*** cdearborn has quit IRC | 00:00 | |
*** jaosorior has quit IRC | 00:00 | |
*** amotoki has joined #openstack-ironic | 00:04 | |
openstackgerrit | Merged openstack/ironic: Add dynamic interfaces fields to nodes API https://review.openstack.org/424720 | 00:07 |
---|---|---|
*** ijw has quit IRC | 00:09 | |
*** sayantani01 has joined #openstack-ironic | 00:10 | |
*** dims_ has joined #openstack-ironic | 00:10 | |
*** dims has quit IRC | 00:11 | |
*** sayantani01 has quit IRC | 00:15 | |
*** vsaienko has quit IRC | 00:26 | |
*** MattMan has quit IRC | 00:32 | |
*** MattMan has joined #openstack-ironic | 00:32 | |
*** causten has quit IRC | 00:43 | |
*** rama_y has quit IRC | 00:44 | |
*** causten has joined #openstack-ironic | 00:44 | |
*** causten has quit IRC | 00:44 | |
*** crushil has quit IRC | 00:45 | |
*** crushil has joined #openstack-ironic | 00:45 | |
*** rajinir has quit IRC | 00:46 | |
*** sdake has quit IRC | 00:50 | |
openstackgerrit | Bob Fournier proposed openstack/ironic-inspector: Add plugin to process basic management LLDP TLVs https://review.openstack.org/406496 | 00:51 |
openstackgerrit | Merged openstack/ironic: [Devstack] Add stack user to libvirt group https://review.openstack.org/426664 | 00:52 |
*** sdake has joined #openstack-ironic | 00:52 | |
*** absubram has joined #openstack-ironic | 00:55 | |
*** rpioso has quit IRC | 01:03 | |
*** baoli has joined #openstack-ironic | 01:24 | |
*** Syed__ has quit IRC | 01:25 | |
*** rloo has quit IRC | 01:29 | |
*** maratoid_ has quit IRC | 01:32 | |
*** Goneri has joined #openstack-ironic | 01:54 | |
*** baoli has quit IRC | 02:15 | |
*** aNuposic has quit IRC | 02:15 | |
*** baoli has joined #openstack-ironic | 02:15 | |
*** Sukhdev has quit IRC | 02:16 | |
*** yuanying has quit IRC | 02:21 | |
*** Goneri has quit IRC | 02:41 | |
openstackgerrit | Naohiro Tamura proposed openstack/ironic: Update the existing APIs due to adding get_supported_power_states https://review.openstack.org/382194 | 02:43 |
openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add supported_power_states field to api-ref https://review.openstack.org/427000 | 02:46 |
*** dsneddon has joined #openstack-ironic | 03:07 | |
*** dsneddon has quit IRC | 03:07 | |
*** owalsh is now known as owalsh-afk | 03:23 | |
*** crushil has quit IRC | 03:28 | |
*** vikrant has joined #openstack-ironic | 03:34 | |
*** mtanino has quit IRC | 03:43 | |
*** absubram has quit IRC | 03:50 | |
*** sdake has quit IRC | 03:53 | |
*** baoli has quit IRC | 03:58 | |
*** aparnav has joined #openstack-ironic | 03:58 | |
*** owalsh has joined #openstack-ironic | 03:59 | |
*** owalsh-afk has quit IRC | 04:00 | |
*** crushil has joined #openstack-ironic | 04:02 | |
*** crushil has quit IRC | 04:20 | |
*** crushil has joined #openstack-ironic | 04:21 | |
*** rbudden has quit IRC | 04:31 | |
*** crushil has quit IRC | 04:31 | |
*** udesale has joined #openstack-ironic | 04:37 | |
*** sdake has joined #openstack-ironic | 04:44 | |
*** Nisha_Agarwal has joined #openstack-ironic | 04:56 | |
*** sasha2 has quit IRC | 05:01 | |
*** sdake_ has joined #openstack-ironic | 05:05 | |
*** sdake has quit IRC | 05:05 | |
*** wrouesnel has quit IRC | 05:08 | |
*** wrouesnel has joined #openstack-ironic | 05:08 | |
*** wrouesnel has quit IRC | 05:09 | |
*** wrouesnel has joined #openstack-ironic | 05:09 | |
*** Sukhdev has joined #openstack-ironic | 05:13 | |
*** wrouesnel has quit IRC | 05:27 | |
*** wrouesnel has joined #openstack-ironic | 05:27 | |
*** wrouesnel has quit IRC | 05:29 | |
*** wrouesnel has joined #openstack-ironic | 05:29 | |
*** nmathew has joined #openstack-ironic | 05:35 | |
*** wrouesnel has quit IRC | 05:37 | |
*** wrouesnel has joined #openstack-ironic | 05:37 | |
*** wrouesnel has quit IRC | 05:39 | |
*** wrouesnel has joined #openstack-ironic | 05:39 | |
*** jaganathan has joined #openstack-ironic | 05:45 | |
*** rbudden has joined #openstack-ironic | 05:45 | |
*** ijw has joined #openstack-ironic | 05:48 | |
*** wrouesnel has quit IRC | 05:51 | |
*** wrouesnel has joined #openstack-ironic | 05:51 | |
*** ijw has quit IRC | 05:53 | |
*** wrouesnel has quit IRC | 05:53 | |
*** wrouesnel has joined #openstack-ironic | 05:54 | |
*** wrouesnel has quit IRC | 05:56 | |
*** wrouesnel has joined #openstack-ironic | 05:56 | |
*** Sukhdev has quit IRC | 05:57 | |
*** wrouesnel has quit IRC | 06:00 | |
*** wrouesnel has joined #openstack-ironic | 06:00 | |
*** Nisha_Agarwal has quit IRC | 06:00 | |
*** wrouesnel has quit IRC | 06:04 | |
*** wrouesnel has joined #openstack-ironic | 06:04 | |
*** yarkot has quit IRC | 06:06 | |
*** wrouesnel has quit IRC | 06:06 | |
*** wrouesnel has joined #openstack-ironic | 06:07 | |
*** wrouesnel has quit IRC | 06:10 | |
*** wrouesnel has joined #openstack-ironic | 06:10 | |
*** wrouesnel has quit IRC | 06:12 | |
*** wrouesnel has joined #openstack-ironic | 06:12 | |
*** wrouesnel has quit IRC | 06:12 | |
*** wrouesnel has joined #openstack-ironic | 06:13 | |
*** wrouesnel has quit IRC | 06:13 | |
*** wrouesnel has joined #openstack-ironic | 06:13 | |
*** wrouesnel has quit IRC | 06:14 | |
*** wrouesnel has joined #openstack-ironic | 06:15 | |
*** wrouesnel has quit IRC | 06:15 | |
*** wrouesnel has joined #openstack-ironic | 06:15 | |
*** sdake_ is now known as sdake | 06:16 | |
*** yolanda has quit IRC | 06:23 | |
*** chedda has joined #openstack-ironic | 06:24 | |
*** rcernin has joined #openstack-ironic | 06:29 | |
*** wrouesnel has quit IRC | 06:30 | |
*** wrouesnel has joined #openstack-ironic | 06:30 | |
*** wrouesnel has quit IRC | 06:31 | |
*** wrouesnel has joined #openstack-ironic | 06:31 | |
*** yarkot has joined #openstack-ironic | 06:31 | |
*** Nisha_Agarwal has joined #openstack-ironic | 06:31 | |
*** mjura has joined #openstack-ironic | 06:33 | |
*** wrouesnel has quit IRC | 06:34 | |
*** wrouesnel has joined #openstack-ironic | 06:34 | |
*** wrouesnel has quit IRC | 06:35 | |
*** wrouesnel has joined #openstack-ironic | 06:35 | |
*** wrouesnel has quit IRC | 06:37 | |
*** wrouesnel has joined #openstack-ironic | 06:37 | |
*** rbudden has quit IRC | 06:39 | |
*** wrouesnel has quit IRC | 06:39 | |
*** wrouesnel has joined #openstack-ironic | 06:39 | |
*** absubram has joined #openstack-ironic | 06:49 | |
*** fxpester has joined #openstack-ironic | 06:53 | |
*** absubram has quit IRC | 06:54 | |
*** absubram has joined #openstack-ironic | 07:02 | |
*** tesseract has joined #openstack-ironic | 07:18 | |
*** pcaruana has joined #openstack-ironic | 07:19 | |
*** foexle has joined #openstack-ironic | 07:26 | |
*** xek__ is now known as xek | 07:31 | |
*** amarao has joined #openstack-ironic | 07:34 | |
openstackgerrit | jiang wei proposed openstack/python-ironicclient: Support i18n for baremetal node cmds https://review.openstack.org/357639 | 07:38 |
*** chedda has quit IRC | 07:39 | |
*** ChubYann has quit IRC | 07:42 | |
*** pmannidi has quit IRC | 07:43 | |
*** chedda has joined #openstack-ironic | 07:49 | |
*** chedda has quit IRC | 07:49 | |
*** rbartal has joined #openstack-ironic | 07:51 | |
*** rbartal__ has joined #openstack-ironic | 07:51 | |
*** rbartal_ has joined #openstack-ironic | 07:52 | |
*** yolanda has joined #openstack-ironic | 08:04 | |
*** Nisha_Agarwal has quit IRC | 08:05 | |
*** slagle has quit IRC | 08:05 | |
*** rnoriega has quit IRC | 08:07 | |
*** lhinds has quit IRC | 08:08 | |
*** ralonsoh has joined #openstack-ironic | 08:14 | |
*** ccamacho has joined #openstack-ironic | 08:15 | |
*** mgoddard has joined #openstack-ironic | 08:27 | |
*** foexle has quit IRC | 08:29 | |
*** yuanying has joined #openstack-ironic | 08:30 | |
vdrok | morning ironic! | 08:47 |
*** amotoki has quit IRC | 08:49 | |
*** priteau has joined #openstack-ironic | 08:51 | |
*** dtantsur|afk is now known as dtantsur | 08:55 | |
dtantsur | morning ironic | 08:55 |
*** mgoddard has quit IRC | 08:56 | |
openstackgerrit | Galyna Zholtkevych proposed openstack/ironic: ETAGs maintenance for all api node's requests https://review.openstack.org/392213 | 08:59 |
*** zzzeek has quit IRC | 09:00 | |
*** zzzeek has joined #openstack-ironic | 09:00 | |
openstackgerrit | Galyna Zholtkevych proposed openstack/ironic: ETAGs maintenance for all api node's requests https://review.openstack.org/392213 | 09:01 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-ui: Imported Translations from Zanata https://review.openstack.org/426667 | 09:04 |
*** absubram has quit IRC | 09:04 | |
*** Nisha_Agarwal has joined #openstack-ironic | 09:06 | |
openstackgerrit | Galyna Zholtkevych proposed openstack/ironic: ETAGs maintenance for all api node's requests https://review.openstack.org/392213 | 09:08 |
openstackgerrit | Galyna Zholtkevych proposed openstack/python-ironicclient: Client support resources filecache https://review.openstack.org/420033 | 09:22 |
*** e0ne has joined #openstack-ironic | 09:23 | |
aparnav | hey dtantsur, Can you review the documentation patch https://review.openstack.org/#/c/408441/ when you find time? | 09:25 |
