| *** amotoki has joined #openstack-ironic | 00:00 | |
| *** jaybeale has quit IRC | 00:03 | |
| *** gfhellma has quit IRC | 00:04 | |
| *** jaybeale has joined #openstack-ironic | 00:04 | |
| openstackgerrit | Merged openstack/ironic: IPMITool: add IPMISocatConsole and IPMIConsole class https://review.openstack.org/293873 | 00:05 |
|---|---|---|
| *** amotoki has quit IRC | 00:06 | |
| *** oomichi has quit IRC | 00:06 | |
| *** amotoki has joined #openstack-ironic | 00:12 | |
| *** yuikotakadamori has joined #openstack-ironic | 00:23 | |
| *** fragatina has quit IRC | 00:26 | |
| *** mtanin___ has joined #openstack-ironic | 00:28 | |
| *** harshs has quit IRC | 00:29 | |
| *** mtanino has quit IRC | 00:30 | |
| *** Sukhdev has quit IRC | 00:30 | |
| *** amotoki has quit IRC | 00:30 | |
| *** abc_xyz has joined #openstack-ironic | 00:30 | |
| *** abc_xyz has quit IRC | 00:32 | |
| *** jaybeale has quit IRC | 00:32 | |
| *** abcxyz has quit IRC | 00:32 | |
| *** hoangcx has joined #openstack-ironic | 00:32 | |
| *** Sukhdev has joined #openstack-ironic | 00:34 | |
| *** ahqa_ has quit IRC | 00:36 | |
| *** jaybeale has joined #openstack-ironic | 00:37 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic: Fix rendering for version 1.14 https://review.openstack.org/347084 | 00:39 |
| *** sdake has joined #openstack-ironic | 00:39 | |
| *** rloo has quit IRC | 00:39 | |
| *** jaybeale has quit IRC | 00:41 | |
| *** ayoung has joined #openstack-ironic | 00:45 | |
| *** harshs has joined #openstack-ironic | 00:50 | |
| *** sdake_ has joined #openstack-ironic | 00:50 | |
| *** spandhe has quit IRC | 00:51 | |
| *** sdake has quit IRC | 00:52 | |
| *** harshs has quit IRC | 00:52 | |
| *** sdake has joined #openstack-ironic | 00:55 | |
| *** sdake_ has quit IRC | 00:55 | |
| *** ayoung has quit IRC | 01:00 | |
| *** jaybeale has joined #openstack-ironic | 01:06 | |
| *** jaybeale has quit IRC | 01:08 | |
| devananda | shameless plug for reviews: https://review.openstack.org/#/c/325599/ keystone policy support, already has a +2 | 01:09 |
| devananda | there's a couple smaller follow on patches, as well | 01:09 |
| TheJulia | heh | 01:10 |
| TheJulia | I'll look at it in the morning :) I fried my brain for code review hours ago | 01:10 |
| *** amotoki has joined #openstack-ironic | 01:10 | |
| *** ijw has joined #openstack-ironic | 01:12 | |
| *** joprovost has joined #openstack-ironic | 01:13 | |
| devananda | thanks :) | 01:14 |
| *** clenimar_ has joined #openstack-ironic | 01:14 | |
| *** clenimar_ has quit IRC | 01:15 | |
| *** ijw has quit IRC | 01:17 | |
| *** mtanino has joined #openstack-ironic | 01:18 | |
| *** mtanin___ has quit IRC | 01:19 | |
| *** joprovost has quit IRC | 01:22 | |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: Update documentation for keystone policy support https://review.openstack.org/345556 | 01:23 |
| *** jcoufal has joined #openstack-ironic | 01:23 | |
| *** jcoufal_ has joined #openstack-ironic | 01:28 | |
| *** ijw has joined #openstack-ironic | 01:28 | |
| *** ijw has quit IRC | 01:30 | |
| *** jcoufal has quit IRC | 01:31 | |
| *** wajdi has joined #openstack-ironic | 01:33 | |
| *** ijw has joined #openstack-ironic | 01:33 | |
| *** hoonetorg has quit IRC | 01:35 | |
| *** anush_ has quit IRC | 01:37 | |
| *** mtanino has quit IRC | 01:44 | |
| *** harshs has joined #openstack-ironic | 01:46 | |
| *** joprovost has joined #openstack-ironic | 01:49 | |
| *** hoonetorg has joined #openstack-ironic | 01:49 | |
| *** sdake has quit IRC | 01:59 | |
| *** jcoufal_ has quit IRC | 01:59 | |
| *** jcoufal has joined #openstack-ironic | 02:00 | |
| *** catintheroof has quit IRC | 02:06 | |
| *** bvandewa has quit IRC | 02:07 | |
| openstackgerrit | Andy Yan proposed openstack/ironic-specs: lenovo driver proposal https://review.openstack.org/208319 | 02:12 |
| *** ijw has quit IRC | 02:12 | |
| *** penick has quit IRC | 02:20 | |
| *** Sukhdev has quit IRC | 02:24 | |
| *** harshs has quit IRC | 02:24 | |
| *** jaybeale has joined #openstack-ironic | 02:28 | |
| *** cfarquhar has quit IRC | 02:31 | |
| *** jcoufal_ has joined #openstack-ironic | 02:32 | |
| *** jistr has quit IRC | 02:32 | |
| *** jcoufal has quit IRC | 02:33 | |
| *** jistr has joined #openstack-ironic | 02:33 | |
| *** jcoufal_ has quit IRC | 02:38 | |
| *** joprovost1 has joined #openstack-ironic | 02:41 | |
| *** joprovost has quit IRC | 02:42 | |
| *** joprovost1 is now known as joprovost | 02:42 | |
| *** Nisha_away has joined #openstack-ironic | 02:47 | |
| *** bvandewa has joined #openstack-ironic | 02:48 | |
| *** bvandewa has quit IRC | 02:53 | |
| *** harshs has joined #openstack-ironic | 02:58 | |
| *** harshs has quit IRC | 02:58 | |
| *** cfarquhar has joined #openstack-ironic | 03:01 | |
| *** cfarquhar has quit IRC | 03:01 | |
| *** cfarquhar has joined #openstack-ironic | 03:01 | |
| *** harshs has joined #openstack-ironic | 03:06 | |
| *** Nisha_away has quit IRC | 03:08 | |
| *** siva_krish has joined #openstack-ironic | 03:10 | |
| *** Nisha_away has joined #openstack-ironic | 03:17 | |
| *** jaybeale has quit IRC | 03:23 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Enhance ilo drivers to do inband inspection https://review.openstack.org/201904 | 03:35 |
| *** bvandewa has joined #openstack-ironic | 03:42 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Add new out-of-band properties to ilo drivers https://review.openstack.org/338138 | 03:42 |
| *** xavierr has quit IRC | 03:46 | |
| *** xavierr has joined #openstack-ironic | 03:46 | |
| *** bvandewa has quit IRC | 03:47 | |
| *** bvandewa has joined #openstack-ironic | 03:48 | |
| *** PollyZ has joined #openstack-ironic | 03:51 | |
| *** appu103_ has joined #openstack-ironic | 03:53 | |
| *** links has joined #openstack-ironic | 04:03 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic: Fix iLO drivers inconsistent boot mode default value https://review.openstack.org/345231 | 04:05 |
| *** PollyZ has quit IRC | 04:05 | |
| *** amotoki has quit IRC | 04:07 | |
| *** bvandewa has quit IRC | 04:12 | |
| *** harshs has quit IRC | 04:12 | |
| *** harshs has joined #openstack-ironic | 04:14 | |
| *** joprovost has quit IRC | 04:16 | |
| *** bvandewa has joined #openstack-ironic | 04:21 | |
| *** Sukhdev has joined #openstack-ironic | 04:25 | |
| *** Sukhdev has quit IRC | 04:29 | |
| *** spandhe has joined #openstack-ironic | 04:31 | |
| *** harlowja_at_home has joined #openstack-ironic | 04:33 | |
| *** harshs has quit IRC | 04:34 | |
| *** pcaruana has quit IRC | 04:34 | |
| *** jtomasek has quit IRC | 04:39 | |
| *** appu103_ is now known as aparnav | 04:44 | |
| *** harshs has joined #openstack-ironic | 04:45 | |
| *** amotoki has joined #openstack-ironic | 04:46 | |
| *** harshs has quit IRC | 04:55 | |
| *** jtomasek has joined #openstack-ironic | 04:58 | |
| *** harshs has joined #openstack-ironic | 05:00 | |
| *** harlowja_at_home has quit IRC | 05:01 | |
| *** amotoki has quit IRC | 05:10 | |
| *** siva_krish has quit IRC | 05:14 | |
| *** amotoki has joined #openstack-ironic | 05:17 | |
| *** gabriel-bezerra has quit IRC | 05:23 | |
| *** gabriel-bezerra has joined #openstack-ironic | 05:24 | |
| *** bvandewa has quit IRC | 05:26 | |
| *** bvandewa has joined #openstack-ironic | 05:29 | |
| *** amotoki has quit IRC | 05:35 | |
| *** harshs has quit IRC | 05:36 | |
| *** jtomasek has quit IRC | 05:38 | |
| *** ptrlv has quit IRC | 05:38 | |
| *** ptrlv has joined #openstack-ironic | 05:39 | |
| *** ChubYann has quit IRC | 05:43 | |
| *** gabriel-bezerra has quit IRC | 05:46 | |
| *** gabriel-bezerra has joined #openstack-ironic | 05:46 | |
| *** amotoki has joined #openstack-ironic | 05:54 | |
| openstackgerrit | Andy Yan proposed openstack/ironic-specs: lenovo driver proposal https://review.openstack.org/208319 | 05:55 |
| *** ptrlv has quit IRC | 05:59 | |
| *** ptrlv has joined #openstack-ironic | 05:59 | |
| *** skramaja_afk is now known as skramaja | 06:00 | |
| *** amotoki has quit IRC | 06:04 | |
| openstackgerrit | Tang Chen proposed openstack/ironic: Move "server_profile_template_uri" to REQUIRED_ON_PROPERTIES https://review.openstack.org/347127 | 06:04 |
| *** mjura has joined #openstack-ironic | 06:05 | |
| *** tangchen has joined #openstack-ironic | 06:06 | |
| *** gabriel-bezerra has quit IRC | 06:06 | |
| *** gabriel-bezerra has joined #openstack-ironic | 06:07 | |
| *** fumie has joined #openstack-ironic | 06:11 | |
| *** amotoki has joined #openstack-ironic | 06:12 | |
| *** Sukhdev has joined #openstack-ironic | 06:12 | |
| *** mjura has quit IRC | 06:13 | |
| *** mjura has joined #openstack-ironic | 06:13 | |
| *** milan has joined #openstack-ironic | 06:15 | |
| *** hshiina has joined #openstack-ironic | 06:26 | |
| *** ptrlv has quit IRC | 06:28 | |
| *** ptrlv has joined #openstack-ironic | 06:35 | |
| *** bks has joined #openstack-ironic | 06:36 | |
| *** vmud213 has joined #openstack-ironic | 06:38 | |
| *** tesseract- has joined #openstack-ironic | 06:39 | |
| *** spandhe has quit IRC | 06:39 | |
| *** moshele has joined #openstack-ironic | 06:42 | |
| *** yibo has quit IRC | 06:44 | |
| *** yibo has joined #openstack-ironic | 06:44 | |
| *** Sukhdev has quit IRC | 06:46 | |
| *** tangchen_ has quit IRC | 06:46 | |
| *** tangchen has quit IRC | 06:46 | |
| *** pcaruana has joined #openstack-ironic | 06:55 | |
| *** tangchen_ has joined #openstack-ironic | 06:57 | |
| *** tangchen has joined #openstack-ironic | 06:57 | |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 07:00 |
| *** skramaja is now known as skramaja_afk | 07:03 | |
| *** skramaja_afk is now known as skramaja | 07:05 | |
| *** rcernin has joined #openstack-ironic | 07:06 | |
| *** milan has quit IRC | 07:21 | |
| *** skramaja is now known as skramaja_afk | 07:28 | |
| *** ifarkas_afk is now known as ifarkas | 07:28 | |
| *** milan has joined #openstack-ironic | 07:35 | |
| *** rbartal has joined #openstack-ironic | 07:36 | |
| *** jpich has joined #openstack-ironic | 07:40 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic: Fix iLO drivers inconsistent boot mode default value https://review.openstack.org/345231 | 07:40 |
| milan | #pixiesay Good morning, Ironic! -m flexing | 07:43 |
| *** bvandewa has quit IRC | 07:52 | |
| *** bvandewa has joined #openstack-ironic | 07:53 | |
| openstackgerrit | Galyna Zholtkevych proposed openstack/ironic: Deny chassis with too long description https://review.openstack.org/345248 | 07:56 |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-ironic | 08:00 | |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: Add a simple smoke test to be run in the grenade gate https://review.openstack.org/336532 | 08:02 |
| *** Nisha_away has quit IRC | 08:03 | |
| *** milan has quit IRC | 08:05 | |
| *** jpich has quit IRC | 08:07 | |
| *** fragatina has joined #openstack-ironic | 08:11 | |
| *** fragatina has quit IRC | 08:11 | |
| *** fragatina has joined #openstack-ironic | 08:12 | |
| *** bvandewa has quit IRC | 08:13 | |
| *** bvandewa has joined #openstack-ironic | 08:26 | |
| *** jpich has joined #openstack-ironic | 08:27 | |
| *** derekh has joined #openstack-ironic | 08:29 | |
| *** Nisha_away has joined #openstack-ironic | 08:32 | |
| *** jtomasek has joined #openstack-ironic | 08:33 | |
| *** skramaja_afk is now known as skramaja | 08:35 | |
| *** e0ne has joined #openstack-ironic | 08:40 | |
| *** Goneri has joined #openstack-ironic | 08:52 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic-lib: Add parse_root_device_hints to utils.py https://review.openstack.org/346068 | 08:52 |
| *** electrofelix has joined #openstack-ironic | 08:55 | |
| *** e0ne has quit IRC | 08:55 | |
| *** e0ne has joined #openstack-ironic | 08:56 | |
| *** ljjjustin has quit IRC | 08:57 | |
| *** amotoki has quit IRC | 08:57 | |
| *** milan has joined #openstack-ironic | 08:58 | |
| *** amotoki has joined #openstack-ironic | 08:59 | |
| *** ljjjustin has joined #openstack-ironic | 09:00 | |
