defining more than one provisioning network, good old rule of thumb states 500-1000 machines max in a vlan | 11:36 |
| dtantsur | a use case has arrived ^^ :) | 11:36 |
| mat128 | (and independent of deployment "process") | 11:36 |
| * jroll is fairly certain he has more than 1000 in a vlan :) | 11:36 | |
| jroll | but yeah | 11:37 |
| mat128 | not too much broadcast? | 11:37 |
| jroll | nah | 11:37 |
| jroll | I guess they aren't all there at once though | 11:37 |
| mat128 | interesting | 11:37 |
| mat128 | we are limiting IPMI vlan/subnets to 500 because of this | 11:37 |
| jroll | mat128: a better solution for you, though, would be CONF.provisioning_nets = [foo, bar] | 11:38 |
| mat128 | we just have one per X racks | 11:38 |
| jroll | and use one with free IPs | 11:38 |
| mat128 | jroll: and let ironic decide of hosts / provisioning network? | 11:38 |
| mat128 | interesting | 11:38 |
| jroll | but anyway, I already +1'd the RFE ¯\_(ツ)_/¯ | 11:38 |
| jroll | mat128: well, you'd decide by making it a /23 | 11:39 |
| mat128 | yeah | 11:39 |
| mat128 | i like that idea | 11:39 |
| jroll | :D | 11:39 |
| mat128 | (doesnt help dtantsur lol) | 11:39 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironic-inspector-client: Fix functional tests broken by the latest refactoring https://review.openstack.org/357795 | 11:39 |
| jroll | I'd love to just make it magic | 11:39 |
| jroll | have a dedicated ironic user and feel free to use any networks that user owns :) | 11:40 |
| mat128 | how do you distinguish provisioning/cleaning ? | 11:40 |
| mat128 | metadata? | 11:40 |
| dtantsur | thinking of all this node.* bloat, I'd love to have a "node template" entity | 11:40 |
| jroll | idk, just spitballing | 11:40 |
| dtantsur | that would contain things like driver_info/deploy_kernel, driver_info/cleaning_network, properties/cpu_arch, maybe even driver | 11:40 |
| mat128 | jroll: I think sambetts mentioned having ironic create the network | 11:40 |
| dtantsur | and then a node would be created like $ ironic node-create --template my-x86-64 --name my-node | 11:41 |
| mat128 | dtantsur: driver type? | 11:41 |
| dtantsur | well... something even more lightweight | 11:41 |
| jroll | mat128: you'd have to have ironic know the vlan to use and such, but could be cool | 11:41 |
| mat128 | jroll: yeah, but I was more thinking like, if Ironic creates it, it could keep track of it's uuid and that solves dtantsur's problem | 11:42 |
| jroll | dtantsur: I still like the node grouping thing we've talked about, having those configs ("okay rack a-12 is all x86, ironic-gold, this ramdisk") | 11:42 |
| sambetts | mat128: this ^ | 11:42 |
| sambetts | jroll: this ^ | 11:42 |
| sambetts | also | 11:42 |
| sambetts | :-P | 11:42 |
| jroll | heh | 11:42 |
| *** baoli has joined #openstack-ironic | 11:42 | |
| dtantsur | hmm, create both provisioning and cleaning network? | 11:42 |
| jroll | mat128: yeah, but you just end up putting the IP range and vlans in config instead | 11:43 |
| mat128 | sambetts: are you looking for @channel by any chance? :) | 11:43 |
| mat128 | jroll: yeah.. removes the flexibility of creating it in neutron | 11:43 |
| dtantsur | that would work even better for me, though we'll need to pass the hardware network (which is totally fine, as it's responsibility of the deployer) | 11:43 |
| mat128 | and forces us to define all possible parameters | 11:43 |
| jroll | I guess you could use either | 11:43 |
| dtantsur | cleaning_network_uuid = auto | 11:43 |
| jroll | define it up front or just give it a block to create it for | 11:43 |
| dtantsur | or a real UUID :) | 11:43 |
| *** moshele has joined #openstack-ironic | 11:44 | |
| jroll | realistically when you have to mess around creating ACLs and routes and base switch config for everything... net-create isn't a big deal :P | 11:44 |
| * jroll should open source our bootstrapping repo one day | 11:45 | |
| dtantsur | jroll, mat128, do you mean something like https://etherpad.openstack.org/p/jq37kmQ908 ? | 11:46 |
| sambetts | dtantsur: yup, pretty much | 11:46 |
| mat128 | dtantsur: segmentation_id, too | 11:46 |
| mat128 | but yes | 11:46 |
| dtantsur | well, yeah, a couple more options, but still | 11:46 |
| dtantsur | then we have to somehow remember the UUID (or just use a predefined UUID?) | 11:47 |
| dtantsur | (or just use a well-known name?) | 11:47 |
| * dtantsur is ready to hack on this instead of his RFE | 11:47 | |
| mat128 | if it's a well known name and ironic does a lookup, you can create it *after* deployment | 11:47 |
| mat128 | but prior to deploying any node | 11:47 |
| mat128 | which also solves your problem in a different way | 11:48 |
| mat128 | dtantsur: ^ | 11:48 |
| dtantsur | right. I'm fine with both approaches. the less things to create - the better | 11:48 |
| jroll | yeah, I'd prefer not to have ironic manage provider net create | 11:48 |
| jroll | provider networks can get too special, I suspect | 11:48 |
| dtantsur | true | 11:48 |
| mat128 | jroll: would you be okay with a lookup by name instead of uuid in the config? | 11:48 |
| jroll | if we do the "well known name" thing, let's... that | 11:48 |
| jroll | name or uuid | 11:48 |
| *** moshele has quit IRC | 11:48 | |
| jroll | default to ironic-provision-net | 11:48 |
| jroll | or something | 11:48 |
| mat128 | I think that solves dtantsur problem in a clean way | 11:49 |
| jroll | and actually do a net-show at startup maybe? | 11:49 |
| mat128 | or when we need it | 11:49 |
| mat128 | (it might change) | 11:49 |
| dtantsur | jroll, then ironic will fail to start until we create it | 11:49 |
| -openstackstatus- NOTICE: OSIC has burned through the problematic IP range with failures, things should be back to normal now. | 11:49 | |
| jroll | right | 11:49 |
| dtantsur | so it's the same problem essentially | 11:49 |
| dtantsur | however, right now we don't check UUID on existence | 11:49 |
| jroll | I'm just trying to help the users that just leave the default and don't create the network | 11:49 |
| dtantsur | so it might be fair to not check the name too | 11:50 |
| mat128 | this is exactly like the RFC that states "whereever you can provide a hostname, you can also provide an IP address, and vice-versa") | 11:50 |
| * dtantsur is confused with how many proposal we have here simultaneously... | 11:50 | |
| openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Add update notifications for ironic resources https://review.openstack.org/357799 | 11:50 |
| jroll | dtantsur: right, but if there's now a default, I'd prefer to check it, maybe validate time is okay | 11:50 |
| dtantsur | so, proposal #1: allow per-node setting | 11:50 |