patchbot | patch 408441 - ironic - Doc for sanitize erase support in iLO drivers | 09:25 |
dtantsur | sure, I can't promise when exactly though | 09:27 |
*** amotoki has joined #openstack-ironic | 09:30 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Do not change admin_state for tenant port https://review.openstack.org/427603 | 09:32 |
*** mgoddard_ has joined #openstack-ironic | 09:33 | |
aarefiev | morning all! | 09:37 |
*** amotoki has quit IRC | 09:38 | |
*** lucas-afk is now known as lucasagomes | 09:38 | |
lucasagomes | morning all! | 09:40 |
aarefiev | dtantsur: morning, why are going to delete jq from newton? | 09:40 |
aarefiev | morning lucasagomes | 09:40 |
dtantsur | aarefiev, did we use it in newton? | 09:40 |
dtantsur | also morning lucasagomes, aarefiev | 09:41 |
aarefiev | dtantsur: we have bash exercise in newton | 09:41 |
dtantsur | oh, do we? | 09:41 |
aarefiev | yeah | 09:42 |
*** Nisha_Agarwal has quit IRC | 09:42 | |
dtantsur | then why does the CI pass? :) | 09:42 |
dtantsur | anyway, abandoning then | 09:43 |
aarefiev | dtantsur: we are using tempest on ci, but keep exercise in repo | 09:43 |
dtantsur | correct, yeah. anyway, it's just a small optimization | 09:44 |
aarefiev | dtantsur: would you like to review https://review.openstack.org/#/c/264257/ | 09:45 |
patchbot | patch 264257 - ironic-inspector - Adding InfiniBand Support | 09:45 |
*** amotoki has joined #openstack-ironic | 09:45 | |
*** amotoki has quit IRC | 09:47 | |
dtantsur | aarefiev, not really :) feel free to approve, if you think it's good | 09:47 |
aarefiev | yeah, I'm good , last 5 patch sets was about docs, notes :) | 09:48 |
dtantsur | we're close to FF, so if we want it landed - we should go for it | 09:48 |
dtantsur | ditto for the LLDP thingy | 09:49 |
aarefiev | I've +2 on LLDP, there are lot of nits I'd like to see, nut nothing critical | 09:50 |
*** salmankhan has joined #openstack-ironic | 09:50 | |
*** amotoki has joined #openstack-ironic | 09:50 | |
dtantsur | cool | 09:51 |
dtantsur | pas-ha, hey! I wonder if you want to revive https://review.openstack.org/#/c/394221/ yourself, if it's still important for you. | 09:52 |
patchbot | patch 394221 - python-dracclient - Add support for idrac_card and lifecycle_controlle... | 09:52 |
openstackgerrit | Galyna Zholtkevych proposed openstack/ironic: ETAGs maintenance for all api node's requests https://review.openstack.org/392213 | 09:52 |
dtantsur | aarefiev, actually it seems like bfournie has fixed your nits, so it could use another +2 | 09:53 |
aarefiev | oh, I missed that, I'll take a look | 09:53 |
*** e0ne has quit IRC | 09:56 | |
*** udesale has quit IRC | 10:04 | |
dtantsur | JayF, please remove and add back your W+1 on https://review.openstack.org/#/c/418359/ to push it through the gate | 10:06 |
patchbot | patch 418359 - python-ironicclient - Add args to CLI 'node-create' for selecting hardwa... | 10:06 |
openstackgerrit | Oksana Voshchana proposed openstack/ironic: WIP: TEST GRENADE DEV https://review.openstack.org/397126 | 10:08 |
openstackgerrit | Pavlo Shchelokovskyy proposed openstack/bifrost: Remove local actions for ssh keys https://review.openstack.org/426188 | 10:08 |
openstackgerrit | Oksana Voshchana proposed openstack/ironic: WIP: TEST GRENADE DEV https://review.openstack.org/397126 | 10:10 |
vdrok | morning dtantsur aarefiev and lucasagomes ! | 10:21 |
aarefiev | morning vdrok :) | 10:21 |
lucasagomes | o/ | 10:22 |
openstackgerrit | Sofiia Andriichenko proposed openstack/python-ironicclient: Fix cleanup of resources https://review.openstack.org/424528 | 10:31 |
vsaienk0 | morning lucasagomes, dtantsur, vdrok, sambetts, aarefiev milan and all Ironic'ers! | 10:39 |
*** athomas has joined #openstack-ironic | 10:39 | |
*** salmankhan has quit IRC | 10:40 | |
vsaienk0 | dtantsur could you please have look at new patchset https://review.openstack.org/#/c/426700 | 10:40 |
patchbot | patch 426700 - openstack-infra/project-config - Update experimental ironic-standalone job | 10:40 |
vdrok | morning vsaienk0 | 10:40 |
jroll | morning ironic | 10:41 |
*** milan has joined #openstack-ironic | 10:41 | |
milan | morning Ironic! : | 10:41 |
jroll | zuul should be happy now yes? | 10:42 |
milan | :) | 10:42 |
jroll | hey milan | 10:42 |
milan | morning jroll! :) | 10:42 |
openstackgerrit | Sofiia Andriichenko proposed openstack/python-ironicclient: Fix cleanup of resources https://review.openstack.org/424545 | 10:47 |
vdrok | morning milan and jroll | 10:49 |