| openstackgerrit | Tang Chen proposed openstack/ironic: Move "server_profile_template_uri" to REQUIRED_ON_PROPERTIES https://review.openstack.org/347127 | 09:03 |
| *** amotoki has quit IRC | 09:09 | |
| *** amotoki has joined #openstack-ironic | 09:11 | |
| *** mgould|afk is now known as mgould | 09:13 | |
| mgould | morning Ironic! | 09:13 |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic: Fix iLO drivers inconsistent boot mode default value https://review.openstack.org/345231 | 09:18 |
| *** hshiina has quit IRC | 09:23 | |
| openstackgerrit | Galyna Zholtkevych proposed openstack/ironic: Updated tests for db migration scripts https://review.openstack.org/344246 | 09:24 |
| *** mbound has joined #openstack-ironic | 09:25 | |
| *** e0ne has quit IRC | 09:26 | |
| *** stendulker has joined #openstack-ironic | 09:27 | |
| *** bvandewa has quit IRC | 09:30 | |
| sambetts|afk | Morning all | 09:31 |
| *** sambetts|afk is now known as sambetts | 09:32 | |
| mgould | morning sambetts | 09:33 |
| *** e0ne has joined #openstack-ironic | 09:34 | |
| sambetts | o/ mgould | 09:34 |
| lucasagomes | morning | 09:36 |
| mgould | morning lucasagomes | 09:40 |
| openstackgerrit | Imre Farkas proposed openstack/ironic: DRAC RAID configuration https://review.openstack.org/176319 | 09:41 |
| sambetts | hi lucasagomes | 09:41 |
| *** vnogin has joined #openstack-ironic | 09:44 | |
| *** hwcomcn has joined #openstack-ironic | 09:49 | |
| mgould | milan: https://review.openstack.org/#/c/346923/ | 09:49 |
| *** hwcomcn has quit IRC | 09:52 | |
| *** hwcomcn has joined #openstack-ironic | 09:53 | |
| *** lennyb has quit IRC | 09:54 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add CLI tests to check JSON response body https://review.openstack.org/291239 | 09:58 |
| vdrok | good morning ironic, milan mgould sambetts and lucasagomes ! | 10:02 |
| sambetts | hey vdrok | 10:03 |
| milan | good morning lucasagomes mgould sambetts vdrok :) | 10:03 |
| *** hoangcx has quit IRC | 10:05 | |
| milan | mgould, +1 on the documentation request :) | 10:09 |
| *** Goneri has quit IRC | 10:13 | |
| *** hshiina has joined #openstack-ironic | 10:17 | |
| tangchen | Hi guys, begging for review of this one. https://review.openstack.org/#/c/243430/ | 10:17 |
| tangchen | Thanks. :) | 10:17 |
| yibo | lucasagomes, would you check https://bugs.launchpad.net/ironic/+bug/1582964. is it possible to add rfe-approved? thanks. | 10:17 |
| openstack | Launchpad bug 1582964 in Ironic "[RFE] Support multi arch deployments" [Wishlist,In progress] - Assigned to Yibo Cai (cyb70289) | 10:17 |
| mgould | morning vdrok | 10:19 |
| *** vmud213 has quit IRC | 10:22 | |
| *** Goneri has joined #openstack-ironic | 10:23 | |
| *** amotoki has quit IRC | 10:24 | |
| *** amotoki has joined #openstack-ironic | 10:25 | |
| *** e0ne has quit IRC | 10:29 | |
| openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic-specs: Add CRUD notifications to Ironic https://review.openstack.org/347242 | 10:37 |
| *** e0ne has joined #openstack-ironic | 10:37 | |
| *** lennyb has joined #openstack-ironic | 10:37 | |
| *** hoangcx has joined #openstack-ironic | 10:39 | |
| *** amotoki has quit IRC | 10:41 | |
| *** vmud213 has joined #openstack-ironic | 10:42 | |
| *** e0ne has quit IRC | 10:47 | |
| *** abcxyz has joined #openstack-ironic | 10:49 | |
| *** hoangcx has quit IRC | 10:53 | |
| *** itooon has joined #openstack-ironic | 10:53 | |
| *** stendulker has quit IRC | 11:04 | |
| *** amotoki has joined #openstack-ironic | 11:07 | |
| *** skramaja is now known as skramaja_afk | 11:07 | |
| openstackgerrit | Merged openstack/ironic: Centralize config options - [api] https://review.openstack.org/309186 | 11:08 |
| *** vnogin has quit IRC | 11:09 | |
| *** amotoki has quit IRC | 11:11 | |
| openstackgerrit | Merged openstack/ironic: Centralize config options - [deploy] https://review.openstack.org/309206 | 11:13 |
| *** itooon has quit IRC | 11:13 | |
| *** trown|outtypewww is now known as trown | 11:17 | |
| moshele | sambetts: can you check my comment on https://review.openstack.org/#/c/264263/ ? does it make sense to you? or you have a better way to solve this | 11:19 |
| *** vnogin has joined #openstack-ironic | 11:22 | |
| *** Goneri has quit IRC | 11:24 | |
| *** Romanenko_K has joined #openstack-ironic | 11:26 | |
| *** skramaja_afk is now known as skramaja | 11:29 | |
| *** Goneri has joined #openstack-ironic | 11:31 | |
| *** ljjjustin has quit IRC | 11:31 | |
| *** ljjjustin has joined #openstack-ironic | 11:32 | |
| sambetts | moshele: hmm interesting, thats frustrating :/ I'm really not sure what the best way of handling it is, it just felt a little weird to me to be injecting an option at that point when we defined them all elsewhere, but I guess thats a result of how the current DHCP provider system works | 11:33 |
| *** lucasagomes is now known as lucas-hungry | 11:38 | |
| *** e0ne has joined #openstack-ironic | 11:39 | |
| moshele | sambetts: if I have a node with several ports how does ironic select the port to pxe from? | 11:40 |
| sambetts | moshele: the ports now have a pxe_enabled flag | 11:41 |
| moshele | sambetts: yes but in my debug I saw that all of them set to pxe_enabled=True | 11:42 |
| moshele | sambetts: but that can be because I use the ironic-inspecor to add all port | 11:42 |
| moshele | sambetts: Is the admin should set this flag? | 11:43 |
| *** vmud213 has quit IRC | 11:44 | |
| *** vmud213 has joined #openstack-ironic | 11:45 | |
| sambetts | the person/service enrolling the node should set that flag on the ports to define if they are the ports they want to PXE boot, the default is True though so that behaviour is the same as before this field was added | 11:46 |
| sambetts | when you create a new port | 11:46 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add baremetal port set command to OSC plugin https://review.openstack.org/347287 | 11:48 |
| *** yuikotakadamori has quit IRC | 11:50 | |
| moshele | sambetts: how about creating new method dhcp_options_for_port pxe_utils.py and move the client-id logic to it | 11:56 |
| sambetts | moshele: but where would we call that function from? the PXEBootInterface, and change the logic in there?? | 11:57 |
| sambetts | moshele: I think I remember someone else talking about adding feature that looks and feels similar to this, Allowing Ironic user defined DHCP option per port, and we'd need to implement logic like this for that too, but I can't remember how the conversation went with that | 12:00 |
| sambetts | lucas-hungry: do you remember anything about this ^ \ | 12:00 |
| *** Nisha_away has quit IRC | 12:02 | |
| *** thrash|g0ne is now known as thrash | 12:07 | |
| TheJulia | Good morning everyone | 12:07 |
| sambetts | Morning TheJulia | 12:07 |
| mgould | morning TheJulia | 12:08 |
| moshele | sambetts: I suggested it as port of the infiniband spec, but it was reject due to no other use case for this see https://review.openstack.org/#/c/265635/6/specs/approved/add-infiniband-support.rst read comments below Line 53 | 12:11 |
| moshele | s/port/part | 12:11 |
| sambetts | moshele: ah :/ | 12:15 |
| *** MattMan has quit IRC | 12:16 | |
| *** MattMan has joined #openstack-ironic | 12:16 | |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 12:19 |
| *** jjohnson2__ has joined #openstack-ironic | 12:22 | |
| sambetts | moshele: so I think I might have an idea, but you'll need to help me clarify because I've got little infiniband experience | 12:23 |
| moshele | sambetts: ok sure | 12:24 |
| sambetts | we're passing the client ident DHCP option because we want to operate using IP over infiniband right? | 12:24 |
| moshele | sambetts: yes | 12:24 |
| vdrok | good morning TheJulia and moshele | 12:24 |
| moshele | morning: vdrok | 12:25 |
| sambetts | moshele: so we'll need to have that option set on the port regardless of whether we're PXE booting or not right? to enable communication between ramdisk + ironic, and the tenant image and the internet? | 12:25 |
| *** vnogin_ has joined #openstack-ironic | 12:27 | |
| *** Goneri has quit IRC | 12:30 | |
| moshele | sambetts: we need the client-id to be able to get ip address from the dhcp-server when we are using IP over infiniband | 12:31 |
| *** vnogin has quit IRC | 12:32 | |
| *** vnogin_ has quit IRC | 12:32 | |
| *** vnogin has joined #openstack-ironic | 12:32 | |
| sambetts | moshele: so if thats the case then we always need it right? Because the dhcp providers are only used if your using PXEBoot, if you use a different type of boot they won't be called and therefore the DHCP option would never get configured so you'd never get an IP address right? | 12:33 |
| moshele | sambetts: yes | 12:34 |
| *** vnogin has quit IRC | 12:34 | |
| sambetts | if we always need it like this then I suggest that we build this logic into the network interface (flat and neutron) and not the dhcp provider | 12:35 |
| sambetts | https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/network/flat.py and https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/network/neutron.py | 12:36 |
| sambetts | and as part of add_provisioning_network and add_cleaning_network and configure_tenant_networks we should ensure this option gets set | 12:36 |
| *** lucas-hungry is now known as lucasagomes | 12:39 | |
| lucasagomes | sambetts, moshele so, nova will pick a random port from ironic when deploying, I'm not even sure if the driver already check for pxe_enabled when passing the list of macs to nova | 12:40 |
| lucasagomes | that's a problem we have | 12:40 |
| * lucasagomes checks nova driver | 12:40 | |
| *** jcoufal has joined #openstack-ironic | 12:40 | |
| moshele | sambetts: ok I will look on that direction | 12:40 |
| sambetts | lucasagomes: it looks like we'll require this DHCP option even for ports with PXE disabled to provide networking for local boot tenant networks | 12:44 |
| *** catintheroof has joined #openstack-ironic | 12:46 | |
| *** siva_krish has joined #openstack-ironic | 12:47 | |
| *** baoli has joined #openstack-ironic | 12:47 | |
| *** Goneri has joined #openstack-ironic | 12:48 | |
| *** skramaja is now known as skramaja_afk | 12:52 | |
| *** aparnav has quit IRC | 12:52 | |
| *** abcxyz has quit IRC | 12:54 | |
| *** hoangcx has joined #openstack-ironic | 12:57 | |
| *** skramaja_afk is now known as skramaja | 12:57 | |
| openstackgerrit | Vadim Hmyrov proposed openstack/ironic: Block some port updates https://review.openstack.org/347360 | 13:01 |
| *** rloo has joined #openstack-ironic | 13:03 | |
| *** baoli has quit IRC | 13:03 | |
| *** wajdi has quit IRC | 13:04 | |
| *** hshiina has quit IRC | 13:06 | |
| openstackgerrit | xiexs proposed openstack/virtualbmc: Adds power_soft support to vbmc https://review.openstack.org/347367 | 13:16 |
| openstackgerrit | xiexs proposed openstack/virtualbmc: Adds power diag support to vbmc https://review.openstack.org/347368 | 13:16 |
| *** vmud213 has quit IRC | 13:21 | |
| xavierr | good morning Ironic | 13:21 |
| openstackgerrit | lokesh s proposed openstack/ironic: Add support for the audit middleware https://review.openstack.org/272658 | 13:24 |
| *** bks has quit IRC | 13:24 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add baremetal port set command to OSC plugin https://review.openstack.org/347287 | 13:26 |
| vdrok | morning xavierr and NobodyCam :) | 13:31 |
| openstackgerrit | lokesh s proposed openstack/ironic: Add support for the audit middleware https://review.openstack.org/272658 | 13:31 |
| rloo | morning xavierr, vdrok | 13:35 |
| vdrok | good morning rloo ! | 13:36 |
| rloo | hi vdrok, question about https://review.openstack.org/#/c/297895/9/nova/virt/ironic/driver.py | 13:36 |
| vdrok | yup | 13:37 |
| rloo | vdrok: if node.network_interface is None, can't it be neutron if CONF.default_network_interface is neutron? | 13:37 |
| jroll | morning everyone | 13:37 |
| rloo | morning jroll | 13:37 |
| mgould | morning jroll | 13:37 |