| jroll | it's less good but it might be ok | 11:51 |
| dtantsur | proposal #2: accept names and do not check existence on start up | 11:51 |
| mat128 | dtantsur: mine is "dont add anything to the node, instead declare a name in the config and lookup will be dont at deploy time" | 11:51 |
| dtantsur | proposal #3: have a default and......? | 11:51 |
| TheJulia | Good morning | 11:51 |
| dtantsur | morning TheJulia | 11:51 |
| mat128 | o/ TheJulia | 11:51 |
| sambetts | o/ | 11:51 |
| dtantsur | mat128, so proposal #3: default to a well-known name, check its presence right before deployment, right? | 11:51 |
| *** bfournie has quit IRC | 11:52 | |
| mat128 | dtantsur: and allow a network "name" in that config option | 11:52 |
| dtantsur | right | 11:52 |
| jroll | well, check presence is it's own question | 11:52 |
| * jroll making notes in https://etherpad.openstack.org/p/jq37kmQ908 | 11:52 | |
| mat128 | dont check its presence, just do your regular attach thing | 11:52 |
| dtantsur | if we check presence on start up, we're back to the initial problem :) | 11:52 |
| mat128 | if it fails, handle it | 11:52 |
| mat128 | dtantsur: yes, "resolution" must be done at deploy time | 11:53 |
| jroll | dtantsur: if we assert the only reason to do this is to allow deploying ironic before creating a network, yes | 11:53 |
| jroll | I don't think we've asserted that | 11:53 |
| * TheJulia reads backscroll for current discussion | 11:53 | |
| jroll | (multiple networks is another reason) | 11:53 |
| jroll | TheJulia: see you in 15 minutes :P | 11:53 |
| jroll | (and good morning) | 11:53 |
| mat128 | jroll: if we do this, then the original RFE changes, there is no "per-node" anymore | 11:53 |
| TheJulia | heh | 11:53 |
| dtantsur | jroll, my personal problem is with deployment, mat128 has also a use case for several networks, but seems like it's possible to overcome (?) | 11:54 |
| TheJulia | Eh, its not that long of a read, but coffee first | 11:54 |
| jroll | mat128: right, so I think we need to establish what we want to solve here | 11:54 |
| TheJulia | ^^ that | 11:54 |
| jroll | if we still allow only one name or uuid, mat128 still has a problem | 11:54 |
| mat128 | dtantsur: I could always tell my network people that there's someone else doing more than 1000 without problem | 11:54 |
| jroll | maybe the answer is do both! | 11:54 |
| mat128 | but they'd be hard to convince | 11:54 |
| jroll | mat128: don't quote me on that! | 11:54 |
| mat128 | dont worry | 11:54 |
| jroll | :P | 11:55 |
| mat128 | our hosts arent always up | 11:55 |
| mat128 | so having 1000 actually *deploying* at the same time | 11:55 |
| mat128 | in a given region | 11:55 |
| jroll | ah yeah, that's another point, if you don't leave agents up it doesn't matter | 11:55 |
| jroll | mat128: I think that's a problem we'd both love to have :D | 11:55 |
| dtantsur | jroll, I'm totally fine with the "do both" thing | 11:55 |
| mat128 | jroll: we do have huge spikes though | 11:56 |
| dtantsur | however, the first thing I want to do is to be able to start Ironic before networks are created | 11:56 |
| jroll | mat128: yeah, but 1000 is $$$$ | 11:56 |
| mat128 | =) | 11:56 |
| dtantsur | so, should I capture the name thingy in another RFE? objections? | 11:57 |
| jroll | dtantsur: I think "accept names" is a generally useful thing - maybe let's skip the well-known/default question for now though | 11:57 |
| mat128 | no objections, but will you also mention that the lookup needs to be done at deploy time in there? | 11:57 |
| jroll | mat128: that changes with per-node anyway | 11:58 |
| jroll | but I'd say s/deploy time/validate time/ | 11:58 |
| dtantsur | jroll, I'm fine, but I think we tend not to make outside calls in validation | 11:58 |
| jroll | dtantsur: we call ipmitool no? | 11:58 |
| jroll | if neutron GET /networks/foo ever becomes slower than a BMC, let's rewrite neutron | 11:59 |
| dtantsur | LOL | 11:59 |
| * dtantsur rewrites neutron | 11:59 | |
| sambetts | HAHA | 11:59 |
| mat128 | jroll: dont you think one day BMCs will be fast? :) | 11:59 |
| * mat128 dreams | 11:59 | |
| jroll | mat128: I also dream neutron will be fast! | 11:59 |
| jroll | dtantsur: different topic, want to decide what stays and goes in newton with me today or monday? maybe monday we can pull in ruby | 12:01 |
| jroll | I'm not ready right this minute but am going to look over things this morning | 12:01 |
| dtantsur | yeah, I'd prefer Monday, but can do it today | 12:01 |
| jroll | dtantsur: monday is fine, I have lots to do today :) | 12:02 |
| jroll | oh, also of note, I'm out tuesday/wednesday next week | 12:02 |
| * sambetts gets ready the rm -rf drivers/modules/ssh* | 12:02 | |
| jroll | so I'll probably put up final ironic-lib release monday | 12:02 |
| dtantsur | so folks, did I capture it right: https://bugs.launchpad.net/ironic/+bug/1614938 ? | 12:03 |
| openstack | Launchpad bug 1614938 in Ironic "[RFE] Allow names in {provisioning,cleaning}_network option" [Wishlist,Confirmed] - Assigned to Dmitry Tantsur (divius) | 12:03 |
| sambetts | dtantsur: option 4 is have Ironic create the networks for us | 12:03 |
| jroll | lucas-hungry: any update on https://review.openstack.org/#/c/348953/ ? | 12:03 |
| dtantsur | yes. but this can be dangerous, as there are plenty of options | 12:04 |
| mat128 | dtantsur: will the lookup be done in the ironic service tenant? | 12:04 |
| dtantsur | mat128, probably? | 12:04 |
| jroll | dtantsur: so I think that looks good, but I think the problem statement is wrong :) | 12:04 |
| mat128 | makes sense, but just checking | 12:04 |
| jroll | dtantsur: "names are useful for humans" is a better problem statement, I think | 12:05 |
| dtantsur | jroll, feel free to fix it whatever makes sense to you | 12:05 |
| dtantsur | yes, it's a good problem statement too | 12:05 |
| * jroll does it | 12:05 | |
| jroll | edited quite a bit, I'm +1 on it but let's have another specs core approve | 12:07 |
| * dtantsur tries to remember who is specs core | 12:07 | |
| dtantsur | lucas-hungry, ^^^ | 12:07 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironic-inspector-client: Add functional test for wait_for_finish https://review.openstack.org/357804 | 12:08 |
| * jroll doesn't like the fancy tox_install junk for libraries to edit constraints files | 12:08 | |
| sambetts | jroll: I agree, but I don't see a way around it :/ | 12:10 |
| jroll | yeah, there isn't | 12:11 |
| jroll | other than fix pip :) | 12:11 |
| *** ccamacho|lunch is now known as ccamacho | 12:12 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-lib: Use constraints for all the things https://review.openstack.org/318692 | 12:12 |