vdrok | appears that we don't have the feature parity of OSC with ironic client yet :( I see https://review.openstack.org/362047 and https://review.openstack.org/381153 | 10:50 |
patchbot | patch 362047 - python-ironicclient - Add command RAID logical disk properties | 10:50 |
patchbot | patch 381153 - python-ironicclient - Add command `driver properties` to osc | 10:50 |
jroll | indeed, we're getting close, though | 10:51 |
milan | morning vdrok! :) | 10:51 |
*** salmankhan has joined #openstack-ironic | 10:58 | |
aarefiev | morning vsaienk0, jroll, vdrok | 10:58 |
dtantsur | morning jroll | 10:58 |
dtantsur | jroll, I've got a request from the RDO team to use Sem-Ver tag explained in http://docs.openstack.org/developer/pbr/#version | 10:59 |
dtantsur | otherwise due to how pbr composes versions, we now have master version smaller than newton | 10:59 |
dtantsur | e.g. https://github.com/openstack/gnocchi/commit/63e1cd95e7bc53e9bff2f2b71262ed729581a6e2 | 10:59 |
sambetts | Morning all | 11:00 |
jroll | dtantsur: the request is to use it each time we cut a branch, or on every commit? | 11:00 |
dtantsur | morning sambetts, vdrok | 11:00 |
dtantsur | jroll, well, pbr developers recommend every commit, but it seems like too much to ask.. so probably after we cut a branch | 11:01 |
jroll | dtantsur: yeah, I have no problem with once per branch | 11:01 |
jroll | dtantsur: I wonder if release folks would like that in their post-branch automation | 11:01 |
dtantsur | dhellmann, ^^^ | 11:02 |
jroll | I think that would make it easier for everyone | 11:02 |
dtantsur | jroll, meanwhile, when do we plan on ironic release? does it make sense to create such commit already to fix upgrades? | 11:03 |
jroll | dtantsur: otherwise, go ahead and add a patch to the release docs | 11:03 |
jroll | dtantsur: "fix upgrades" ? | 11:03 |
jroll | you mean fix RDO upgrades, yes? :) | 11:03 |
dtantsur | well, yeah :) any package-based, when a version is derived from setup.py --version | 11:04 |
jroll | dtantsur: yeah, I'm fine with it | 11:04 |
dtantsur | k, will do a bit later | 11:05 |
jroll | cool | 11:05 |
* jroll afk for a while | 11:05 | |
*** salmankhan has quit IRC | 11:05 | |
*** salmankhan has joined #openstack-ironic | 11:05 | |
openstackgerrit | Merged openstack/python-ironicclient: Add args to CLI 'node-create' for selecting hardware interfaces https://review.openstack.org/418359 | 11:07 |
dtantsur | \o/ | 11:07 |
*** jkilpatr_ has quit IRC | 11:08 | |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Add sem-ver flag so pbr generates correct version https://review.openstack.org/427643 | 11:16 |
dtantsur | jroll, ^^^ | 11:16 |
dtantsur | now setup.py --version is 6.3.0.dev701 | 11:17 |
*** amotoki has quit IRC | 11:18 | |
*** e0ne has joined #openstack-ironic | 11:22 | |
vsaienk0 | jroll: are we going to add scenario tests for portgroups, if yes I'll need to rebase https://review.openstack.org/#/c/381743 | 11:33 |
patchbot | patch 381743 - ironic - Ironic devstack portgroup support | 11:33 |
openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Add dynamic interfaces fields to base node notification https://review.openstack.org/427651 | 11:40 |
*** nmathew has quit IRC | 11:40 | |
openstackgerrit | Merged openstack/ironic: Update guide for PXE multi-architecture setup https://review.openstack.org/386313 | 11:41 |
mat128 | good morning | 11:51 |
openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Remove deprecated heartbeat policy check https://review.openstack.org/427654 | 11:51 |
openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Add dynamic interfaces fields to base node notification https://review.openstack.org/427651 | 11:53 |
*** salmankhan has quit IRC | 11:59 | |
vdrok | morning mat128 | 11:59 |
*** salmankhan has joined #openstack-ironic | 11:59 | |
vdrok | and good morning sambetts | 11:59 |
openstackgerrit | Merged openstack/ironic-inspector: Deprecated log_bmc_address option https://review.openstack.org/425642 | 12:00 |
openstackgerrit | Merged openstack/ironic-inspector: Clean up deprecated configuration options https://review.openstack.org/424659 | 12:00 |
hwoarang | TheJulia: yolanda: Good day. Do you have any idea what this error http://paste.opensuse.org/6463487 observed every now and then in gate-bifrost-integration-tinyipa-ubuntu-xenial jobs is really about? | 12:02 |
openstackgerrit | edan david proposed openstack/ironic-inspector: Adding InfiniBand Support https://review.openstack.org/264257 | 12:02 |