| vdrok | rloo: oh right there is a config option | 13:37 |
| mgould | morning rloo xavierr | 13:37 |
| vdrok | i completely forgot about it :( | 13:37 |
| rloo | vdrok: yeah, that wonderful config option... | 13:37 |
| jroll | rloo: I really wish we returned what it's actually using in the API | 13:37 |
| vdrok | morning jroll | 13:38 |
| jroll | otherwise how do ops know which network interface a node is using? | 13:38 |
| rloo | jroll: i agree with you but i have already forgotten the logic that got us to here. | 13:38 |
| rloo | vdrok: do you remember? | 13:38 |
| sambetts | jroll, rloo, vdrok: this is why I prefered actaully having the field set on node create if unspecified instead of just making it an if None then fall back to default | 13:38 |
| vdrok | rloo: dtantsur|pto's spec | 13:39 |
| vdrok | saying that None == use default | 13:39 |
| jroll | we have to fix this, else there's no way for nova to know :/ | 13:39 |
| jroll | either that or we always bind late and in the flat/noop case have ironic do the port-update before pxe booting | 13:40 |
| jroll | ... which it already does, we could just add the host_id | 13:40 |
| jroll | hrm | 13:40 |
| rloo | there seem to be two 'settings'. the node's 'x_interface', and the actual node.driver's interface. | 13:41 |
| rloo | it would be two calls (or is there a call) to get the node's driver, then the driver's interfaces? | 13:42 |
| jroll | there will be a call to get the list of possible (and default) interfaces for a node, but it doesn't exist yet | 13:43 |
| rloo | jroll: we don't want the list of possible/default, we want the actual. | 13:43 |
| jroll | rloo: right, we could combine the two, though | 13:43 |
| jroll | rloo: first call gets {driver: foo, network_interface: None}, second finds the default network interface for driver:foo | 13:44 |
| sambetts | I definatly think if you do a node-show you should get back a node with the interfaces currently being used by that node | 13:44 |
| sambetts | regardless if in the db they are None or not | 13:44 |
| *** links has quit IRC | 13:44 | |
| jroll | not optimal at all, I'd rather return in the GET /nodes/uuid | 13:44 |
| jroll | yeah | 13:44 |
| rloo | how long is dtantsur|pto on pto? | 13:45 |
| jroll | all week | 13:45 |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Add tests for testing node-delete command https://review.openstack.org/251458 | 13:45 |
| jroll | I think he's back monday | 13:45 |
| rloo | can this wait til next week? | 13:45 |
| vdrok | if it will be shown only in API, we'll have to have some magic to ensure that it's not accidentally saved to DB | 13:45 |
| jroll | I guess? | 13:45 |
| sambetts | IMO I'd prefer it if we actually set the unspecficied interfaces on the node at node-create time, instead of saving them as None in the DB | 13:46 |
| rloo | jroll: if we wait til next week, you can discuss with dtantsur|pto, and i'll be away so i won't have to think about it :D | 13:46 |
| jroll | rloo: your opinion is valuable though :( | 13:47 |
| rloo | sambetts: if we set the interface at node.create time, then that means the global config won't work. | 13:47 |
| rloo | jroll: heh. | 13:47 |
| sambetts | rloo: no it does work, once, at node-create time | 13:47 |
| jroll | well, the global config would help determine what's set at... that | 13:47 |
| rloo | sambetts: that's what i mean. only once. which isn't the way global configs work, is it? i'd need to think about that | 13:48 |
| *** ayoung has joined #openstack-ironic | 13:48 | |
| *** wajdi has joined #openstack-ironic | 13:48 | |
| jroll | depends on the config help text :) | 13:48 |
| sambetts | ^ | 13:49 |
| sambetts | hehe | 13:49 |
| jroll | "default to set node.network_interface to at node-create time" vs "default network_interface if node.network_interface is NULL" | 13:49 |
| jroll | that said, it'll be an inconvenience for upgrades, but meh? | 13:49 |
| sambetts | I think it'll make upgrades more consistant though | 13:49 |
| sambetts | because if Ironic changes the default for the config, it won't change the nodes | 13:50 |
| jroll | yeah, I meant the first upgrade | 13:50 |
| rloo | so right now, lets set config to 'flat'. create node.n_iface=None. node.driver.n_iface='flat'. if i then set config to 'noop' and restart conductor? will node.driver.n_iface='noop'? | 13:51 |
| jroll | in sam's proposal? no, it will still be flat | 13:51 |
| rloo | jroll: no with current code/dmitry's composition. | 13:51 |
| jroll | rloo: currently, that's correct | 13:52 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add baremetal port unset command to OSC plugin https://review.openstack.org/347402 | 13:54 |
| rloo | i need to look at dmitry's spec again. sigh. | 13:54 |
| openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Replace DB API call to object's method in iLO drivers https://review.openstack.org/347403 | 13:54 |
| *** vnogin has joined #openstack-ironic | 13:54 | |
| *** jcoufal has quit IRC | 13:54 | |
| TheJulia | vdrok: I think accidental saving to the API becomes an implicit operator decision to default the node to that behavior. It is not like an operator can bulk change everything to tenant isolated networking anyway, they will have to explicitly choose their path forward | 13:56 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Add node.resource_class field https://review.openstack.org/345055 | 13:56 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-specs: Boot from Volume - Reference Drivers https://review.openstack.org/294995 | 13:56 |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Add tests for testing node-delete command https://review.openstack.org/251458 | 13:57 |
| vdrok | TheJulia: with current code they still can switch between flat/noop easily, just updating config | 13:57 |
| TheJulia | vdrok: yup, and those are similar configurations, but the configuration information and possibly even the pre-existing infrastucture configuraiton has to be radically changed in order to just magically switch to tenat isolation | 13:58 |
| TheJulia | i.e. they would have to have their switches explicitly already configured and integrated in prior | 13:58 |
| vdrok | yeah I agree with you about isolation case | 13:58 |
| TheJulia | the other two cases, dumb switch, even a hub would work | 13:59 |
| TheJulia | cross over cable :) | 13:59 |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Add tests for testing node-delete command https://review.openstack.org/251458 | 13:59 |
| TheJulia | which reminds me, I keep putting off getting a thunderbolt to ethernet adapter for my laptop so I can play with the nuc that I have | 13:59 |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Add tests for testing node-delete command https://review.openstack.org/251458 | 14:00 |
| rloo | on another note, has anyone noticed that the gate for ipa is flaky? | 14:01 |
| rloo | something to do with the py27 or py34 tests. | 14:02 |
| *** siva_krish has quit IRC | 14:02 | |
| *** siva_krish has joined #openstack-ironic | 14:04 | |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Add tests for testing node-delete command https://review.openstack.org/251458 | 14:05 |
| *** mgoddard_ has joined #openstack-ironic | 14:05 | |
| *** jcoufal has joined #openstack-ironic | 14:06 | |
| jroll | rloo: I haven't, got a link? | 14:06 |
| rloo | jroll: https://review.openstack.org/#/c/346963/ | 14:07 |
| *** PollyZ has joined #openstack-ironic | 14:07 | |
| jroll | rloo: thanks | 14:07 |
| rloo | jroll: I noticed something failing last Friday (well, sometime before Monday) and didn't get a chance yesterday to look. And I think I've seen more failures from yesterday (prob cuz I reviewed a bunch of patches yest) | 14:08 |
| *** mgoddard has quit IRC | 14:08 | |
| rloo | lucasagomes: did you look into these failures on your ipa patch: https://review.openstack.org/#/c/344164/ | 14:09 |
| lucasagomes | rloo, hi there, checking | 14:09 |
| lucasagomes | rloo, yeah I've rechecked cause it looks unrelated | 14:09 |
| lucasagomes | http://logs.openstack.org/64/344164/6/check/gate-ironic-python-agent-python34/b981071/console.html#_2016-07-25_23_21_00_291033 | 14:09 |
| jroll | that's... why are we checking if _check_for_iscsi is called | 14:10 |
| rloo | lucasagomes: i'm seeing failures in other ipa patches too ^^ jroll is looking | 14:10 |
| lucasagomes | lemme take a more deep look into it | 14:10 |
| jroll | feels like it shouldn't be | 14:10 |
| jroll | lucasagomes: that's part of evaluate_hardware_support, which shouldn't be called in that method best I can tell | 14:10 |
| lucasagomes | rloo, jroll oh alright, yeah it seems that we have some random failures/passes in the unittests there | 14:10 |
| lucasagomes | jroll, lemme try to reproduce it locally | 14:11 |
| * lucasagomes sets a loop | 14:11 | |
| *** siva_krish has quit IRC | 14:11 | |
| lucasagomes | jroll, it seems that it's mocking some nested methods, for unittests is better to just mock the upper level method/function directly | 14:12 |
| *** thiagop has joined #openstack-ironic | 14:12 | |
| rloo | jroll, lucasagomes: here's one where py27 fails. randomness would explain... https://review.openstack.org/#/c/345513/ | 14:12 |
| thiagop | Good morning, Ironichood | 14:12 |
| jroll | lucasagomes: in dispatch_to_managers, _get_managers() is called, I bet that's cached | 14:13 |
| jroll | er, I think that's cached | 14:13 |
| jroll | which would explain things | 14:13 |
| lucasagomes | rloo, yup, yeah probably it's random... probably it's a module variable controller when that call is invoked and when is not | 14:13 |
| lucasagomes | and parallel tests are messing up | 14:13 |
| lucasagomes | jroll, yeah something like that | 14:13 |
| jroll | yeah, it is cached | 14:13 |
| *** sdake has joined #openstack-ironic | 14:14 | |
| *** joprovost has joined #openstack-ironic | 14:14 | |
| jroll | lucasagomes: so we just shouldn't be asserting that checK_for_iscsi is called, unless we're testing for it | 14:14 |
| lucasagomes | ++ | 14:14 |
| jroll | and mock it at the class level | 14:14 |
| * jroll does it | 14:14 | |
| lucasagomes | jroll, thanks, we can fast merge that once you get the patch up | 14:15 |
| jroll | cool | 14:15 |
| lucasagomes | cause random failures are a PITA | 14:15 |
| rloo | lucasagomes: +++++ | 14:15 |
| jroll | thanks for catching that rloo | 14:17 |
| * jroll running tests | 14:17 | |
| rloo | thank YOU jroll & lucasagomes for fixing. I should have mentioned it yesterday. I think I got as far as trying to reproduce and got distracted. | 14:17 |
| *** sdake_ has joined #openstack-ironic | 14:18 | |
| lucasagomes | rloo, thank you and jroll (he's the one actually fixing it :D) | 14:18 |
| *** baoli has joined #openstack-ironic | 14:18 | |
| * lucasagomes is trying to reproduce locally tho | 14:18 | |
| rloo | lucasagomes: team effort :D | 14:18 |
| *** baoli_ has joined #openstack-ironic | 14:20 | |
| *** sdake has quit IRC | 14:20 | |
| *** PollyZ has quit IRC | 14:21 | |
| *** mtanino has joined #openstack-ironic | 14:22 | |
| *** baoli has quit IRC | 14:23 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-python-agent: Fix races in advertise_address unit tests https://review.openstack.org/347421 | 14:25 |
| jroll | lucasagomes: rloo: that should do it | 14:26 |
| rloo | jroll: thx. | 14:26 |
| jroll | np | 14:26 |
| apughazh | good morning | 14:26 |
| *** e0ne has quit IRC | 14:27 | |
| apughazh | I uploaded a patch | 14:27 |
| apughazh | https://review.openstack.org/#/c/346894/ | 14:27 |
| * jroll is happy there's no rechecks on the patch that added that bug | 14:27 | |
| jroll | morning apughazh | 14:27 |
| *** michauds has joined #openstack-ironic | 14:28 | |
| *** gabriel-bezerra has quit IRC | 14:33 | |
| *** siva_krish has joined #openstack-ironic | 14:34 | |
| *** harshs has joined #openstack-ironic | 14:35 | |
| *** yuikotakadamori has joined #openstack-ironic | 14:35 | |
| *** michauds_ has joined #openstack-ironic | 14:39 | |
| *** hwcomcn has quit IRC | 14:40 | |