| jroll | dtantsur: sambetts: here's ironic-lib, I'll do the rest of our repos after I get the animals some breakfast, if you're good with that | 12:12 |
| jroll | oh wait | 12:12 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-lib: Use constraints for all the things https://review.openstack.org/318692 | 12:13 |
| jroll | there, it's ready now | 12:13 |
| jroll | want to get those done before lib freeze | 12:13 |
| dtantsur | cool, let's wait for the CI | 12:13 |
| sambetts | sure :) I'll dig into them in just a minute | 12:13 |
| jroll | thanks! | 12:13 |
| * jroll runs locally | 12:14 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-lib: Use constraints for all the things https://review.openstack.org/318692 | 12:15 |
| jroll | *now* it works | 12:15 |
| jroll | silly +x | 12:15 |
| jroll | passes locally though | 12:15 |
| *** thrash|g0ne is now known as thrash | 12:15 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironic-inspector-client: Fix functional tests broken by the latest refactoring https://review.openstack.org/357795 | 12:17 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironic-inspector-client: Add functional test for wait_for_finish https://review.openstack.org/357804 | 12:18 |
| *** MattMan has quit IRC | 12:24 | |
| *** MattMan has joined #openstack-ironic | 12:24 | |
| *** sdake has joined #openstack-ironic | 12:35 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/virtualbmc: Use upper-constraints for all tox targets https://review.openstack.org/357825 | 12:37 |
| *** alaski is now known as lascii | 12:38 | |
| *** bfournie has joined #openstack-ironic | 12:38 | |
| sambetts | jroll: is virtualbmc not in globalreqs yet? | 12:39 |
| openstackgerrit | Jim Rollenhagen proposed openstack/python-ironicclient: Use constraints for all the things https://review.openstack.org/318672 | 12:40 |
| jroll | sambetts: guess not, I don't think it needs to be though | 12:40 |
| jroll | nothing imports it | 12:40 |
| sambetts | jroll: good point, I was thinking it was a library but its not | 12:40 |
| *** lucas-hungry is now known as lucasagomes | 12:41 | |
| lucasagomes | dtantsur, I like the names suggestions yes | 12:42 |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Negative tests for testing actions with chassis-update https://review.openstack.org/344715 | 12:43 |
| dtantsur | jroll, lucasagomes, rfe-appoved then? | 12:43 |
| lucasagomes | +1 for me | 12:43 |
| *** wajdi has quit IRC | 12:43 | |
| lucasagomes | from* | 12:43 |
| jroll | dtantsur: yeah, go ahead | 12:44 |
| lucasagomes | there's a +1 from jroll in the bug, so I've just changed the tag | 12:44 |
| jroll | or that! | 12:44 |
| dtantsur | cool, thanks! I'll try to come up with something today | 12:45 |
| *** jtomasek has joined #openstack-ironic | 12:48 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/bifrost: Use upper-constraints for all tox targets https://review.openstack.org/357836 | 12:49 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-python-agent: Use upper-constraints for all tox targets https://review.openstack.org/357838 | 12:50 |
| *** baoli has quit IRC | 12:51 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/python-ironic-inspector-client: Use constraints for all the things https://review.openstack.org/318693 | 12:51 |
| jroll | all the constraints! https://review.openstack.org/#/q/topic:use-constraints+status:open | 12:53 |
| sambetts | \o/ | 12:53 |
| jroll | I think ironic-lib, wsmanclient, dracclient are the ASAP ones there, we should release those next week | 12:54 |
| jroll | lucasagomes: shall we release virtualbmc after this merges? https://review.openstack.org/#/c/357825/ | 12:56 |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Tests for testing port-create command https://review.openstack.org/291802 | 12:56 |
| lucasagomes | jroll, yes! It's been a while since last release | 12:56 |
| lucasagomes | jrist, +a btw | 12:56 |
| jroll | lucasagomes: I don't see any tags, have we had a release at all? | 12:56 |
| lucasagomes | jrist, sorry heh | 12:56 |
| jroll | http://git.openstack.org/cgit/openstack/virtualbmc/refs/ | 12:57 |
| lucasagomes | jroll, it wasn't under /openstack before | 12:57 |
| jroll | ah right | 12:57 |
| *** dprince has quit IRC | 12:57 | |
| jroll | shall we 1.0.0 it? :) | 12:57 |
| lucasagomes | jroll, go for it | 12:57 |
| jroll | cool | 12:57 |
| jroll | I'll do that when that constraints thing lands | 12:57 |
| lucasagomes | ty | 12:58 |
| jroll | I'm not sure if this counts as a library or a service, so idk which freeze applies :P | 12:58 |
| * lucasagomes have to release the staging drivers as well | 12:58 | |
| jroll | oh, release:independent, guess it doesn't matter | 12:58 |
| lucasagomes | jroll, we use it as a service, but yeah ^ (was about to write that) | 12:58 |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Negative tests for testing actions with Chassis. https://review.openstack.org/343796 | 12:59 |
| *** dprince has joined #openstack-ironic | 13:02 | |
| *** baoli has joined #openstack-ironic | 13:03 | |
| dtantsur | sambetts, aarefiev, can we merge https://review.openstack.org/#/c/357763/ please? it blocks fixing the client func test | 13:04 |
| *** rbudden has joined #openstack-ironic | 13:05 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: DO NOT REVIEW https://review.openstack.org/296432 | 13:09 |
| *** jtomasek has quit IRC | 13:11 | |
| *** mtanino has joined #openstack-ironic | 13:13 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironic-inspector-client: Add functional tests for CLI https://review.openstack.org/357864 | 13:15 |
| *** _vsaienko has quit IRC | 13:18 | |
| openstackgerrit | Merged openstack/ironic-inspector: functional: allow passing kwargs to all calls https://review.openstack.org/357763 | 13:18 |
| dtantsur | vsaienk0, any plans to continue with https://review.openstack.org/328921 ? | 13:19 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Update release notes for Newton https://review.openstack.org/357868 | 13:20 |
| *** cdearborn has joined #openstack-ironic | 13:21 | |
| *** sabeen has joined #openstack-ironic | 13:21 | |
| *** mtanino has quit IRC | 13:21 | |
| *** _vsaienko has joined #openstack-ironic | 13:28 | |
| jroll | who wants to land multitenant network docs? https://review.openstack.org/#/c/228496/93 | 13:29 |
| *** baoli has quit IRC | 13:34 | |
| vsaienk0 | dtantsur: I will update them on next week | 13:35 |
| dtantsur | thanks! | 13:35 |
| *** causten has joined #openstack-ironic | 13:36 | |
| *** jcoufal has joined #openstack-ironic | 13:36 | |
| *** _vsaienko has quit IRC | 13:38 | |