aparnav | hey yuriyz, vdrok, Can you take a look at these patches https://review.openstack.org/408441 https://review.openstack.org/#/c/422572/ when you find time? | 12:03 |
patchbot | patch 408441 - ironic - Doc for sanitize erase support in iLO drivers | 12:03 |
patchbot | patch 422572 - ironic - Support `update_firmware` to hpsum based firmware ... | 12:03 |
mat128 | hwoarang: My bet is that this happens on the ipv6-only nodes | 12:03 |
mat128 | hwoarang: ['ansible_' + ans_network_interface]['ipv4'] and it complains that it can't find the ipv4 key in that dictionary | 12:03 |
yuriyz | hi ironic aparnav look today later | 12:03 |
aparnav | yuriyz, thank you :) | 12:04 |
mat128 | hwoarang: there are a few places where ipv4 is assumed (e.g. https://github.com/openstack/bifrost/blob/master/playbooks/roles/bifrost-ironic-install/templates/inspector-default-boot-ipxe.j2 and https://github.com/openstack/bifrost/blob/master/playbooks/roles/ironic-enroll-dynamic/defaults/main.yml) | 12:07 |
mat128 | hwoarang: I suggest you open a bug in bifrost | 12:07 |
hwoarang | mat128: ok I will. Thank you for the information! | 12:07 |
mat128 | TheJulia: transient gate failure in bifrost might be related to ^ and using ipv6-only nodes on the OSIC cloud | 12:08 |
mat128 | just so she knows what it's all about :) | 12:08 |
hwoarang | TheJulia: mat128: https://bugs.launchpad.net/bifrost/+bug/1660953 | 12:12 |
openstack | Launchpad bug 1660953 in bifrost "transient gate failures on gate-bifrost-integration-tinyipa-ubuntu-xenial" [Undecided,New] | 12:12 |
hwoarang | lets see | 12:12 |
*** jkilpatr_ has joined #openstack-ironic | 12:13 | |
mat128 | hwoarang: thank you | 12:13 |
openstackgerrit | Oksana Voshchana proposed openstack/ironic: WIP: TEST_GRENADE_IRONIC https://review.openstack.org/414988 | 12:14 |
*** dprince has joined #openstack-ironic | 12:21 | |
*** athomas has quit IRC | 12:29 | |
*** slagle has joined #openstack-ironic | 12:29 | |
*** aparnav has quit IRC | 12:30 | |
*** athomas has joined #openstack-ironic | 12:37 | |
openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Allow duplicate execution of update node DB api method https://review.openstack.org/427673 | 12:45 |
*** awiddersheim has quit IRC | 12:45 | |
openstackgerrit | Merged openstack/ironic: DevStack: Only install edk2-ovmf on Fedora https://review.openstack.org/427142 | 12:46 |
*** vikrant has quit IRC | 12:46 | |
sambetts | lucasagomes: ^ woop \o/ | 12:46 |
lucasagomes | \o/ | 12:46 |
openstackgerrit | Merged openstack/ironic: Don't override device_owner for tenant network ports https://review.openstack.org/424248 | 12:49 |
*** nmathew has joined #openstack-ironic | 12:51 | |
*** bfournie has quit IRC | 12:51 | |
*** nmathew has quit IRC | 12:52 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: [Devstack] fix waiting resources on subnode https://review.openstack.org/427675 | 12:54 |
*** mtanino has joined #openstack-ironic | 12:56 | |
*** lucasagomes is now known as lucas-hungry | 13:00 | |
openstackgerrit | Markos Chandras proposed openstack/bifrost: roles: bifrost-ironic-install: Always download the latest iPXE files https://review.openstack.org/427682 | 13:00 |
yuriyz | dtantsur hello. Looks After driver composition reform node can use only one vendor interface at the same moment. This is OK for other interfaces. There is no limitation? For example I have a node with IPMI-compatible BMC and Intel Node Manager and want to use methods from both interfaces. I should merge code? Or something easier? | 13:05 |
*** alineb has quit IRC | 13:06 | |
*** bnemec has quit IRC | 13:06 | |
*** salmankhan has quit IRC | 13:06 | |
dtantsur | yuriyz, I think we need IntelNodeManagerVendor inheriting IPMIVendor. then IntelIPMIHardware inheriting IPMIHardware. or something like that, I haven't put much thought in it | 13:07 |
dtantsur | jroll, opinions welcome ^^^ | 13:07 |
*** ccamacho is now known as ccamacho|lunch | 13:07 | |
*** ralonsoh_ has joined #openstack-ironic | 13:07 | |
*** bapalm has quit IRC | 13:07 | |
*** bnemec has joined #openstack-ironic | 13:07 | |
yuriyz | dtantsur this is only an example, node can contain different devices for different vendors with OOB management | 13:08 |
yuriyz | dtantsur this is only an example, node can contain different devices from different vendors with OOB management | 13:08 |
openstackgerrit | Merged openstack/python-ironicclient: Functional tests for port groups in ironicclient https://review.openstack.org/297727 | 13:08 |