| *** zhenguo has quit IRC | 14:41 | |
| lucasagomes | jroll, thanks! | 14:42 |
| *** harshs has quit IRC | 14:42 | |
| lucasagomes | will approve as soon as the tests are returned by jenkins | 14:42 |
| *** fumie has quit IRC | 14:42 | |
| TheJulia | vdrok: Hey, out of curiosity, do you have a pending revision or a local patch ready for https://git.openstack.org/cgit/openstack/ironic/tree/ironic/tests/unit/api/utils.py#n97 ? | 14:43 |
| openstackgerrit | Akilan Pughazhendi proposed openstack/ironic: Updating dev quickstart to include compatiblity for newest distros https://review.openstack.org/346894 | 14:44 |
| vdrok | TheJulia: huh, no, I don't | 14:45 |
| vdrok | but can do it fast :) | 14:45 |
| *** PollyZ has joined #openstack-ironic | 14:45 | |
| TheJulia | cool, thank you :) | 14:47 |
| *** harshs has joined #openstack-ironic | 14:48 | |
| *** michauds_ has quit IRC | 14:54 | |
| vdrok | TheJulia: otoh, I don't know if the popping should be removed | 14:54 |
| vdrok | tests that are using an old api version start to fail so need to be changed too | 14:55 |
| vdrok | it seems like it's the first POSTable attribute we added since the introduction of microversioning | 14:55 |
| *** spandhe has joined #openstack-ironic | 14:56 | |
| *** spandhe has quit IRC | 14:56 | |
| jroll | vdrok: yeah, I ran into the same thing adding resource_class | 14:56 |
| vdrok | jroll: so maybe just include network_interface if it was requested in arguments to post_get_test_node? | 14:57 |
| jroll | vdrok: like you have it now? | 14:57 |
| vdrok | s/if/only if/ | 14:57 |
| vdrok | yeah | 14:57 |
| jroll | yeah, that worked for me | 14:57 |
| vdrok | just change TODO -> NOTE :D | 14:57 |
| jroll | https://review.openstack.org/#/c/345055/8/ironic/tests/unit/api/utils.py | 14:58 |
| jroll | heh, yeah | 14:58 |
| *** harshs has quit IRC | 15:00 | |
| *** vnogin has quit IRC | 15:00 | |
| *** mjura has quit IRC | 15:01 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Change comment regarding network_interface https://review.openstack.org/347441 | 15:04 |
| *** rcernin has quit IRC | 15:05 | |
| *** mbound has quit IRC | 15:07 | |
| thiagop | Folks, vote in us: https://www.openstack.org/summit/barcelona-2016/vote-for-speakers/presentation/16078/?q=red%20pill | 15:07 |
| thiagop | sambetts rajinir mjturek1 ^^ | 15:08 |
| rajinir | thiagop:+2 | 15:08 |
| mjturek1 | thiagop: will do | 15:08 |
| thiagop | :D | 15:08 |
| vdrok | :) | 15:08 |
| yuikotakadamori | jroll, hi, all of Ironic console patches have been merged, then we've got some comments from Nova guys to Nova patch. They say that we can FFE request, then could you please talk to Nova PTL? | 15:09 |
| jroll | yuikotakadamori: it's on my todo list, I'm hoping to get the networks patch ready and talk to PTL about both at the same time | 15:10 |
| yuikotakadamori | jroll, oh, ya, thanks! | 15:11 |
| rajinir | thiagop: I noticed the abstract is blank. Any idea? | 15:11 |
| thiagop | rajinir: nope... | 15:11 |
| jroll | rajinir: thiagop: the voting site doesn't let you link directly to a presentation anymore | 15:12 |
| jroll | because you aren't supposed to ask people to vote for you | 15:12 |
| jroll | so... please don't ask people to vote for you | 15:12 |
| rajinir | jroll: wonder why we cannot see the abstracts? | 15:12 |
| jroll | rajinir: I'm not sure, I assumed the random talk that came up when you clicked the link doesn't have one | 15:13 |
| thiagop | jroll: why? | 15:13 |
| jroll | but you're right, I don't see any | 15:13 |
| jroll | thiagop: because people are supposed to vote based on what they want to see, not what their friends ask them to vote for | 15:13 |
| jroll | otherwise it's a popularity contest | 15:13 |
| jroll | which isn't healthy for the talk sessions | 15:13 |
| *** harshs has joined #openstack-ironic | 15:14 | |
| *** vnogin has joined #openstack-ironic | 15:14 | |
| thiagop | jroll: one should vote in what is interesting for him. Not only by the request of a friend. This way I can't even point a presentation I found interesting to someone | 15:14 |
| jroll | thiagop: right, it's a tradeoff | 15:15 |
| *** rbartal has quit IRC | 15:15 | |
| thiagop | jroll: I found it interesting that OSF thinks that a poll of such smart people would vote only based on friendship... | 15:15 |
| jroll | thiagop: but there's been years of abuse of people asking their entire company to vote for talks just because they're from the same company | 15:15 |
| thiagop | a pool* | 15:15 |
| jroll | thiagop: imagine CEO sends a list of talks and asks for votes | 15:15 |
| jroll | keep in mind you don't need to be ATC or even attending summit to vote | 15:15 |
| jroll | so suddenyl every $large_corp talk has 20k votes | 15:16 |
| thiagop | jroll: but large corps have their own sponsored space at summits, right? | 15:16 |
| TheJulia | vdrok: perhaps some test refactoring then :( | 15:17 |
| jroll | it does bum me out that I can't link interesting talks to friends, but I guess the OSF thinks it's worth it | 15:17 |
| jroll | thiagop: that's irrelevant, I think? | 15:17 |
| openstackgerrit | Akilan Pughazhendi proposed openstack/ironic: Updating dev quickstart to include compatiblity for newest distros https://review.openstack.org/346894 | 15:18 |
| thiagop | jroll: I don't think so. If the large corp has it's sponsored space, why submit coporate talks and ask for such an amount of votes for it to be selected? It makes no sense to me | 15:19 |
| thiagop | unless the politics of summit is far worse than I expect... | 15:19 |
| thiagop | jroll: nonetheless, sorry for asking votes anyway. It won't happen again. | 15:20 |
| jroll | thiagop: 1) they aren't always corporate talks, some may be actually interesting talks by devs; 2) because corporate space AND a bunch of talks == more marketing than just corporate space | 15:20 |
| thiagop | ^that corroborates my theory | 15:21 |
| jroll | thiagop: the politics of the summit are a large amount of the reason the summit is splitting :) | 15:21 |
| *** Nisha_away has joined #openstack-ironic | 15:22 | |
| JayF | For your consideration, I do have a talk on agents + hardware managers, that I know some folks in here would be interested in, if you search for "007" you can see it. But it seems like abstracts aren't showing up so you probably can't make a good decision on it yet :) | 15:22 |
| JayF | jroll: I try to avoid the friend-upboats by not being friendly | 15:22 |
| jroll | JayF: "vote for my talk you jerks" | 15:23 |
| jroll | :P | 15:23 |
| * thiagop tries to not reach the foundation channel to ask about abstracts | 15:23 | |
| JayF | thiagop: I already emailed speakersupport@ about it | 15:23 |
| thiagop | last time I showed them a couple of things wrong at openstack.org | 15:24 |
| thiagop | (submission dates and another thing I don't remember now) | 15:24 |
| * rloo prefers that there aren't any talks at the summit. | 15:32 | |
| *** harshs has quit IRC | 15:32 | |
| thiagop | rloo: look at the bright side: free time | 15:33 |
| thiagop | :P | 15:33 |
| rloo | thiagop: free time for what? :D | 15:33 |
| thiagop | rloo: wander thru the city or meet with other OS folks that doesn't like talks either to chat/work | 15:34 |
| rloo | thiagop: the latter should == design sessions + ;) | 15:35 |
| thiagop | rloo: I thought more of a "partial consensus" | 15:36 |
| thiagop | :) | 15:36 |
| *** harshs has joined #openstack-ironic | 15:36 | |
| *** jaybeale has joined #openstack-ironic | 15:37 | |
| rajinir | thiagop: jroll: The abstracts are showing up now | 15:37 |
| jroll | woot | 15:37 |
| jroll | JayF: ^ | 15:37 |
| thiagop | rajinir: w00t | 15:37 |
| thiagop | jroll: about removing tests of tempest, I bumbed into the "no facility to check if plugin is installed problem" | 15:38 |
| thiagop | jroll: will look if they have something in course to solve this later today | 15:39 |
| jroll | thiagop: they have a concept of feature flags, we may have to create some | 15:39 |
| *** harshs has quit IRC | 15:39 | |
| thiagop | jroll: using that @service tag? | 15:40 |
| thiagop | I thought it was meant only for scenario tests... | 15:40 |
| jroll | thiagop: similar to live_migrate_enabled or whatever, I think | 15:40 |
| jroll | I'm not sure | 15:40 |
| thiagop | jroll: will take a look on it | 15:40 |
| *** ifarkas is now known as ifarkas_afk | 15:40 | |
| jroll | thiagop: thanks, mtreinish is a good person to talk to if you need suggestions | 15:41 |
| thiagop | jroll: ack | 15:41 |
| *** tesseract- has quit IRC | 15:43 | |
| mtreinish | thiagop: don't use the service decorator in a plugin, it's not part of the stable interface | 15:44 |
| mtreinish | I've seen several projects try to use that and complain when we removed something from tree | 15:44 |
| thiagop | mtreinish: Hi! Not on the plugin, what I'm trying to do is remove the old baremetal tests and keep the ones that has a baremetal flag running, but there is a lot of checking around the config file | 15:45 |
| thiagop | mtreinish: baremetal flag *in other services* | 15:45 |
| mtreinish | right, and that will likely have to be removed. When we remove the ironic tests from tempest all that goes away. It was a hacky workaround that got in because ironic was a service, it wouldn't have been allowed for any other nova driver | 15:46 |
| mtreinish | the best path forward would be to leverage specific feature flags on what ironic can and can't support to make sure everything ironic can't do is skipped | 15:47 |
| mtreinish | but the easiest path (while not the best) is a whitelist file (with os-testr or tempest run) where you maintain that list of what to run somewhere | 15:48 |
| thiagop | mtreinish: you mean removing tests such as https://github.com/openstack/tempest/blob/master/tempest/api/compute/admin/test_baremetal_nodes.py ? | 15:49 |
| thiagop | mtreinish: this one is inside "compute" | 15:49 |
| mtreinish | that test will stay, (for the time being at least) because it tests a nova api | 15:49 |
| *** PollyZ has quit IRC | 15:49 | |
| mtreinish | but the service available skip will have to be removed | 15:50 |
| mtreinish | because that flag will go away | 15:50 |
| *** rama_y has joined #openstack-ironic | 15:51 | |
| mtreinish | if there are missing feature flags on the nova side we can add them | 15:51 |
| *** PollyZ has joined #openstack-ironic | 15:55 | |
| thiagop | mtreinish: sorry, I'm not very tempest fluent. What do you mean by "feature flag"? the @test.attr? | 15:55 |
| openstackgerrit | Merged openstack/ironic: Remove unused code when failing to start console https://review.openstack.org/346524 | 15:56 |
| mtreinish | thiagop: a config option. Basically things in groups like: https://github.com/openstack/tempest/blob/master/tempest/config.py#L294-L395 | 15:57 |
| thiagop | mtreinish: uhmm | 15:57 |
| mtreinish | most services have a config group like that, it used to tell tempest what configurable or optional features are available on the deployment you're running against | 15:58 |
| openstackgerrit | Akilan Pughazhendi proposed openstack/ironic: Updating dev quickstart to include compatiblity for newest distros https://review.openstack.org/346894 | 15:58 |
| thiagop | mtreinish: my idea was to detect somehow if the ironic plugin is active and if so, run these tests | 15:59 |
| *** trown is now known as trown|lunch | 15:59 | |
| thiagop | mtreinish: but I saw a doc yesterday saying that doing service discovery on tempest is bad practice.. | 15:59 |
| *** harshs has joined #openstack-ironic | 16:00 | |
| mtreinish | thiagop: right, that would be the wrong way to handle this | 16:01 |
| mtreinish | it's also harder than you think because running tests is decoupled from the plugin mechanism (and the test code itself) | 16:01 |
| mtreinish | we need to be explicit about what we're running | 16:01 |
| mtreinish | which is why the current usage of that ironic service available flag is bad, and was just a fast hack to fix things and something we should have pushed back on | 16:02 |