| *** derekh has quit IRC | 13:42 | |
| *** wajdi has joined #openstack-ironic | 13:42 | |
| *** mari0jv has joined #openstack-ironic | 13:42 | |
| *** dprince has quit IRC | 13:43 | |
| *** dprince has joined #openstack-ironic | 13:43 | |
| *** foexle has quit IRC | 13:43 | |
| *** foexle has joined #openstack-ironic | 13:45 | |
| *** derekh has joined #openstack-ironic | 13:45 | |
| *** mari0jv_ has joined #openstack-ironic | 13:46 | |
| *** baoli has joined #openstack-ironic | 13:47 | |
| *** mari0jv__ has joined #openstack-ironic | 13:47 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient: Add --wait flag for provision actions and wait_for_provision_state function https://review.openstack.org/328191 | 13:48 |
| dtantsur | vdrok, ^^^ | 13:48 |
| *** dansmith is now known as superdan | 13:48 | |
| vdrok | dtantsur: yup, will take a look now | 13:48 |
| *** mari0jv_ has left #openstack-ironic | 13:49 | |
| vdrok | btw morning all :) | 13:49 |
| *** mari0jv__ has quit IRC | 13:49 | |
| *** mari0jv has quit IRC | 13:50 | |
| *** mari0jv has joined #openstack-ironic | 13:50 | |
| dtantsur | morning/afternoon(/evening?) vdrok | 13:51 |
| *** thiagop has joined #openstack-ironic | 13:52 | |
| *** catintheroof has joined #openstack-ironic | 13:54 | |
| *** e0ne_ has joined #openstack-ironic | 13:55 | |
| thiagop | Good morning, Ironic | 13:55 |
| sambetts | jroll: is dracclient not in global constraints? /me doesn't know if it should be or not | 13:55 |
| TheJulia | Good morning thiagop | 13:55 |
| *** e0ne has quit IRC | 13:55 | |
| thiagop | TheJulia: o/ | 13:56 |
| jroll | sambetts: guess not, because it's an extra requirement (not in requirements.txt) | 13:56 |
| jroll | it couldn't hurt to put it there though | 13:56 |
| openstackgerrit | Merged openstack/ironic: Operator documentation for multitenancy https://review.openstack.org/228496 | 13:56 |
| *** bfournie1 has joined #openstack-ironic | 13:57 | |
| thiagop | sambetts: it is on driver-requirements though | 13:58 |
| *** bfournie2 has joined #openstack-ironic | 13:58 | |
| *** bfournie1 has quit IRC | 13:58 | |
| *** bfournie3 has joined #openstack-ironic | 13:58 | |
| *** bfournie has quit IRC | 13:58 | |
| *** anush_ has joined #openstack-ironic | 14:00 | |
| *** catintheroof has quit IRC | 14:01 | |
| *** bfournie2 has quit IRC | 14:02 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: DO NOT REVIEW https://review.openstack.org/296432 | 14:03 |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: Introducing node introspection state management https://review.openstack.org/348943 | 14:10 |
| *** sabeen1 has joined #openstack-ironic | 14:12 | |
| *** chlong has joined #openstack-ironic | 14:13 | |
| *** sabeen has quit IRC | 14:14 | |
| * jroll kicks off the driver removal question thread | 14:15 | |
| openstackgerrit | Merged openstack/ironic-lib: Use constraints for all the things https://review.openstack.org/318692 | 14:16 |
| *** hjanssen-hpe has joined #openstack-ironic | 14:17 | |
| *** hpe-hj has joined #openstack-ironic | 14:17 | |
| *** wznoinsk has quit IRC | 14:19 | |
| mat128 | jroll: 0-based indexing.. interesting ;) | 14:19 |
| sambetts | mat128: we're programers we always start counting from zero :-P | 14:20 |
| * mat128 realizes I've been doing 1-based index in all prior emails :( | 14:21 | |
| mat128 | s/I've/he's | 14:21 |
| jroll | yeah, always 0-based for me | 14:21 |
| mat128 | not used to that | 14:21 |
| jroll | most do 1 though | 14:21 |
| mat128 | well the first like you have did 0-based too | 14:21 |
| mat128 | the cinder thread | 14:21 |
| jroll | heh, nice | 14:21 |
| *** ksumit has joined #openstack-ironic | 14:21 | |
| * jroll aims to change the world one email sender at a time | 14:22 | |
| sambetts | jroll: will the ssh driver have to go through the same thing? | 14:23 |
| sambetts | given its not a "real" driver | 14:23 |
| jroll | sambetts: yeah, we should do it for all drivers for consistency | 14:24 |
| jroll | also, turns out there's production deployments using that driver >.> | 14:24 |
| sambetts | <.< ... >.> ... <.< ... | 14:25 |
| sambetts | who even does that :-P | 14:25 |
| jroll | I wish I knew | 14:25 |
| *** jtomasek has joined #openstack-ironic | 14:27 | |
| *** jtomasek has quit IRC | 14:28 | |
| *** mtanino has joined #openstack-ironic | 14:28 | |
| openstackgerrit | Merged openstack/virtualbmc: Use upper-constraints for all tox targets https://review.openstack.org/357825 | 14:28 |
| *** jtomasek has joined #openstack-ironic | 14:29 | |
| mat128 | remove it and you'll know :) | 14:29 |
| mat128 | baremetal using virtual machines on compute nodes isnt baremetal =) | 14:29 |
| jroll | no, but it apparently works well enough that people actually use it | 14:29 |
| jroll | which is just so strange to me | 14:29 |
| *** wznoinsk has joined #openstack-ironic | 14:30 | |
| jroll | especially if they use it with nova! | 14:30 |
| mat128 | but it still does virtual machines | 14:30 |
| mat128 | thats it | 14:30 |
| jroll | well yeah | 14:30 |
| mat128 | might as well use nova with libvirt | 14:30 |
| jroll | but they use it to manage VMs | 14:30 |
| jroll | ¯\_(ツ)_/¯ | 14:30 |
| jroll | you don't have to tell me :) | 14:30 |
| mat128 | or maybe ironic is the "virt" abstraction that should have always been | 14:30 |
| mat128 | and nova could "only" manage the instance part | 14:30 |
| jroll | heh | 14:31 |
| sambetts | maybe virtualBMC is ;) | 14:31 |
| mat128 | :P | 14:31 |
| mat128 | on that topic | 14:31 |
| mat128 | virtualPDU is almost done | 14:31 |
| openstackgerrit | Graham Hayes proposed openstack/ironic: Get ready for os-api-ref sphinx theme change https://review.openstack.org/357922 | 14:33 |
| sambetts | mat128: oh nice! | 14:34 |
| *** michauds has joined #openstack-ironic | 14:36 | |
| mat128 | see, the driver deprecation / removal forced us to open source a component, helping everyone :) | 14:37 |
| sambetts | \o/ | 14:37 |
| jroll | heh, woo | 14:39 |
| mat128 | jroll: if you wanted to completely follow policy, you could introduce "enable_unsupported_drivers" defaulting to True, then notifying that in X release setting will go to False | 14:41 |
| mat128 | but that takes 1 year of fooling around | 14:41 |
| sambetts | mat128: in X release we'll just remove the drviers though ;) | 14:42 |
| jroll | mat128: but then *later* changing a driver supported to False breaks people, still | 14:42 |
| mat128 | but they will (supposedly) know because we told them enable_supported would turn to false | 14:43 |
| *** foexle has quit IRC | 14:43 | |
| jroll | well, no | 14:43 |
| jroll | so | 14:43 |
| jroll | we get through X like you say right? | 14:43 |