*** ralonsoh has quit IRC | 13:09 | |
dtantsur | yuriyz, this may be a case of actually changing a vendor interface via API | 13:10 |
yuriyz | yes this will work | 13:11 |
*** ralonsoh_ is now known as ralonsoh | 13:11 | |
openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient: Add test to create a port with specific port group UUID https://review.openstack.org/411826 | 13:11 |
*** chedda has joined #openstack-ironic | 13:12 | |
*** bapalm has joined #openstack-ironic | 13:12 | |
*** mtanino has quit IRC | 13:13 | |
dtantsur | folks, could you merge https://review.openstack.org/#/c/427643/ please? it will help the packagers (at least RDO) a lot | 13:15 |
patchbot | patch 427643 - ironic - Add sem-ver flag so pbr generates correct version | 13:15 |
*** chedda has quit IRC | 13:17 | |
*** dtantsur is now known as dtantsur|brb | 13:17 | |
*** mgould|afk is now known as mgould | 13:17 | |
*** g3ek has quit IRC | 13:19 | |
*** jaganathan has quit IRC | 13:22 | |
openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Add dynamic interfaces fields to base node notification https://review.openstack.org/427651 | 13:23 |
*** bapalm has quit IRC | 13:24 | |
*** rnoriega has joined #openstack-ironic | 13:25 | |
vdrok | dtantsur|brb: so this sem-ver: feature thing, which number will it bump, major, minor or patch? | 13:25 |
*** lhinds has joined #openstack-ironic | 13:26 | |
*** g3ek has joined #openstack-ironic | 13:28 | |
openstackgerrit | Merged openstack/ironic: Duplicated code in ..api.get_active_driver_dict() https://review.openstack.org/427335 | 13:28 |
TheJulia | good morning | 13:30 |
vdrok | morning TheJulia | 13:31 |
*** vsaienko has joined #openstack-ironic | 13:32 | |
TheJulia | mat128: hwoarang: my thought was that the interface was not up when the facts on the node were collected and thus no interface was available. Unless libvirt's default network config is moving towards v6 only, it is the only thing that I can think of :( | 13:32 |
vdrok | dtantsur|brb: I mean, if it is going to be the version used for ocata, maybe we want to do sem-ver: api-break? | 13:33 |
*** jaganathan has joined #openstack-ironic | 13:34 | |
openstackgerrit | edan david proposed openstack/ironic-inspector: Adding InfiniBand Support https://review.openstack.org/264257 | 13:34 |
openstackgerrit | Sofiia Andriichenko proposed openstack/python-ironicclient: Fix cleanup of resources https://review.openstack.org/424545 | 13:36 |
mat128 | TheJulia: that's another possibility. The errors hwoarang seemed to have were related to host "facts" more than VMs. | 13:42 |
mat128 | TheJulia: like generating the kernel options to be pushed via pxe @ https://github.com/openstack/bifrost/blob/master/playbooks/roles/bifrost-ironic-install/templates/inspector-default-boot-ipxe.j2 | 13:43 |
*** bfournie has joined #openstack-ironic | 13:43 | |
TheJulia | Yeah, what is odd to me is we added an explicit recollection of facts and it is still occurring. :( | 13:43 |
*** salmankhan has joined #openstack-ironic | 13:46 | |
*** dprince has quit IRC | 13:46 | |
mgould | morning TheJulia | 13:46 |
*** makowals_ has joined #openstack-ironic | 13:47 | |
TheJulia | Good morning mgould | 13:47 |
mat128 | hwoarang, TheJulia: Is it possible that network_interface (and thus ans_network_interface) is configured to an interface that isn't up? https://github.com/openstack/bifrost/blob/master/playbooks/roles/ironic-enroll-dynamic/defaults/main.yml | 13:49 |
*** jaganathan has quit IRC | 13:49 | |
*** makowals has quit IRC | 13:49 | |
mat128 | According to an ansible issue (https://github.com/ansible/ansible/issues/4848) the exact errors happens when trying to fetch the ipv4 address of a non-active interface | 13:49 |
*** lucas-hungry is now known as lucasagomes | 13:50 | |
*** jaganathan has joined #openstack-ironic | 13:50 | |
hwoarang | does this host come and go? or is the libvirt network re-created on every run? because you would think that once virbr0 is up it will stay up until something destructive happens to the host | 13:51 |
TheJulia | mat128: So, that may be a possibility if libvirt is doing something slightly funky. What is interesting is if hwoarang failed on the the template you noted, that means that it was able to access the same fact when writing ironic.conf. | 13:51 |
mat128 | O_o | 13:51 |
TheJulia | hwoarang: we tell libvirt to bring up it's default bridge | 13:51 |
sambetts | hwoarang: OpenStack CI jobs are always run in a fresh VM | 13:51 |
hwoarang | hmm | 13:52 |