| mtreinish | but hindsight is 20/20 | 16:02 |
| jroll | so to be clear, we'd rather a feature flag like os_baremetal_extension_available ? | 16:02 |
| jroll | (for that os-baremetal api extension specifically) | 16:02 |
| mtreinish | that already exists, it's the api_extensions list option in that group | 16:03 |
| mtreinish | just remove it from the list | 16:03 |
| jroll | got it | 16:03 |
| mtreinish | the problem is that's actually installed, but nova isn't configured so that it'll do something | 16:03 |
| mtreinish | so we probably need just a baremetal option in that group, to say that nova has baremetal anything configured | 16:04 |
| mtreinish | we might want a better name that, but we can bikeshed on review when it's ready | 16:04 |
| mtreinish | this is the standard discoverability problem that most projects' APIs have | 16:04 |
| jroll | so CONF.compute.baremetal or whatever | 16:05 |
| jroll | yeah, indeed | 16:05 |
| mtreinish | probably CONF.compute_feature_enabled.baremetal, but yeah | 16:05 |
| jroll | yeah | 16:05 |
| sambetts | can't we just look at the which compute driver is enabled/ | 16:05 |
| sambetts | ? | 16:05 |
| mtreinish | sambetts: no, that's not user exposed information | 16:05 |
| mtreinish | and even if it was tempest couldn't rely on an api response to determine whether to skip a test. What if there was a bug in the api that made it lie, you'd skip the test and never know | 16:06 |
| sambetts | oh I guess I misunderstand which conf files we're looking at then | 16:06 |
| mtreinish | sambetts: this is the tempest config file | 16:06 |
| *** mbound has joined #openstack-ironic | 16:08 | |
| sambetts | ah I see, perhaps we should put a config option to store the compute driver in tempest, so that we can switch on that, and then it leaves it open for other compute drivers to switch on/off stff if they don't support the full feature set | 16:08 |
| thiagop | mtreinish: but even if we check another config flag instead of the one we check today, there is the problem of how tempest will know how to instantiate the baremetal_client | 16:09 |
| mtreinish | sambetts: that's not something a user should need to know (or likely will in many cases) Tempest's job isn't to maintain a matrix of which features work with which virt drivers | 16:09 |
| mtreinish | it also couldn't keep up with that, because nova changes pretty quickly | 16:09 |
| jroll | yeah, I agree it should be individual feature flags | 16:09 |
| jroll | and a baremetal flag for the os-baremetal extension availability | 16:09 |
| sambetts | so there should things like be a CONF.compute.interface_attach_possible = True/False? | 16:10 |
| *** PollyZ has quit IRC | 16:10 | |
| mtreinish | sambetts: like: https://github.com/openstack/tempest/blob/master/tempest/config.py#L363 | 16:10 |
| sambetts | right | 16:11 |
| jroll | sambetts: yep | 16:11 |
| jroll | so what we need to do is run through the cruft that uses service_enabled('ironic') in tempest and fix it with real feature flags like that | 16:11 |
| mtreinish | jroll: ++ | 16:11 |
| mtreinish | thiagop: for nova tests, the config flag is fine. Those api clients aren't going anywhere and the tests just need a condition to not execute unless you configured nova to use baremetal | 16:12 |
| mtreinish | thiagop: for ironic tests those will be in the plugin and will be included when you install it | 16:12 |
| *** mbound has quit IRC | 16:13 | |
| thiagop | mtreinish jroll so, we won't remove the baremetal_client from tempest? What can happen when we supersed it with the plugin's client (it will supersed, right?)? | 16:14 |
| jroll | eh? we need to remove the baremetal_client from tempest, there's no reason for it to be there | 16:15 |
| jroll | tempest/api/compute/admin/test_baremetal_nodes.py will stay, but that uses the nova client | 16:15 |
| mtreinish | thiagop: clients are per api endpoint. There is a nova baremetal extension, that client stays in tempest. Things that talk directly to ironic migrate with the plugin | 16:15 |
| thiagop | jroll: see L32 | 16:15 |
| jroll | oh, because it creates the nodes in ironic | 16:16 |
| jroll | ew | 16:16 |
| mtreinish | oh, yuck | 16:16 |
| sambetts | oh it shouldn't be doing that... | 16:16 |
| *** skramaja has quit IRC | 16:16 | |
| sambetts | devstack should have done that | 16:16 |
| jroll | tempest isn't always run with devstack :) | 16:16 |
| mtreinish | yeah lets either fix the test to be only nova api calls, move it to the ironic plugin, and/or delete it | 16:17 |
| thiagop | ¯\_(ツ)_/¯ | 16:17 |
| jroll | is it safe to assume ironic will always have nodes? I'm not sure we assume that in our plugin | 16:17 |
| jroll | fwiw that endpoint is deprecated | 16:18 |
| *** PollyZ has joined #openstack-ironic | 16:18 | |
| thiagop | jroll: I don't think so, but unsure | 16:18 |
| jroll | mtreinish: I assume the rule about deleting endpoints is to keep the tempest tests until you're ready to delete? | 16:18 |
| thiagop | thiagop: what I know is that some of the tests here on OneView CI aren't running exactly on HW... just the baremetal basic ops ones. Tempest really creates some nodes on the run | 16:19 |
| thiagop | jroll: ^ | 16:19 |
| jroll | right | 16:19 |
| mtreinish | jroll: well normally we keep it around until there aren't any stable branches using it, we just skip it on master so the delete patch can land | 16:19 |
| mtreinish | but api deletes are pretty rare | 16:19 |
| sambetts | thiagop: thats the Ironic API only tests right? | 16:20 |
| sambetts | those create nodes but with fake information | 16:20 |
| jroll | mtreinish: ah, right. well, the nova baremetal endpoint is deprecated and going away next cycle afaik | 16:20 |
| thiagop | sambetts: I think these nova tests enters this category too... | 16:20 |
| thiagop | sambetts: lemme check | 16:21 |
| mtreinish | jroll: yeah, iir the proxy api discussion last week that was the conclusion | 16:21 |
| jroll | indeed | 16:21 |
| jroll | I was wondering if we could just drop that test, but I guess not | 16:21 |
| mtreinish | so for this case I'd say we probably could since it's a bad test and doesn't provide much value | 16:21 |
| mtreinish | for the the other proxy apis it's probably not so straightforward. Especially when you factor in that some are in the defcore list :( | 16:22 |
| jroll | yeah | 16:22 |
| jroll | I'm fine with dropping it | 16:22 |
| jroll | nova-baremetal has been gone for years | 16:22 |
| jroll | that proxy was there to keep the api working | 16:22 |
| thiagop | a humble +1 for dropping | 16:22 |
| *** Romanenko_K has quit IRC | 16:23 | |
| *** Sukhdev has joined #openstack-ironic | 16:23 | |
| mtreinish | so technically there is: http://docs.openstack.org/developer/tempest/test-removal.html but pretend I didn't share that link :) | 16:24 |
| mtreinish | just push the patch and lets hope no one remembers that doc | 16:24 |
| thiagop | mtreinish: lol | 16:24 |
| jroll | heh | 16:24 |
| mtreinish | otherwise it'll be at least week for public comment before we can remove it | 16:24 |
| jroll | I think it's worth it given the complexity it'd require to keep it | 16:24 |
| thiagop | jroll mtreinish I'm reading the last sentence and smiling | 16:26 |
| thiagop | anything that lives in tempest which doesn't test one of these projects can be removed [...] | 16:26 |
| jroll | well, it's testing nova | 16:27 |
| mtreinish | thiagop: right, but this is technically a nova test :) | 16:27 |
| mtreinish | but we can use that to justify fast tracking a removal | 16:28 |
| mtreinish | it's only partially a lie | 16:28 |
| thiagop | alright | 16:29 |
| *** derekh has quit IRC | 16:30 | |
| *** aNupoisc has joined #openstack-ironic | 16:31 | |
| openstackgerrit | Akilan Pughazhendi proposed openstack/ironic: Updating dev quickstart to include compatiblity for newest distros https://review.openstack.org/346894 | 16:34 |
| *** jpich has quit IRC | 16:35 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Updating dev quickstart to include compatiblity for newest distros https://review.openstack.org/346894 | 16:39 |
| lucasagomes | rloo, jenkins didn't return yet for #347421, mind keeping an eye on it and approve once it does? I will call it a day | 16:43 |
| *** yuikotakadamori has quit IRC | 16:43 | |
| rloo | lucasagomes: will do! have a good evening! | 16:43 |
| lucasagomes | thanks | 16:43 |
| lucasagomes | have a great ugt evening all, see you tomorrow | 16:44 |
| *** lucasagomes is now known as lucas-dinner | 16:44 | |
| vdrok | might lucas-dinner | 16:46 |
| vdrok | night :) | 16:46 |
| *** gfhellma has joined #openstack-ironic | 16:47 | |
| NobodyCam | good Morning Ironicers | 16:47 |
| *** Sukhdev has quit IRC | 16:48 | |
| *** baoli_ has quit IRC | 16:49 | |
| *** ccamacho is now known as ccamacho|awawawa | 16:50 | |
| *** ccamacho|awawawa is now known as ccamacho|away | 16:50 | |
| mgould | morning NobodyCam, also jroll rloo mtreinish thiagop | 16:50 |
| NobodyCam | morning mgould | 16:50 |
| *** bvandewa has joined #openstack-ironic | 16:51 | |
| * mgould reads the subteam status report and learns the word "xenodochial" | 16:52 | |
| mgould | can't imagine how I've gone through life not knowing that :-) | 16:52 |
| *** bvandewa has quit IRC | 16:52 | |
| *** fragatina has quit IRC | 16:54 | |
| rloo | hi mgould. you're probably the only person that reads the status report :) | 16:54 |
| rloo | mgould: i wasn't quite sure i used that word properly ;) | 16:55 |
| NobodyCam | morning rloo :) | 16:55 |
| *** harshs has quit IRC | 16:55 | |
| rloo | morning NobodyCam | 16:55 |
| NobodyCam | :) | 16:55 |
| *** harshs has joined #openstack-ironic | 16:56 | |
| mgould | rloo: it's certainly grammatically correct | 16:57 |
| *** gfhellma1 has joined #openstack-ironic | 16:57 | |
| rloo | mgould: :D | 16:57 |
| * mgould had assumed you were using a script and a big list of positive adjectives :-) | 16:57 | |
| *** gfhellma has quit IRC | 16:59 | |
| rloo | mgould: the fun part was picking an adjective :) | 17:00 |
| * sambetts looks forward to see what word rloo will use each week ;) | 17:00 | |
| *** fragatina has joined #openstack-ironic | 17:00 | |
| rloo | mgould: i only plan to do this once per letter | 17:00 |
| thiagop | morning mgould | 17:01 |
| thiagop | morning NobodyCam | 17:02 |
| *** bvandewa has joined #openstack-ironic | 17:06 | |
| *** ChubYann has joined #openstack-ironic | 17:06 | |
| NobodyCam | Morning thiagop ;) | 17:06 |
| *** fragatina has quit IRC | 17:07 | |
| *** jtomasek has quit IRC | 17:10 | |
| *** PollyZ has quit IRC | 17:12 | |
| *** trown|lunch is now known as trown | 17:14 | |
| *** gfhellma1 has quit IRC | 17:15 | |
| *** bvandewa has quit IRC | 17:17 | |
| *** harshs has quit IRC | 17:21 | |
| * mgould -> home; good night! | 17:26 | |
| openstackgerrit | Akilan Pughazhendi proposed openstack/ironic: Updating dev quickstart to include compatiblity for newest distros https://review.openstack.org/346894 | 17:26 |
| *** mgould is now known as mgould|afk | 17:27 | |
| *** milan has quit IRC | 17:28 | |
| thiagop | good night mgould|afk | 17:29 |
| sambetts | o/ mgould|afk | 17:29 |
| *** sdake_ is now known as sdake | 17:37 | |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Add volume_connector table to save connector information https://review.openstack.org/200983 | 17:38 |
| *** mgoddard_ has quit IRC | 17:38 | |
| *** gfhellma has joined #openstack-ironic | 17:41 | |
| *** livelace has joined #openstack-ironic | 17:41 | |
| *** bvandewa_ has joined #openstack-ironic | 17:42 | |
| *** Sukhdev has joined #openstack-ironic | 17:42 | |
| *** aNupoisc has quit IRC | 17:42 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Fix races in advertise_address unit tests https://review.openstack.org/347421 | 17:43 |
| *** spandhe has joined #openstack-ironic | 17:44 | |
| *** spandhe has quit IRC | 17:44 | |
| *** rpioso has joined #openstack-ironic | 17:44 | |
| *** aNupoisc has joined #openstack-ironic | 17:44 | |