| mat128 | I think current proposal is perfect, if operator finds out his driver is unsupported, he can just enable the "enable_unsupported_drivers" and move on, notify his vendor or something | 14:43 |
| mat128 | yes | 14:43 |
| jroll | then in X+1 ipmitool stops being tested | 14:43 |
| *** syed_ has joined #openstack-ironic | 14:43 | |
| jroll | we mark it supported=False | 14:43 |
| mat128 | ahhh | 14:43 |
| mat128 | got it | 14:43 |
| jroll | :) | 14:43 |
| mat128 | it was set to True when we did the initial thing | 14:43 |
| jroll | yeah | 14:44 |
| *** mari0jv has quit IRC | 14:46 | |
| *** bfournie has joined #openstack-ironic | 14:49 | |
| *** bfournie3 has quit IRC | 14:51 | |
| *** rama_y has joined #openstack-ironic | 15:01 | |
| *** milan has quit IRC | 15:03 | |
| *** ElCoyote_ has quit IRC | 15:08 | |
| *** ElCoyote_ has joined #openstack-ironic | 15:08 | |
| *** dprince has quit IRC | 15:09 | |
| *** mari0jv has joined #openstack-ironic | 15:12 | |
| *** ksumit has quit IRC | 15:13 | |
| *** ksumit has joined #openstack-ironic | 15:14 | |
| *** Sylvester has quit IRC | 15:16 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] Switch Inspector interface to pass keystoneauth sessions https://review.openstack.org/357963 | 15:18 |
| *** rcernin has quit IRC | 15:24 | |
| *** pcaruana has quit IRC | 15:26 | |
| *** ccamacho is now known as ccamacho|away | 15:29 | |
| *** Goneri has quit IRC | 15:33 | |
| phschwartz | what would the suggestion be for having VM's and BM in a single region? I do not want to run cells. | 15:34 |
| *** e0ne_ has quit IRC | 15:39 | |
| sambetts | phschwartz: host aggregates | 15:40 |
| *** ametts has joined #openstack-ironic | 15:41 | |
| *** ksumit has quit IRC | 15:41 | |
| sambetts | phschwartz: you can setup multiple nova computes, 1 running the ironic virt driver, all others running VM virt drivers, then tag your flavors with host aggregates to ensure that the baremetal flavors land on the right nova compute for Ironic | 15:42 |
| *** ksumit has joined #openstack-ironic | 15:43 | |
| phschwartz | sambetts: I had not thought of that. I was messing with getting 2 nova-computes running on a single host with different configs for each. With that method I could use it in this manor as I am doing this as an allinone setup with VM's on the allinone (preferred), and 50 bm's to enroll for ironic. | 15:45 |
| *** ametts has quit IRC | 15:47 | |
| sambetts | phschwartz: host aggregates sounds like the right solution for you :) I have had a test bed configured exactly like that with 2 nova computes using host aggregates to put some flavors on BMs and some on VMs | 15:48 |
| *** links has quit IRC | 15:49 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-lib: Add developer documentation on metrics https://review.openstack.org/356796 | 15:49 |
| JayF | vsaienk0: ^ addresesed your comments | 15:50 |
| *** jtomasek has quit IRC | 15:50 | |
| *** ametts has joined #openstack-ironic | 15:51 | |
| *** rama_y has quit IRC | 15:53 | |
| *** ksumit has quit IRC | 15:53 | |
| openstackgerrit | Graham Hayes proposed openstack/ironic: Get ready for os-api-ref sphinx theme change https://review.openstack.org/357922 | 15:54 |
| *** derekh has quit IRC | 15:56 | |
| *** wajdi has quit IRC | 15:59 | |
| *** ChrisAusten has joined #openstack-ironic | 15:59 | |
| *** wajdi has joined #openstack-ironic | 15:59 | |
| *** rbrud has joined #openstack-ironic | 16:00 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] Support names for {cleaning,provisioning}_network https://review.openstack.org/357983 | 16:02 |
| *** ChrisAusten has quit IRC | 16:09 | |
| mat128 | phschwartz, sambetts: which host manager are you using for scheduler in that setup? | 16:10 |
| mat128 | phschwartz, sambetts: https://github.com/openstack/nova/blob/master/nova/conf/scheduler.py#L215 | 16:13 |
| mat128 | brb | 16:13 |
| *** mat128 is now known as mat128|afk | 16:13 | |
| sambetts | mat128: I think I used the default when I did it, because you don't want the whole using up a whole host thing for the VM side of the deployment | 16:15 |
| sambetts | mat128|afk: I basically just reset all the nova schedular settings to a standard VM openstack configuration | 16:16 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Neutron notifications https://review.openstack.org/345963 | 16:19 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Test patch that verify external events https://review.openstack.org/357995 | 16:19 |
| *** foexle has joined #openstack-ironic | 16:20 | |
| phschwartz | sambetts: so I would not do anything special scheduler wise? Just setup host aggs based on flavor? | 16:20 |
| sambetts | phschwartz: thats the way I think I did it, it was nearly a year ago I was testing this out those | 16:21 |
| sambetts | though* | 16:21 |
| sambetts | so I don't remember if I did anything specific | 16:21 |
| *** dprince has joined #openstack-ironic | 16:21 | |
| NobodyCam | Good Morning Ironic'ers | 16:21 |
| NobodyCam | and OFC T.G.I.F :p | 16:22 |
| *** amotoki_ has quit IRC | 16:22 | |
| *** fragatina has quit IRC | 16:22 | |
| *** foexle has quit IRC | 16:24 | |
| wznoinsk | hi, trying to get ironic-ui (1.1.2) working on with stable/mitaka horizon, getting [Fri Aug 19 16:23:34.506408 2016] [:error] [pid 18:tid 139734410958592] Could not process panel theme_preview: Dashboard with slug "developer" is not registered. | 16:24 |
| wznoinsk | , anyone have any suggestions? | 16:24 |
| sambetts | wznoinsk: have you tried asking in #openstack-horizon ? | 16:25 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic-lib: Extend parse_root_device_hints to support operators https://review.openstack.org/348953 | 16:25 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic-lib: Add match_root_device_hints() to the utils.py module https://review.openstack.org/358000 | 16:25 |
| wznoinsk | sambetts: not yet, horizon is working fine for last weeks, when I enable ironic_ui it breaks | 16:25 |
| lucasagomes | jroll, ^ sorry you asked later... I think that's all we need in ironic-lib | 16:25 |
| lucasagomes | jroll, now I just need to change IPA and Ironic to use the new methods | 16:25 |
| jroll | lucasagomes: ok you tested that and you're happy with it? | 16:27 |
| jroll | betherly: if you're around could you help out wznoinsk ? | 16:27 |
| *** jpich has quit IRC | 16:28 | |
| lucasagomes | jroll, tested partially, I'm modifying IPA now | 16:28 |
| jroll | lucasagomes: ok cool | 16:28 |
| jroll | thanks | 16:28 |
| sambetts | wznoinsk: if you head over the #openstack-horizon robcresswell might be able to help you out | 16:30 |
| robcresswell | o/ | 16:30 |