TheJulia | and what sambetts said, specifically in bifrost's case, it is essentially a bare system. | 13:52 |
sambetts | TheJulia: could it be related too https://review.openstack.org/#/c/420104/ ?? | 13:52 |
patchbot | patch 420104 - bifrost - Fix race in virbr0 startup (MERGED) | 13:52 |
TheJulia | sambetts: we thought it was, and the failure rate seemed to drop some. And then there was https://review.openstack.org/#/c/426702/ yesterday which should collect the interface again | 13:53 |
patchbot | patch 426702 - bifrost - Refresh Ansible facts after installing packages (MERGED) | 13:53 |
*** vsaienko has quit IRC | 13:54 | |
TheJulia | hwoarang: mat128: so the bug markos opened is ironic.conf's write out where it failed. Was there one where it was specifically the ipxe template write out? | 13:56 |
mat128 | TheJulia: no, that was just me finding the same occurences of ['ansible' + ans_network_interface]['ipv4'] | 13:56 |
mat128 | in fact, I couldn't find that in ironic.conf.j2, which was surprising to me | 13:57 |
milan | morning hwoarang, mat128, sambetts, TheJulia! :) | 13:57 |
TheJulia | mat128: oh... OH | 13:58 |
TheJulia | we have a race on fact population then | 13:58 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: [Devstack] fix waiting resources on subnode https://review.openstack.org/427675 | 13:58 |
* mat128 has absolutely no clue how you found that out :) | 13:58 | |
yolanda | hi hwoarang , is that repeating over time, or a transient error? | 13:58 |
yolanda | i've seen this error when there are network config failures mostly... | 13:59 |
TheJulia | the underlying fact is assembled as a default, it must be collecting the fact and the error then, and it is only appearing later on when the variable is accessed | 13:59 |
mat128 | yolanda: he mentioned (repeating that here since he seems away) "observed every now and then in gate-bifrost-integration-tinyipa-ubuntu-xenial" | 13:59 |
openstackgerrit | Dao Cong Tien proposed openstack/python-ironicclient: Extends driver-list, driver-show supporting new hardware types https://review.openstack.org/419274 | 13:59 |
*** amotoki has joined #openstack-ironic | 13:59 | |
mat128 | TheJulia: interesting | 14:00 |
hwoarang | yolanda: i think it's transient. i see it every now and then in this gate job | 14:00 |
TheJulia | https://github.com/openstack/bifrost/blob/299a7259aabd09793a0bd015606bee409e889391/playbooks/roles/bifrost-ironic-install/defaults/main.yml#L233 | 14:00 |
*** alineb has joined #openstack-ironic | 14:00 | |
TheJulia | https://github.com/openstack/bifrost/blob/299a7259aabd09793a0bd015606bee409e889391/playbooks/roles/bifrost-ironic-install/templates/ironic.conf.j2#L32 | 14:01 |
*** ccamacho|lunch is now known as ccamacho | 14:01 | |
*** sasha2 has joined #openstack-ironic | 14:02 | |
*** trown|outtypewww is now known as trown | 14:03 | |
*** jcoufal has joined #openstack-ironic | 14:03 | |
*** salmankhan has quit IRC | 14:03 | |
*** amotoki has quit IRC | 14:08 | |
*** dtantsur|brb is now known as dtantsur | 14:08 | |
lucasagomes | vsaienk0, my project-config-fu is not the greatest, but I tried to add the UEFI job here: https://review.openstack.org/#/c/424576 | 14:09 |
patchbot | patch 424576 - openstack-infra/project-config - Add non-voting UEFI job for ironic | 14:09 |
dtantsur | vdrok, sem-ver: feature bumps the minor version. and unless we actually broke some API, it's the right thing to do | 14:09 |
lucasagomes | maybe we should think of just swaping the pxe_ipmitool-partition with UEFI directly and leave BIOS test for agent_ | 14:09 |
lucasagomes | to condense different scenarios in 1 job | 14:10 |
jroll | vsaienk0: yeah, I'd like to do scenario tests for portgroups, if the community thinks that's a useful thing to do :) | 14:10 |
lucasagomes | wdyt ? | 14:10 |
jroll | dtantsur: I merged the semver thing | 14:11 |
jroll | er, approved | 14:11 |
dtantsur | thanks! | 14:11 |
*** cdearborn has joined #openstack-ironic | 14:12 | |
jroll | dtantsur: on yuriyz's question, I think someone that wanted to combine many of these would do something like: https://gist.github.com/jimrollenhagen/1a8e0c00e91c4c76282f3469506aaca2 | 14:14 |
jroll | or rather that's my instinct | 14:15 |
*** Goneri has joined #openstack-ironic | 14:15 | |
dtantsur | yeah, maybe | 14:15 |
jroll | for folks that want to truly mix a bunch of unrelated things | 14:16 |
TheJulia | they exist! | 14:16 |
jroll | oh I'm sure they do :) | 14:16 |
openstackgerrit | Merged openstack/ironic: Log reason for hardware type registration failure https://review.openstack.org/427389 | 14:17 |