| JayF | jroll: would something like a "change-ramdisk.sh" in devstack/tools/ironic/scripts/ be appropriate? the idea would be it would take arguments for new kernel+initrd for ramdisk, and upload it to glance and make it the deploy_kernel and ramdisk for it | 17:45 |
| *** electrofelix has quit IRC | 17:46 | |
| *** spandhe has joined #openstack-ironic | 17:46 | |
| *** baoli has joined #openstack-ironic | 17:46 | |
| *** harshs has joined #openstack-ironic | 17:50 | |
| *** Sukhdev has quit IRC | 17:50 | |
| *** Sukhdev has joined #openstack-ironic | 17:51 | |
| * sambetts -> home | 17:56 | |
| sambetts | night all | 17:56 |
| *** sambetts is now known as sambetts|afk | 17:56 | |
| thiagop | good night sambetts|afk | 17:56 |
| *** priteau has joined #openstack-ironic | 17:57 | |
| *** gfhellma has quit IRC | 17:57 | |
| *** gfhellma has joined #openstack-ironic | 17:57 | |
| jroll | JayF: yeah, that seems sane to me | 17:58 |
| *** jaybeale has quit IRC | 18:00 | |
| *** harshs has quit IRC | 18:01 | |
| jroll | anyone up for v2 api meeting? should be short | 18:03 |
| jroll | -meeting-3 | 18:03 |
| *** gfhellma1 has joined #openstack-ironic | 18:03 | |
| *** gfhellma has quit IRC | 18:06 | |
| *** Nisha_away has quit IRC | 18:07 | |
| *** harshs has joined #openstack-ironic | 18:09 | |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Introduce VolumeConnector object https://review.openstack.org/214584 | 18:12 |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Introduce new RPCs to support volume connector operation https://review.openstack.org/214585 | 18:17 |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Add volume_targets table to save target information https://review.openstack.org/285218 | 18:21 |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Introduce VolumeTarget object https://review.openstack.org/285219 | 18:21 |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Introduce new RPCs to support volume target operations https://review.openstack.org/285220 | 18:22 |
| *** PollyZ has joined #openstack-ironic | 18:33 | |
| *** gfhellma1 has quit IRC | 18:35 | |
| *** sdake has quit IRC | 18:35 | |
| *** sdake has joined #openstack-ironic | 18:35 | |
| *** e0ne has joined #openstack-ironic | 18:36 | |
| *** moshele has quit IRC | 18:42 | |
| *** ahqa has joined #openstack-ironic | 18:42 | |
| openstackgerrit | Mark Hamzy proposed openstack/ironic: Support use_insecure_erase_with_wipefs option https://review.openstack.org/347515 | 18:43 |
| *** ahqa has left #openstack-ironic | 18:44 | |
| openstackgerrit | Mark Hamzy proposed openstack/ironic-python-agent: Support use_insecure_erase_with_wipefs option https://review.openstack.org/347518 | 18:46 |
| *** spandhe has quit IRC | 18:46 | |
| *** ahqa has joined #openstack-ironic | 18:47 | |
| jroll | rloo: so. resource_class. I still think we need it in newton so that people can prep for ocata if they want to use this stuff | 18:47 |
| jroll | otherwise they need to update their flavors and nodes at the same time | 18:47 |
| rloo | jroll: ok, if that's the reason. | 18:47 |
| devananda | jroll: random note - we should publish an update to the specs repo indicating what was completed in the 6.0 release (it's not mentioned here http://specs.openstack.org/openstack/ironic-specs/ ) | 18:47 |
| rloo | jroll: on the other hand, do we really want that avail before it is actually useable? | 18:48 |
| devananda | and then incorporate that into our release process | 18:48 |
| jroll | rloo: I think so | 18:48 |
| jroll | devananda: yeah, we should | 18:48 |
| * jroll makes a todo | 18:48 | |
| rloo | jroll: what if nova stuff doesn't happen (sorry, i know i should be more optimistic) | 18:48 |
| devananda | rloo: from discussions with nova folks last week, I agree - we need to get that into newton in ironic, even if we're not actually using the field yet internaly | 18:48 |
| *** Sukhdev has quit IRC | 18:49 | |
| devananda | rloo: then we bump the microversion and remove it? :) | 18:49 |
| jroll | lol | 18:49 |
| jroll | rloo: doesn't happen in newton or doesn't happen at all? | 18:49 |
| rloo | devananda: why do we need to get that into newton? it won't be useable before nova gets their part done | 18:49 |
| * devananda isnt sure whethre he's joking or not | 18:49 | |
| rloo | jroll: either | 18:49 |
| devananda | rloo: for rolling upgrades | 18:49 |
| rloo | devananda: ironic rolling upgrades which doesn't work yet? | 18:50 |
| devananda | rloo: so that operators can begin to populate that data in a fully-newton(ian system) | 18:50 |
| JayF | rloo: that don't /officially/ work yet. They do work in practice for a lot of folks. | 18:50 |
| JayF | rloo: plus I think we'd hope that it would be working by ocata | 18:50 |
| rloo | devananda: yet another reason not to do it in a newtonian world. | 18:51 |
| jroll | rloo: we need it in newton so that the resource tracker in nova can start using it during newton, and the data is there from t=0 on an ocata deploy of nova | 18:51 |
| devananda | let's (optimistically) assume nova lands their part in newton, and activates that feature in ocata | 18:51 |
| devananda | an operator will then be able to input that data into their newton-based cloud, and have it functional as soon as they upgrade to ocata | 18:51 |
| devananda | if they DONT do that, the when they upgrade to ocata, certain components of the new scheduler won't work | 18:52 |
| devananda | (that's the plan, anyway) | 18:52 |
| rloo | jroll: so the ironic-virt drive can be modified to set resource_class=Null if the node doesn't have a 'resource_class', right? | 18:52 |
| jroll | rloo: so that the ironic virt driver can be modified to pass the resource_class that is set up to the resource tracker, whether NULL or not | 18:53 |
| * devananda afk's briefly | 18:53 | |
| rloo | look, i'm not against having node.resource_class in newton, i just don't understand why it seems so urgent. let's say the new scheduler is in ocata and we add node.resource_class in ocata. does that break anything? no. things should work as they had worked before the new scheduler, right? | 18:54 |
| * jroll is trying to think through all the cases | 18:57 | |
| rloo | and don't forget, we/ironic can release frequently :) | 18:57 |
| jroll | right, but some deployers only use the stable version | 18:57 |
| JayF | The long pole here is the nova virt driver change, which we have agreement from them to get in for Newton | 18:57 |
| JayF | the problem isn't Ironic deploying quickly enough, it's getting that info exposed so the nova driver will be willing to read it | 18:57 |
| JayF | aiui | 18:58 |
| jroll | correct | 18:58 |
| rloo | JayF: i am fine modifying the nova virt driver. | 18:58 |
| jroll | but we've been told that if someone doesn't populate it, the old way will still work in ocata | 18:58 |
| JayF | rloo: they won't let us modify the virt driver to read a field that doesn't exist yet | 18:59 |
| rloo | JayF: that code could just look for 'resource_name' from the node, if it isn't there, set it to NULL | 18:59 |
| jroll | which is the catch | 18:59 |
| jroll | or rather, which is why "wait until ocata" might work | 18:59 |
| rloo | that's odd. they should allow that, so that new nova 'might' work with an older ironic. | 18:59 |
| jroll | work/be okay | 18:59 |
| *** spandhe has joined #openstack-ironic | 18:59 | |
| rloo | jroll: so the nova code that will be using resource_class won't be in newton, it'll be in ocata? | 19:00 |
| jroll | rloo: new nova won't work with old ironic because microversions | 19:00 |
| JayF | rloo: the objection would be more, we can't write a client to an api that doesn't exist. So we can't modify the virt driver to read in the field until it exists. | 19:00 |
| jroll | we pin to a certain microversion | 19:00 |
| jroll | rloo: the nova code that reads resource class will be in newton (if we have the support in ironic for it), the code that uses resource_class for scheduling will be in ocata | 19:01 |
| *** jaybeale has joined #openstack-ironic | 19:01 | |
| rloo | jroll: so we're going to add node.resource_class in version 1.24 (eg), and then pin nova to version 1.24 in nova-newton? | 19:01 |
| jroll | rloo: correct | 19:01 |
| jroll | rloo: so, backing up, the reason why it's urgent is because nova folks would like this in during newton to help further their scheduling objectives (which happen to help us a ton, it reduces a lot of problems we have due to slow resource tracker things) | 19:03 |
| rloo | jroll: so if we don't have node.resource_class in newton, what will the nova-side code look like wrt the resource class value for any bm nodes? | 19:03 |
| jroll | and they follow normal feature freeze, so we can't just land this at end of cycle | 19:03 |
| *** yuanying has quit IRC | 19:03 | |
| jroll | rloo: the nova side wouldn't use it at all, and it wouldn't put resource data for ironic nodes in the resource_providers table | 19:03 |
| rloo | sigh. seems like it would all work if they just changed the ironic-virt driver. | 19:04 |
| *** ahqa has quit IRC | 19:04 | |
| jroll | to do what? | 19:04 |
| jroll | I'm confused | 19:04 |
| rloo | well, to add a NULL resource_class | 19:04 |
| jroll | I mean... we can | 19:04 |
| *** aNupoisc has quit IRC | 19:04 | |
| jroll | but that doesn't help prep for scheduling things | 19:04 |
| rloo | look, whatever they do, should be backwards compatible for folks that don't care to set node.resource_class etc. | 19:05 |
| jroll | the point of doing this in newton is so that the data is there, so when someone upgrades to ocata it can be used immediately | 19:05 |
| jroll | it will be, jaypipes pointed that out | 19:05 |
| JayF | /can be/ being the core word, not everyone has to use the new way | 19:05 |
| JayF | but the new way is technically superior so people should want to | 19:05 |
| rloo | right, so if it is backwards compatible, there isn't urgency. it is just a 'nice to have so operators can use it 'immediately'. | 19:05 |
| JayF | There is urgency for people running Ironic in the real world who want this class of bug squashed ;( | 19:06 |
| jroll | well, this is one of two major problems with the nova/ironic interactions, so we'd like to fix it sooner than later | 19:06 |
| jroll | and not doing it potentially pushes some amount of planned work in nova back by a cycle | 19:06 |
| rloo | i don't see how it pushes the planned work in nova back. | 19:06 |
| devananda | rloo: is there a specific reason you _don't_ want to allow this in during newton? | 19:07 |
| jroll | I'm told it does, I don't know all the details or the entire scope of work in nova | 19:07 |
| rloo | oh, i have no problem with allowing it. just that there seems to be something i am not grokking as to why it is necessary. | 19:07 |
| jroll | but they won't just leave our driver behind | 19:07 |
| devananda | I see a lot of reasons to do this soon (help a class of users solve a real problem, build goodwill with nova team, make upgrades to ocata better) | 19:07 |
| devananda | the risk is quite low (an unused API field ... we already have an entire unused API resource type ...) | 19:08 |
| JayF | It also enables us to potentially start coding in ocata for active updates from Ironic -> Placement API | 19:08 |
| devananda | and the effort is pretty low | 19:08 |
| devananda | so I can't see a reason to block it | 19:08 |
| *** jaybeale_ has joined #openstack-ironic | 19:08 | |
| rloo | i'm not blocking it | 19:08 |
| rloo | i'm trying to understand why i'd want to spend time on this as opposed to the other priorities we have for newton. | 19:08 |
| jroll | goodwill++ that's what I was trying to get to earlier, the nova team is doing a ton of work to help us in the realm of scheduling, and they've asked us to do this one thing (quickly) to help them move along on that | 19:08 |
| rloo | ok, so goodwill is a good reason. | 19:08 |
| *** aNupoisc has joined #openstack-ironic | 19:08 | |