| robcresswell | Dont worry, lurking here sambetts :p | 16:30 |
| TheJulia | wznoinsk: betherly and ppiela are the two people that know the ironic-ui best. | 16:30 |
| vdrok | morning NobodyCam ! | 16:31 |
| NobodyCam | morning vdrok :) | 16:31 |
| robcresswell | wznoinsk: I'm just looking into it now | 16:33 |
| robcresswell | wznoinsk: Stack should take another few minutes then I'll have an answer. | 16:33 |
| wznoinsk | robcresswell: thanks | 16:34 |
| ppiela | wznoinsk: If you remove the _2200_ironic.py file in the enabled directory does Horizon work correctly? | 16:36 |
| wznoinsk | ppiela: yes | 16:36 |
| wznoinsk | I might have installed master of ironic-ui on it before, maybe something left in some configuration somwhere? | 16:37 |
| *** rama_y has joined #openstack-ironic | 16:39 | |
| ppiela | wznoinsk: I dont think so | 16:39 |
| *** ohamada has quit IRC | 16:39 | |
| openstackgerrit | Vladyslav Drok proposed openstack/python-ironicclient: Add docs target to tox.ini https://review.openstack.org/358011 | 16:41 |
| vdrok | jroll: just noticed that^^ | 16:41 |
| vdrok | JayF: ^^ | 16:42 |
| robcresswell | wznoinsk: So this is 1.1.2 on stable/mitaka, correct? | 16:42 |
| *** krotscheck is now known as the_other_clarkb | 16:42 | |
| wznoinsk | robcresswell: pip freeze | grep ironic-ui | 16:43 |
| wznoinsk | -e git+https://git.openstack.org/openstack/ironic-ui@09f0b88cc561e4439f1d3243ad8c2bb399a02d83#egg=ironic_ui | 16:43 |
| wznoinsk | ironic-ui (1.1.2, /horizon-source/horizon-9.0.0/ironic-ui) | 16:43 |
| robcresswell | so... yes :) | 16:43 |
| *** the_other_clarkb is now known as krotscheck | 16:43 | |
| openstackgerrit | Merged openstack/bifrost: Use upper-constraints for all tox targets https://review.openstack.org/357836 | 16:44 |
| wznoinsk | I will only add openstack was deployed using kolla stable/mitaka, shouldn't make a diff tho right? | 16:44 |
| *** fragatina has joined #openstack-ironic | 16:45 | |
| devananda | morning, all | 16:45 |
| *** Goneri has joined #openstack-ironic | 16:45 | |
| sambetts | Morning devananda | 16:45 |
| robcresswell | It shouldn't matter. is DEBUG True? Or is this without debug on? Trying to understand why its attempting to interact with the dev panel. | 16:46 |
| robcresswell | wznoinsk: ^^ | 16:46 |
| vdrok | morning devananda | 16:47 |
| wznoinsk | I've enabled debug in local_settings and restart, and ... started working ;-) | 16:48 |
| wznoinsk | with that difference the content of the panel is empty for Ironic.. | 16:48 |
| openstackgerrit | Merged openstack/bifrost: Add SUSE support in scripts/env-setup.sh https://review.openstack.org/356964 | 16:48 |
| *** athomas has quit IRC | 16:49 | |
| openstackgerrit | Merged openstack/bifrost: Fix /etc/hosts before starting the rabbitmq server https://review.openstack.org/357718 | 16:49 |
| robcresswell | Okay, so something is bugged about the handling of that panel. *makes note* | 16:50 |
| *** fragatina has quit IRC | 16:51 | |
| openstackgerrit | Merged openstack/bifrost: Restore stable-2.0 as the default Ansible branch https://review.openstack.org/357268 | 16:52 |
| *** fragatina has joined #openstack-ironic | 16:52 | |
| *** tesseract- has quit IRC | 16:53 | |
| *** amotoki has joined #openstack-ironic | 16:55 | |
| *** aNupoisc has joined #openstack-ironic | 16:58 | |
| *** ppiela has quit IRC | 17:02 | |
| *** anush_ has quit IRC | 17:02 | |
| *** ppiela has joined #openstack-ironic | 17:04 | |
| NobodyCam | morning devananda :) | 17:07 |
| vdrok | jroll: idk the reason but we do generate the docs for unittests everywhere else | 17:09 |
| openstackgerrit | Tim Burke proposed openstack/ironic: Pass swiftclient header values as strings https://review.openstack.org/358020 | 17:10 |
| jroll | vdrok: yeah, it's kinda silly :D | 17:10 |
| jroll | I'm fine as-is though, thanks for that | 17:10 |
| vdrok | base class docs may be useful but yeah | 17:10 |
| vdrok | np :) | 17:10 |
| lucasagomes | folks I'm calling it a day | 17:10 |
| sambetts | o/ lucasagomes | 17:10 |
| lucasagomes | have a great evening all, enjoy the weekend | 17:10 |
| vdrok | night lucasagomes | 17:10 |
| vdrok | you too! | 17:10 |
| lucasagomes | jroll, I think the patches are good, but I will finish the tests on monday | 17:11 |
| lucasagomes | testing root device hitns is a PITA | 17:11 |
| TheJulia | goodnight lucasagomes | 17:11 |
| lucasagomes | too many permutations | 17:11 |
| jroll | lucasagomes: ok cool, I'm out tuesday/wednesday so was thinking about release monday | 17:11 |
| jroll | but I can have dtantsur handle releases | 17:11 |
| jroll | and we can get that in | 17:12 |
| sambetts | scumbag jroll releases code then runs away ;) | 17:12 |
| jroll | no better way to do it! | 17:12 |
| sambetts | haha | 17:12 |
| * dtantsur can handle releases for sure :) | 17:12 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Use upper-constraints for all tox targets https://review.openstack.org/357838 | 17:12 |
| lucasagomes | jroll, cool, no worries :D | 17:14 |
| lucasagomes | jroll, we can release multiple times as well right? | 17:14 |
| jroll | lucasagomes: in one week? sure, but no need | 17:15 |
| jroll | I'm not going to release one monday if we have something in the queue | 17:16 |
| lucasagomes | cool | 17:16 |
| lucasagomes | yeah I wasn't meaning multiple releases in a week, but in few weeks time | 17:16 |
| jroll | oh | 17:16 |
| jroll | next week is non-client library freeze | 17:16 |
| lucasagomes | anyway, appreciate if we can hold it few days but I will try to finish it up on monday anyway | 17:16 |
| lucasagomes | oh right | 17:16 |
| jroll | so that's stable/newton for ironic-lib | 17:16 |
| lucasagomes | gotcha | 17:16 |
| lucasagomes | yeah I will hurry with it | 17:16 |
| jroll | cool | 17:17 |
| jroll | I mean, we could also just bump the whole feature to ocata | 17:17 |
| jroll | plenty of other things to do :) | 17:17 |
| lucasagomes | right, we will see cause it's pretty close | 17:18 |
| *** _vsaienko has joined #openstack-ironic | 17:18 | |
| lucasagomes | g'night all | 17:18 |
| *** lucasagomes is now known as lucas-afk | 17:18 | |
| jroll | yeah sounds good | 17:18 |
| jroll | night | 17:18 |
| * jroll steps away for a while | 17:18 | |
| openstackgerrit | Merged openstack/python-ironicclient: Use constraints for all the things https://review.openstack.org/318672 | 17:19 |
| JayF | given we're trying to release ironic-lib by monday, I'd /really/ appreciate reviews on https://review.openstack.org/#/c/356796/ so it can land before release | 17:19 |
| JayF | and I'll go work on project-config patches for ironic-lib docs build now | 17:19 |
| jroll | no, we're going to release it later next week, so we have time for lucas' thing | 17:20 |
| jroll | :) | 17:20 |
| jroll | but we should still hurry up and finish that, agree | 17:20 |
| mat128|afk | JayF: reviewing | 17:20 |
| JayF | shhh don't encourage people to not review my stuff :P | 17:20 |
| jroll | that's not my intention | 17:20 |
| jroll | you already got an rloo so that's one +2 next week | 17:20 |
| openstackgerrit | Merged openstack/bifrost: Only set hostname on 127.0.0.1 if not present in /etc/hosts https://review.openstack.org/355778 | 17:21 |
| jroll | anyway, bbl | 17:21 |
| robcresswell | sambetts: ironic-ui problem solved boss :) | 17:21 |
| sambetts | robcresswell: awesome :) thanks dude | 17:21 |
| *** electrofelix has quit IRC | 17:24 | |
| *** _vsaienko has quit IRC | 17:27 | |
| * sambetts is calling it a day | 17:27 | |
| sambetts | night every one, see y'all Monday | 17:28 |
| *** sambetts is now known as sambetts|afk | 17:28 | |
| devananda | g'night sambetts|afk, lucas-afk ! | 17:28 |
| *** _vsaienko has joined #openstack-ironic | 17:28 | |
| openstackgerrit | Merged openstack/ironic: Update code to use Pike as the code name https://review.openstack.org/357334 | 17:28 |
| NobodyCam | night sambetts|afk | 17:29 |
| NobodyCam | have a good weekend | 17:29 |
| *** baoli has quit IRC | 17:29 | |
| openstackgerrit | Vladyslav Drok proposed openstack/python-ironicclient: Add docs for create command https://review.openstack.org/358025 | 17:29 |
| vdrok | if someone is willing to do yet another doc review :) ^^ | 17:29 |
| vdrok | night sambetts|afk | 17:29 |
| * dtantsur leaves as well | 17:30 | |
| *** bfournie has quit IRC | 17:30 | |
| dtantsur | have a nice weekend | 17:30 |
| NobodyCam | have a great weekend dtantsur | 17:30 |
| *** baoli has joined #openstack-ironic | 17:30 | |
| vdrok | night dtantsur | 17:31 |
| vdrok | time for europe to leave :) | 17:31 |
| vdrok | good night everyone! | 17:31 |
| NobodyCam | night vdrok have a great weekend too :) | 17:32 |
| *** dtantsur is now known as dtantsur|afk | 17:36 | |
| *** baoli has quit IRC | 17:40 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic-specs: Boot from Volume - Reference Drivers https://review.openstack.org/294995 | 17:41 |
| *** baoli has joined #openstack-ironic | 17:46 | |
| *** ElCoyote_ has left #openstack-ironic | 17:47 | |
| *** e0ne has joined #openstack-ironic | 17:55 | |
| openstackgerrit | Fabio Dassan proposed openstack/pyghmi: Fix raid driver bitmask https://review.openstack.org/358031 | 17:55 |
| xavierr | does OSCli has a command for validate node interfaces? | 17:55 |
| *** baoli has quit IRC | 17:56 | |
| *** baoli has joined #openstack-ironic | 17:56 | |
| *** bfournie has joined #openstack-ironic | 18:02 | |
| *** Sukhdev has joined #openstack-ironic | 18:08 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic: [WIP] Add storage_interface to base driver class https://review.openstack.org/348006 | 18:08 |
| *** Goneri has quit IRC | 18:11 | |
| *** amotoki has quit IRC | 18:16 | |
| openstackgerrit | Vasyl Saienko proposed openstack/python-ironicclient: Add create external event functionality https://review.openstack.org/354770 | 18:16 |
| *** baoli has quit IRC | 18:21 | |
| *** e0ne has quit IRC | 18:23 | |
| *** vishwanathj_zzz is now known as vishwanathj | 18:23 | |
| *** baoli has joined #openstack-ironic | 18:26 | |
| *** baoli has quit IRC | 18:33 | |
| openstackgerrit | Thiago Paiva Brito proposed openstack/ironic: [WIP] Reusing oneview_client when possible https://review.openstack.org/358041 | 18:35 |
| thiagop | TheJulia: would like you to take a look on this before carry on^ | 18:35 |
| thiagop | if possible :) | 18:35 |
| NobodyCam | morning thiagop :) | 18:36 |
| thiagop | #NOTE: OneView CI is down due to underlying infra issues | 18:36 |
| thiagop | morning NobodyCam | 18:36 |
| *** baoli has joined #openstack-ironic | 18:36 | |
| * thiagop is beginning to get pissed with the lab's cloud | 18:36 | |
| NobodyCam | :( | 18:37 |
| *** sambetts|afk has quit IRC | 18:42 | |
| *** sambetts has joined #openstack-ironic | 18:44 | |
| *** amotoki has joined #openstack-ironic | 18:47 | |
| *** amotoki has quit IRC | 18:50 | |
| openstackgerrit | Ricardo Carrillo Cruz proposed openstack/bifrost: Remove 'auth' fact initialization from bifrost-deploy-nodes-dynamic https://review.openstack.org/354626 | 18:55 |
| rcarrillocruz | NobodyCam: heya, thx for the review, i amended ^ | 18:56 |
| *** amotoki has joined #openstack-ironic | 18:56 | |
| NobodyCam | hey rcarrillocruz :) will take a look at that in a bit | 18:56 |
| NobodyCam | yea sorry for the -1 on it... | 18:57 |
| NobodyCam | but better to have the link | 18:57 |
| *** aNupoisc has quit IRC | 19:04 | |
| *** amotoki has quit IRC | 19:04 | |
| *** aNupoisc has joined #openstack-ironic | 19:04 | |
| *** Goneri has joined #openstack-ironic | 19:04 | |
| rcarrillocruz | nah, it's ok | 19:07 |
| rcarrillocruz | i remembered lp was ok | 19:07 |
| rcarrillocruz | BUT | 19:07 |
| rcarrillocruz | it needs a space between lp and the number | 19:07 |
| rcarrillocruz | :/ | 19:08 |
| rcarrillocruz | also, i think lp works, LP won't | 19:08 |
| rcarrillocruz | i'd need to check jeepyb update_bug regex handling | 19:08 |
| *** rama_y has quit IRC | 19:15 | |
| NobodyCam | :) | 19:15 |
| * TheJulia pokes gerrit | 19:18 | |
| NobodyCam | ??? with a stick ??? :p | 19:19 |
| jroll | TheJulia: agree | 19:19 |
| thiagop | is gerrit dead? | 19:19 |
| TheJulia | I think only infra has the sharpened sticks, so only they know | 19:20 |
| jlvillal | TheJulia: same here | 19:21 |
| jlvillal | And we are back... | 19:22 |
| TheJulia | thiagop: at a glance, it looks super strait forward. Have you guys managed to benchmark it yet? | 19:23 |
| thiagop | TheJulia: not yet, but we will. The python-oneviewclient patch to get a new session when the token expires merged yesterday | 19:25 |
| TheJulia | Ahhh | 19:25 |
| thiagop | TheJulia: aaaaaaand the CI is down... :'( | 19:25 |
| TheJulia | thiagop: Well, basically looks like what I would expect for re-use since you have the same base credentials for oneview. I'm really interested in benchmark results :) | 19:27 |
| thiagop | TheJulia: sure, the guys from benchmark will work on this as soon as they finish measuring the current setup for the remaining cases | 19:28 |
| TheJulia | Excellent | 19:29 |
| *** _vsaienko has quit IRC | 19:29 | |
| *** Goneri has quit IRC | 19:32 | |
| *** _vsaienko has joined #openstack-ironic | 19:33 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Test patch that verify external events https://review.openstack.org/357995 | 19:33 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Fix object save after refresh failure https://review.openstack.org/354744 | 19:38 |