| devananda | rloo: so, strictly speaking, it's not necessary. current things should continue to work whether a deployer chooses to populate this field or not | 19:09 |
| devananda | rloo: it's necessary for us to add so that Nova/Ironic interactions can begin to use the new features of the scheduler / placement engine | 19:09 |
| devananda | that'll be optional at first, and probably (I hope!) become required after a cycle or three | 19:09 |
| devananda | so the urgency is for goodwill, and the necessity is more long-term | 19:10 |
| rloo | so necessary in ocata, and really desired in newton to help with ironic upgrade from newton to ocata, right? | 19:10 |
| *** jaybeale has quit IRC | 19:10 | |
| rloo | that is fine then. | 19:10 |
| jroll | right | 19:10 |
| devananda | *to help with both nova and ironic upgrades to ocata | 19:11 |
| devananda | yes | 19:11 |
| rloo | ok | 19:11 |
| rloo | i have to step out for a few minutes, but will review and comment on that spec again in a bit :) | 19:11 |
| jroll | rloo: ok, updating the spec now | 19:12 |
| devananda | rloo: thanks. good questions, btw. thanks for encouraging us to take the time to explain better | 19:12 |
| rloo | thx devananda, jroll! | 19:12 |
| jroll | +1 to what deva said :) | 19:12 |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: [WIP] Hide driver[_internal]_info from observer role https://review.openstack.org/345076 | 19:13 |
| *** Sukhdev has joined #openstack-ironic | 19:15 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Add node.resource_class https://review.openstack.org/345040 | 19:15 |
| jroll | there we are | 19:15 |
| jroll | hopefully I've unconfused that example, someone sanity check me please :) | 19:15 |
| *** sdake has quit IRC | 19:15 | |
| * devananda looks | 19:15 | |
| openstackgerrit | Clif Houck proposed openstack/ironic: Add support for image caching for the agent driver https://review.openstack.org/161832 | 19:16 |
| *** ChrisAusten has joined #openstack-ironic | 19:17 | |
| devananda | jroll: LGTM | 19:17 |
| jroll | cool | 19:17 |
| devananda | jroll: in our release / upgrade notes, we're going to need to point out that the ironic virt drive in newton requires the newton release of ironic | 19:18 |
| *** causten has joined #openstack-ironic | 19:18 | |
| devananda | OR we're going to need to add some handling here to allow it to be backwards compat, which is not mentioned in your spec | 19:18 |
| *** ChrisAusten has quit IRC | 19:19 | |
| jroll | devananda: yeah, I plan to have that in the nova reno | 19:19 |
| devananda | "There will be a small patch that bumps the API version we're using | 19:19 |
| devananda | and passes the field back in the resource_dict. | 19:19 |
| devananda | " | 19:19 |
| devananda | that line in the Nova section - would you mind doing a quick rev to indicate that this is in the plan? | 19:20 |
| jroll | devananda: "this" being add a release note? | 19:20 |
| jroll | I feel like that's implied, but I can add something | 19:20 |
| *** ljjjustin has quit IRC | 19:20 | |
| * devananda rescinds his LGTM, comments on the spec | 19:20 | |
| jroll | heh | 19:21 |
| jroll | devananda: fwiw, this is what I plan the release note to look like https://review.openstack.org/#/c/297895/9/releasenotes/notes/ironic-multitenant-networking-6f124964831d4a6c.yaml | 19:21 |
| *** jcoufal has quit IRC | 19:23 | |
| *** ljjjustin has joined #openstack-ironic | 19:24 | |
| devananda | jroll: cool. that reno looks fine to me | 19:25 |
| jroll | devananda: this is ready to review, btw https://review.openstack.org/#/c/345055/ | 19:25 |
| jroll | the ironic side of this | 19:25 |
| devananda | have we discussed / decided that Nova should not support N-1 version of Ironic ? | 19:25 |
| devananda | I believe we implicitly have supported it, while recommending deployers always upgrade Ironic before Nova | 19:26 |
| jroll | we kinda fell into that when they mandated we pin api versions in our virt driver | 19:26 |
| jroll | else we end up two cycles behind on nova supporting our features | 19:26 |
| devananda | hypothetically, we could pin minimum, and then optionally do cool things if we detect a newer ironic | 19:26 |
| jroll | well, that's what we wanted to do | 19:27 |
| jroll | nobody on the nova team supported that | 19:27 |
| devananda | oh? | 19:27 |
| devananda | hrm | 19:27 |
| jroll | sean was the biggest detractor | 19:27 |
| * jroll finds the patch in question | 19:27 | |
| *** jvgrant_ has quit IRC | 19:27 | |
| devananda | thanks. I'm curious what his reasoning is. that seems like a reasonable approach | 19:27 |
| jroll | oh client changes are passing too https://review.openstack.org/#/c/345080/ | 19:28 |
| jroll | devananda: here's the one in question https://review.openstack.org/#/c/221057/ | 19:29 |
| jroll | that was abandoned and ended up becoming https://review.openstack.org/#/c/221848/ | 19:29 |
| jroll | I feel like there was a few irc conversations there | 19:29 |
| *** livelace has quit IRC | 19:30 | |
| devananda | I see | 19:31 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Add node.resource_class https://review.openstack.org/345040 | 19:31 |
| jroll | there's your update to call out a reno ^ | 19:31 |
| devananda | thanks for digging that up. I disagree with sdague - 10 lines of code for some useful backwards compatibility isn't a huge burden | 19:31 |
| *** gfhellma has joined #openstack-ironic | 19:31 | |
| jroll | devananda: no, but every feature gets some added and it snowballs | 19:32 |
| *** milan has joined #openstack-ironic | 19:32 | |
| jroll | this did come up at the midcycle for a minute | 19:32 |
| jroll | I think if we attempted to document that nova works with n-1 through n+1 ironic, and really make an attempt to make that work (with CI), they'd be more open to it | 19:32 |
| jroll | but this case was like, "because it should work hopefully" | 19:32 |
| JayF | Do we actually know people who want to do that? | 19:33 |
| jroll | idk | 19:33 |
| JayF | And are we going to run Grenade jobs testing all N+1 except Ironic/ | 19:33 |
| JayF | At some point, we have to support what we support, test it well, and expect folks to follow instructions on ordernig of upgrades | 19:33 |
| jroll | right, that's what I'm saying, if we do say we want to officially support and test it, I think nova folks would be more receptive to it | 19:33 |
| devananda | agreed | 19:33 |
| JayF | I'm saying officially supporting and testing it would be a waste of time and resources imo :) | 19:34 |
| jroll | sure, and that's a fair opinion to have | 19:34 |
| devananda | however, the way we're approaching the resource_class field is going to result in NOT supporting Newton Nova + Mitaka Ironic | 19:34 |
| jroll | I'm not voicing an opinion either way right now, just spelling out the way things are today | 19:34 |
| JayF | devananda: My argument would be we never have supported it | 19:34 |
| jroll | devananda: right, but I don't think we've ever supported that | 19:34 |
| JayF | devananda: we're just lucky if it worked | 19:34 |
| devananda | *nod* | 19:35 |
| *** sdake has joined #openstack-ironic | 19:36 | |
| *** jvgrant has joined #openstack-ironic | 19:36 | |
| devananda | apparently, we also don't have documentation on this: http://docs.openstack.org/developer/ironic/deploy/upgrade-guide.html | 19:36 |
| jroll | really? | 19:36 |
| devananda | I thought I remembered there being an upgrade guide . | 19:36 |
| devananda | but now I cant find it | 19:36 |
| openstackgerrit | Chris Krelle proposed openstack/ironic: Add support for the audit middleware https://review.openstack.org/272658 | 19:36 |
| jroll | ugh | 19:37 |
| JayF | I think that ordering was most emphatically stated in Austin | 19:37 |
| jroll | yeah, same | 19:37 |
| JayF | so it wouldn't surprise me that it hasn't been documented yet | 19:37 |
| JayF | but we can fix documentation :) | 19:37 |
| openstackgerrit | Akilan Pughazhendi proposed openstack/ironic: Updating dev quickstart to include compatiblity for newest distros https://review.openstack.org/346894 | 19:37 |
| jroll | it's been stated quite a bit, austin just happened to have 40 minutes of upgrade discussion :) | 19:37 |
| NobodyCam | morning jroll JayF and devananda :) | 19:37 |
| jroll | who wants to take that on? | 19:37 |
| jroll | hi NobodyCam | 19:37 |
| NobodyCam | :) | 19:37 |
| JayF | jroll: I can file a bug for it, and if it doesn't get picked up put it on my todo list | 19:37 |
| jroll | JayF: thanks | 19:38 |
| JayF | jroll: but it looks like the upgrade guide is pretty much release-to-release specific, so maybe 1 bug to add Liberty -> Mitaka upgrade instructions, and indicate it should state that Ironic gets upgraded first | 19:38 |
| jroll | JayF: I feel like we need general upgrade instructions, with special notes for each release (which, hey, we have in reno) | 19:38 |
| devananda | jroll: +1 | 19:39 |
| *** jvgrant has quit IRC | 19:40 | |
| *** aNupoisc has quit IRC | 19:41 | |
| JayF | jroll: devananda: https://bugs.launchpad.net/ironic/+bug/1606675 | 19:41 |
| openstack | Launchpad bug 1606675 in Ironic "Upgrade guide needs to be made generic and improved" [Undecided,New] | 19:41 |
| jroll | thanks JayF | 19:42 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-python-agent: Add metrics support to IPA https://review.openstack.org/119981 | 19:43 |
| *** jvgrant has joined #openstack-ironic | 19:44 | |
| rloo | devananda, jroll, JayF: fwiw, even w/o node.resource_class, if we land the ironic-virt-driver patch for the networking, nova newton will need ironic newton | 19:45 |
| jroll | rloo: yep | 19:46 |
| *** milan is now known as _milan_ | 19:48 | |
| *** milan has joined #openstack-ironic | 19:52 | |
| *** milan has left #openstack-ironic | 19:54 | |
| *** _milan_ has quit IRC | 19:54 | |
| *** spandhe has quit IRC | 19:56 | |
| *** harshs has quit IRC | 19:56 | |
| rloo | jroll: so the resource_class spec has two +2. Shall we wait or +A when Jenkins wakes up? | 19:59 |
| jroll | rloo: I'm fine with now, but I may be biased :) | 20:00 |
| jroll | rloo: better question, do you want anyone else's opinion on that | 20:01 |
| JayF | let me land it :D | 20:01 |
| JayF | my opinion is that I wanna land it, landing things is fun \o/ | 20:01 |
| JayF | lol | 20:01 |
| rloo | jroll: ha ha. i think we have enough folks, esp if JayF lands it :) | 20:01 |
| * jroll stands by to un -2 his own patch | 20:01 | |
| rloo | what i want to know, is why is jenkins so slow today? | 20:01 |
| JayF | they're doing the changeover to xenial, so says the list | 20:02 |
| JayF | but no notes about availability issues around that | 20:02 |
| rloo | JayF: ah. i was worried it would be like this til newton release time. | 20:02 |
| JayF | oh god I hope not :x | 20:02 |
| *** spandhe has joined #openstack-ironic | 20:05 | |
| *** mtanin___ has joined #openstack-ironic | 20:08 | |
| *** mtanino has quit IRC | 20:09 | |
| JayF | jroll: is it at all possible the recent network changes landing have rendered our devstack howto wrong? | 20:10 |
| JayF | jroll: because I have a devstack that's almost 100% stock, reliably failing to boot agents due to dhcp failures (it won't pxe) | 20:11 |
| *** mtanino has joined #openstack-ironic | 20:11 | |
| jroll | JayF: doubtful - are you doing a deploy or just powering them up? | 20:11 |
| JayF | jroll: doing a deploy | 20:12 |
| *** mtanin___ has quit IRC | 20:12 | |
| JayF | jroll: dnsmasq is clearly running; I'm highly suspicious that the network may not be setup correctly | 20:12 |
| jroll | JayF: hrm, that's a problem... maybe compare local.conf to something in the gate? | 20:12 |
| jroll | it's certainly possible we broke our dev docs | 20:12 |
| JayF | the local.conf is copy+pasted from our documentation | 20:13 |
| JayF | Yeah, that's what I'm thinking | 20:13 |
| jroll | right, I got that | 20:13 |
| JayF | and it's cost me almost my whole morning | 20:13 |
| jroll | I just can't tell you 100% if we did break it or not | 20:13 |