| *** sdake has quit IRC | 19:41 | |
| *** hpe-hj has quit IRC | 19:42 | |
| *** hjanssen-hpe has quit IRC | 19:42 | |
| *** jcoufal has quit IRC | 19:44 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic: Updated from global requirements https://review.openstack.org/357697 | 19:47 |
| *** ametts has quit IRC | 19:48 | |
| mari0jv | TheJulia: jlvillal: same here | 19:59 |
| jlvillal | mari0jv: Oh hanging again? It started working again after I sent that. | 19:59 |
| mari0jv | now it's fine | 19:59 |
| mari0jv | heisenbug | 19:59 |
| TheJulia | That or garbage collection | 20:00 |
| *** jcoufal has joined #openstack-ironic | 20:06 | |
| openstackgerrit | Merged openstack/ironic: Fix AgentDeploy take_over() docstring https://review.openstack.org/357076 | 20:09 |
| *** mtanino has quit IRC | 20:10 | |
| openstackgerrit | Merged openstack/ironic: Remove incorrect CONF.import_opt() from test_ipmitool.py https://review.openstack.org/353907 | 20:10 |
| jroll | who wants to land adoption tempest tests? https://review.openstack.org/#/c/344975/5 | 20:14 |
| *** baoli has quit IRC | 20:17 | |
| openstackgerrit | Ricardo Araújo Santos proposed openstack/ironic: Adds instructions to perform nodes migration https://review.openstack.org/357454 | 20:18 |
| *** rama_y has joined #openstack-ironic | 20:28 | |
| NobodyCam | jroll: did i read that we are planning a ironic-lib release next week? | 20:29 |
| jroll | NobodyCam: yes, next week is non-client library freeze | 20:30 |
| NobodyCam | :) sweet :) looking for https://review.openstack.org/#/c/230924 | 20:31 |
| jroll | oh, why didn't you say so? | 20:31 |
| jroll | we can release often :P | 20:31 |
| NobodyCam | :p hehehehe :) | 20:31 |
| jroll | alright, I'm out for the weekend | 20:33 |
| jroll | y'all have a good one :) | 20:33 |
| NobodyCam | have a great weekend jroll :) | 20:33 |
| jroll | you too! | 20:33 |
| *** dprince has quit IRC | 20:35 | |
| *** aNupoisc has quit IRC | 20:39 | |
| *** moshele has joined #openstack-ironic | 20:50 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Build socket list right before select call https://review.openstack.org/267537 | 20:51 |
| *** ccamacho|away has quit IRC | 20:52 | |
| *** aNupoisc has joined #openstack-ironic | 20:59 | |
| *** fragatin_ has joined #openstack-ironic | 21:00 | |
| openstackgerrit | Thiago Paiva Brito proposed openstack/ironic: Bring configurations from tempest to ironic_tempest_plugin https://review.openstack.org/355591 | 21:02 |
| openstackgerrit | Thiago Paiva Brito proposed openstack/ironic: Bring configurations from tempest to ironic_tempest_plugin https://review.openstack.org/355591 | 21:03 |
| *** fragatina has quit IRC | 21:04 | |
| *** wajdi_ has joined #openstack-ironic | 21:05 | |
| thiagop | Folks, I'm calling it a day | 21:05 |
| thiagop | have a nice weekend | 21:05 |
| thiagop | see ya | 21:05 |
| *** thiagop has quit IRC | 21:05 | |
| *** clenimar has quit IRC | 21:05 | |
| *** jcoufal has quit IRC | 21:08 | |
| openstackgerrit | Mario Villaplana proposed openstack/ironic: Add power state change notifications https://review.openstack.org/321865 | 21:08 |
| openstackgerrit | Mario Villaplana proposed openstack/ironic: Add notification base classes and docs https://review.openstack.org/298461 | 21:08 |
| *** wajdi has quit IRC | 21:08 | |
| *** wajdi_ has quit IRC | 21:09 | |
| openstackgerrit | Merged openstack/python-ironicclient: Trivial: Fix doc string for class DeleteBaremetalNode https://review.openstack.org/357731 | 21:20 |
| JayF | mari0jv: do you have a patch up to update the notification spec? | 21:25 |
| JayF | mari0jv: mainly thinking s/notification_priority/notification_level/ | 21:25 |
| mari0jv | JayF: already done | 21:27 |
| mari0jv | oh, not on the spec | 21:28 |
| mari0jv | in the code, yes | 21:28 |
| mari0jv | rloo commented about it and i agreed so switched it | 21:28 |
| *** michauds has quit IRC | 21:30 | |
| JayF | yeah; we usually try to make the spec reflect reality for those | 21:30 |
| JayF | if you wanna push a small patch up reflecting that and any other minor changes | 21:31 |
| *** cdearborn has quit IRC | 21:31 | |
| mari0jv | sure | 21:31 |
| *** sdake has joined #openstack-ironic | 21:35 | |
| *** sdake has quit IRC | 21:37 | |
| *** sdake has joined #openstack-ironic | 21:38 | |
| mari0jv | JayF: actually the spec already says notification_level, maybe someone else updated it | 21:39 |
| *** fragatin_ has quit IRC | 21:39 | |
| JayF | perfect | 21:39 |
| JayF | \o/ | 21:39 |
| *** fragatina has joined #openstack-ironic | 21:39 | |
| *** _vsaienko has quit IRC | 21:40 | |
| *** foexle has joined #openstack-ironic | 21:43 | |
| *** baoli has joined #openstack-ironic | 21:47 | |
| *** baoli has quit IRC | 21:54 | |
| openstackgerrit | Merged openstack/pyghmi: Fix raid driver bitmask https://review.openstack.org/358031 | 21:55 |
| *** causten has quit IRC | 21:56 | |
| *** lascii is now known as alaski | 22:02 | |
| jlvillal | mari0jv: Now I have a gerrit hang. . . | 22:27 |
| mari0jv | :( | 22:27 |
| mari0jv | same jlvillal | 22:27 |
| mari0jv | does infra know about this | 22:28 |
| mari0jv | ? | 22:28 |
| jlvillal | :( | 22:28 |
| *** rbrud has quit IRC | 22:28 | |
| jlvillal | I just mentioned it over in infra | 22:28 |
| mari0jv | same | 22:29 |
| mari0jv | i'll be heading off since there's no gerrit, have an excellent weekend ironic | 22:31 |
| *** mari0jv has quit IRC | 22:31 | |
| *** rbudden has quit IRC | 22:38 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: DO NOT REVIEW https://review.openstack.org/296432 | 22:53 |
| *** ppiela has quit IRC | 23:08 | |
| *** amotoki has joined #openstack-ironic | 23:09 | |
| *** boris-42 has quit IRC | 23:12 | |
| *** patrickeast has quit IRC | 23:12 | |
| *** izaakk has quit IRC | 23:14 | |
| *** Goneri has joined #openstack-ironic | 23:17 | |
| *** boris-42 has joined #openstack-ironic | 23:18 | |
| *** patrickeast has joined #openstack-ironic | 23:18 | |
| *** izaakk has joined #openstack-ironic | 23:18 | |
| *** amotoki has quit IRC | 23:23 | |
| *** moshele has quit IRC | 23:24 | |
| *** jcoufal has joined #openstack-ironic | 23:34 | |
| *** thrash is now known as thrash|g0ne | 23:40 | |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Enable OSProfiler support in Ironic https://review.openstack.org/358142 | 23:42 |
| *** syed_ has quit IRC | 23:42 | |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Enable OSProfiler support in Ironic https://review.openstack.org/358142 | 23:42 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!