| *** aNupoisc has joined #openstack-ironic | 20:14 | |
| *** fragatina has joined #openstack-ironic | 20:15 | |
| rloo | jroll: going back to our short discussion earlier today about node.network_interface not reflecting the actual network_interface being used. I just re-skimmed the driver composition spec. | 20:16 |
| rloo | jroll: there is *nothing* that mentions how to get access to the actual interfaces being used by a node. | 20:16 |
| rloo | jroll: i don't think we can use GET /v1/drivers/<name> to get that info cuz it may be specific to a node | 20:17 |
| rloo | jroll: which makes me think we need two set of node.*_interfaces...?? one to indicate what we'd like to override if anything, and one that indicates what is being used. | 20:17 |
| rloo | jroll: or we add an API to eg GET /v1/nodes/<node>/interfaces | 20:18 |
| *** aNupoisc has quit IRC | 20:20 | |
| *** aNupoisc has joined #openstack-ironic | 20:20 | |
| aNupoisc | Hi cinerama: TheJulia: I was able to manually install shade and set it to work. Somehow ansible was unable to find shade module inside virtual environment for me. | 20:22 |
| aNupoisc | cinerama: TheJulia: but this time I am having a new question, i was able to provision but the ipa image got stuck on the boot. | 20:24 |
| *** spandhe has left #openstack-ironic | 20:25 | |
| *** moshele has joined #openstack-ironic | 20:25 | |
| *** mtanin___ has joined #openstack-ironic | 20:30 | |
| *** moshele has quit IRC | 20:31 | |
| *** mtanino has quit IRC | 20:32 | |
| *** harshs has joined #openstack-ironic | 20:38 | |
| openstackgerrit | Michael Turek proposed openstack/ironic: Add classes for Portgroups API enablement https://review.openstack.org/347549 | 20:39 |
| mjturek1 | fyi - working on splitting up the portgroups API per a discussion during last week's ironic neutron meeting ^ | 20:40 |
| mjturek1 | seeing if the tests actually run | 20:40 |
| *** jtomasek has joined #openstack-ironic | 20:40 | |
| *** bvandewa_ has quit IRC | 20:41 | |
| *** jtomasek has quit IRC | 20:41 | |
| *** jtomasek has joined #openstack-ironic | 20:41 | |
| TheJulia | aNupoisc: define stuck on boot | 20:43 |
| aNupoisc | TheJulia: yes a minute. | 20:43 |
| TheJulia | NobodyCam: https://review.openstack.org/#/c/325599/ | 20:43 |
| *** moshele has joined #openstack-ironic | 20:43 | |
| aNupoisc | ixgbe driver is getting stuck. I can see a message "non blocking pool is initialized" from ixgbe driver | 20:45 |
| aNupoisc | theJulia | 20:45 |
| aNupoisc | does this help? Or is there a better way to get logs? I am just reading from the console | 20:45 |
| aNupoisc | TheJulia | 20:46 |
| *** e0ne has quit IRC | 20:46 | |
| TheJulia | aNupoisc: what IPA image are you using? | 20:46 |
| *** ijw has joined #openstack-ironic | 20:47 | |
| *** ahqa has joined #openstack-ironic | 20:48 | |
| *** bvandewa has joined #openstack-ironic | 20:48 | |
| aNupoisc | TheJulia: debian | 20:49 |
| aNupoisc | ubuntu os | 20:49 |
| *** rama_y has quit IRC | 20:49 | |
| *** joprovost has quit IRC | 20:51 | |
| *** catintheroof has quit IRC | 20:52 | |
| *** ijw has quit IRC | 20:54 | |
| *** bvandewa has quit IRC | 20:54 | |
| *** jjohnson2__ has quit IRC | 20:56 | |
| gmmaha | TheJulia: aNupoisc: the IPA image is coreos. the default. the deployment image is ubuntu | 20:57 |
| aNupoisc | gmmaha: ah okay | 20:58 |
| ahqa | Hi, I have build an Ubuntu Xenial (16.04) image with elements "vm", "simiple-init" and "enable-serial-console". Image can be deployed on server but there is no ip registered to it. I am using config drive to assign network and metadata. I checked via console and I see no config drive mounted on system. I am using DHCP server and PXE boot to deploy images. I am not sure why config drive is not mounting on system. | 20:59 |
| gmmaha | ahqa: a while back there was an issue with ubuntu images and simple-init working. https://bugs.launchpad.net/bifrost/+bug/1541566 | 21:01 |
| openstack | Launchpad bug 1541566 in bifrost "simple-init does not work with latest trusty ubuntu cloud image" [Medium,In progress] | 21:01 |
| gmmaha | not sure if a fix has landed yet, but the symptoms seem to point to the same. | 21:01 |
| gmmaha | try using cloud-init-datasources instead of simple-init | 21:02 |
| *** trown is now known as trown|outtypewww | 21:02 | |
| ahqa | gmmaha: Thank you, I will give it a try. | 21:04 |
| gmmaha | ahqa: good luck1 | 21:04 |
| *** rama_y has joined #openstack-ironic | 21:05 | |
| openstackgerrit | Merged openstack/ironic-specs: Add node.resource_class https://review.openstack.org/345040 | 21:06 |
| *** harshs has quit IRC | 21:09 | |
| mrda | Morning Ironic | 21:10 |
| *** harshs has joined #openstack-ironic | 21:10 | |
| NobodyCam | Morning mrda | 21:12 |
| mrda | hey NobodyCam | 21:13 |
| NobodyCam | :) | 21:14 |
| * jroll out for the night, see y'all tomorrow | 21:14 | |
| NobodyCam | have a good night jroll | 21:14 |
| *** bvandewa has joined #openstack-ironic | 21:15 | |
| *** ijw has joined #openstack-ironic | 21:17 | |
| JayF | jroll: you might wanna un-2 your resource_class patches if you aren't gone yet | 21:17 |
| JayF | ^^ since the spec just merged \o/ | 21:17 |
| jroll | JayF: I did :) | 21:17 |
| JayF | niiiiice | 21:17 |
| mrda | Night jroll | 21:17 |
| jroll | morning mrda ;) | 21:18 |
| *** bvandewa_ has joined #openstack-ironic | 21:19 | |
| *** ijw has quit IRC | 21:21 | |
| *** ijw has joined #openstack-ironic | 21:22 | |
| *** bvandewa has quit IRC | 21:22 | |
| *** bvandewa has joined #openstack-ironic | 21:23 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic-specs: Baremetal Trust using Intel TXT was done in 4.2 https://review.openstack.org/347557 | 21:24 |
| *** rm_work has quit IRC | 21:25 | |
| *** bvandewa_ has quit IRC | 21:26 | |
| JayF | rloo: ^ do you think those are simple enough to +2+A without a second +2? | 21:26 |
| *** rm_work has joined #openstack-ironic | 21:26 | |
| JayF | rloo: I do, just wanted a second opinion | 21:26 |
| *** jaybeale_ has quit IRC | 21:27 | |
| rloo | JayF: well, I do, but I submitted it. | 21:27 |
| JayF | I'm asking more for affirmation, since I know our policy is "use discretion" | 21:27 |
| rloo | JayF: yeah, i think it is fairly straightforward. | 21:27 |
| rloo | JayF: it is just 'moving' it around to indicate it was done. | 21:28 |
| JayF | yep, I landed it :) | 21:28 |
| *** ijw has quit IRC | 21:28 | |
| JayF | ty for doing that cleanup | 21:28 |
| *** harshs has quit IRC | 21:31 | |
| *** harshs has joined #openstack-ironic | 21:38 | |
| * TheJulia feels bad for getting hyper distracted | 21:42 | |
| TheJulia | aNupoisc: It sounds like you need to getinside of your running IPA and see whats going on if it is hanging and not actually doing anything. The most important question would be if it heartbeats to the conductor, you can see that in the driver_internal_info, there is a field that will update whenever it heartbeats with the seconds from epoch of the last heartbeat | 21:45 |
| *** ayoung has quit IRC | 21:45 | |
| rloo | thx JayF. there was something i wanted to ask you. I think it was install guides. Do you know what is involved in getting our install guides incorporated with openstack's? i think it is possible now? | 21:45 |
| *** ccamacho|away has quit IRC | 21:46 | |
| thiagop | ow man, 2hs lost due to a malformed build on jenkins-job-builder... | 21:46 |
| * thiagop wanna cry | 21:47 | |
| * TheJulia slides a beer in front of thiagop | 21:48 | |
| * thiagop drinks it in one shot | 21:49 | |
| * TheJulia blinks | 21:49 | |
| *** ijw has joined #openstack-ironic | 21:49 | |
| thiagop | TheJulia: The CI host vm was taking too much time to build a new image, so I tried to resize it... 3hs later it doesn't. I went to configure a new CI host and kinda solve it in around 30min but the jobs weren't building because of^ | 21:51 |
| *** bvandewa has quit IRC | 21:52 | |
| *** bvandewa has joined #openstack-ironic | 21:52 | |
| thiagop | now it's fine, but will need to craft a new image anyway | 21:53 |
| *** ijw has quit IRC | 21:54 | |
| *** ijw has joined #openstack-ironic | 21:56 | |
| aNupoisc | TheJulia: Oh okay may be i need to dig deeper to find this. Thanks, will check on that thing. | 21:56 |
| openstackgerrit | Chris Krelle proposed openstack/bifrost: WIP: initial work to add ISC-DHCP https://review.openstack.org/343947 | 21:57 |
| *** harshs has quit IRC | 22:01 | |
| *** bvandewa has quit IRC | 22:01 | |
| *** openstackgerrit has quit IRC | 22:03 | |
| *** openstackgerrit has joined #openstack-ironic | 22:04 | |
| NobodyCam | thiagop: :( | 22:05 |
| anteaya | so hopefully the channel grows some lurkers from the scientific working group | 22:06 |
| anteaya | great group of folks | 22:06 |
| *** sdake has quit IRC | 22:09 | |
| openstackgerrit | Merged openstack/ironic-specs: Baremetal Trust using Intel TXT was done in 4.2 https://review.openstack.org/347557 | 22:11 |
| *** ijw has quit IRC | 22:13 | |
| *** bvandewa has joined #openstack-ironic | 22:14 | |
| NobodyCam | Hi anteaya | 22:14 |
| anteaya | NobodyCam: hey there | 22:15 |
| anteaya | how are you? | 22:16 |
| *** wajdi_ has joined #openstack-ironic | 22:22 | |
| *** wajdi has quit IRC | 22:22 | |
| *** priteau has quit IRC | 22:23 | |
| *** wajdi has joined #openstack-ironic | 22:29 | |
| *** wajdi_ has quit IRC | 22:32 | |
| *** wajdi_ has joined #openstack-ironic | 22:33 | |
| NobodyCam | doing well, | 22:33 |
| NobodyCam | trying to get my head back in to the swing of things | 22:34 |
| *** wajdi has quit IRC | 22:34 | |
| NobodyCam | How have you been | 22:34 |
| *** wajdi_ has quit IRC | 22:38 | |
| *** joprovost has joined #openstack-ironic | 22:43 | |
| *** rpioso has quit IRC | 22:44 | |
| *** sdake has joined #openstack-ironic | 22:47 | |
| openstackgerrit | Mario Villaplana proposed openstack/ironic: Add power state change notifications https://review.openstack.org/321865 | 22:47 |
| *** openstackgerrit has quit IRC | 22:48 | |
| *** openstackgerrit has joined #openstack-ironic | 22:48 | |
| *** joprovost1 has joined #openstack-ironic | 22:48 | |
| *** sdake_ has joined #openstack-ironic | 22:49 | |
| *** joprovost has quit IRC | 22:50 | |
| *** joprovost1 is now known as joprovost | 22:50 | |
| *** siva_krish has quit IRC | 22:52 | |
| *** sdake has quit IRC | 22:52 | |
| *** baoli has quit IRC | 22:52 | |
| *** baoli has joined #openstack-ironic | 22:53 | |
| *** michauds has quit IRC | 22:55 | |
| *** PollyZ has quit IRC | 22:58 | |
| *** Sukhdev has quit IRC | 23:00 | |
| *** Sukhdev has joined #openstack-ironic | 23:03 | |
| *** bvandewa has quit IRC | 23:11 | |
| *** bvandew__ has joined #openstack-ironic | 23:11 | |
| NobodyCam | devananda: 325599 looks great +2 any one else want to look before I +a it? | 23:11 |
| NobodyCam | it now has 3 +2's | 23:11 |
| *** Nisha_away has joined #openstack-ironic | 23:11 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Remove discover from test-requirements https://review.openstack.org/345513 | 23:17 |
| *** moshele has quit IRC | 23:17 | |
| *** rama_y has quit IRC | 23:19 | |
| *** bvandewa has joined #openstack-ironic | 23:24 | |
| *** fumie has joined #openstack-ironic | 23:25 | |
| *** bvandew__ has quit IRC | 23:28 | |
| *** fumie has quit IRC | 23:29 | |
| *** fumie has joined #openstack-ironic | 23:29 | |
| devananda | jroll: some ramblings posted to https://etherpad.openstack.org/p/ironic-v2-api | 23:32 |
| devananda | surprisingly, as I have gone through the exercise of enumerating these issues, I'm starting to believe that we actually could do this in v1 | 23:33 |
| *** baoli has quit IRC | 23:38 | |
| *** thrash is now known as thrash|g0ne | 23:38 | |
| *** gfhellma has quit IRC | 23:40 | |
| *** yuanying has joined #openstack-ironic | 23:45 | |
| *** lujinluo has joined #openstack-ironic | 23:51 | |
| *** fragatina has quit IRC | 23:52 | |
| *** hoangcx2 has joined #openstack-ironic | 23:54 | |
| *** hoangcx has quit IRC | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!