*** srobert has joined #openstack-ironic | 14:17 | |
*** sayantani01 has joined #openstack-ironic | 14:18 | |
dtantsur | jroll, I wonder if switching vendor_interface is what we should recommend now.. | 14:18 |
*** rbartal__ has quit IRC | 14:19 | |
dtantsur | well, maybe not | 14:19 |
*** rbartal has quit IRC | 14:19 | |
*** rbartal_ has quit IRC | 14:19 | |
jroll | dtantsur: well, these recommendations are for driver authors | 14:19 |
*** rloo has joined #openstack-ironic | 14:19 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Allow get detailed VIF information https://review.openstack.org/421544 | 14:25 |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add the timeout parameter to set_power_state() in the fake driver https://review.openstack.org/427736 | 14:31 |
*** crushil has joined #openstack-ironic | 14:32 | |
rloo | hi and good morning all! | 14:33 |
*** jheroux has joined #openstack-ironic | 14:33 | |
rloo | jroll, dtantsur, TheJulia, lucasagomes, anyone else out there: your opinion. If we remove a policy rule (because it isn't used), do we need a reno? https://review.openstack.org/#/c/427654/ | 14:34 |
patchbot | patch 427654 - ironic - Remove deprecated heartbeat policy check | 14:34 |
lucasagomes | rloo, morning | 14:34 |
mgould | morning rloo | 14:34 |
dtantsur | morning rloo | 14:34 |
TheJulia | I'm surprised there is no test code updated there | 14:35 |
dtantsur | hmm | 14:35 |
TheJulia | rloo: I think we need a reno because otherwise this is a silent change | 14:35 |
lucasagomes | rloo, hmmm /me thinks | 14:35 |
rloo | TheJulia: that's what I'm thinking. We don't use it so it doesn't hurt if someone has that policy, but they may as well delete it. Oh wait, maybe we should also mention in the reno what replaced it? | 14:36 |
jroll | reno can't hurt | 14:36 |
jroll | though | 14:36 |
jroll | can we not just remove that endpoint altogether? | 14:36 |
jroll | (did we do that already?) | 14:36 |
* rloo wonders if 'though' is better than 'but' | 14:36 | |
rloo | jroll: already gone the endpoint is | 14:37 |
rloo | jroll: which is why that code would never be used | 14:37 |
jroll | rloo: if the code cannot be reached, does it really need a reno? :) | 14:37 |
TheJulia | jroll: manually maintained policy files? | 14:37 |
rloo | jroll: if someone has a policy file with that rule, they might want to delete it. | 14:37 |
jroll | ok sure, do it | 14:38 |
lucasagomes | rloo, yeah perhaps that's the only useful thing that I can think of for having that reno | 14:38 |
rloo | jroll: also, if they actually used that rule, gad, do we indicate which policy rule they should be using instead. | 14:38 |
lucasagomes | so someone can clean the policy file | 14:38 |
lucasagomes | otherwise it's ¯\_(ツ)_/¯ | 14:38 |
rloo | sigh. which makes me wonder for all new apis, do we also mention the associated policy rule in the reno? | 14:38 |
TheJulia | rloo: I might regret this statement.... That can't hurt.... | 14:38 |
openstackgerrit | Hironori Shiina proposed openstack/python-ironicclient: Add volume_target support https://review.openstack.org/285233 | 14:39 |
openstackgerrit | Hironori Shiina proposed openstack/python-ironicclient: Add OSC commands for volume connector https://review.openstack.org/427053 | 14:39 |
openstackgerrit | Hironori Shiina proposed openstack/python-ironicclient: Add volume_connector support https://review.openstack.org/214786 | 14:39 |
openstackgerrit | Hironori Shiina proposed openstack/python-ironicclient: Add OSC commands for volume target https://review.openstack.org/427738 | 14:39 |
rloo | TheJulia: but then i thought, what about metrics and notifications (noooooo) | 14:39 |
* TheJulia gets a concrete truck and offers to pour concrete in to the rabbit hole so we can't go down it further | 14:39 | |
rloo | TheJulia: my thought. automate generation of that stuff and put it in the api-ref. | 14:40 |
* rloo thinks i just found a good intern project :) | 14:40 | |
TheJulia | rloo: I'm kind of wondering how we would do that. It makes total sense though! | 14:41 |
rloo | TheJulia: dunno but i'm sure there is a way or several ways to do it | 14:41 |
rloo | TheJulia: or to hack it :) | 14:41 |
TheJulia | we also discussed something with policy checking endpoints, and I'm not sure we ever implemented some extra testing for it. | 14:42 |
jroll | TheJulia: I'm fairly certain that landed | 14:42 |
jroll | orrrr maybe not, wth | 14:42 |
*** kjw3 has joined #openstack-ironic | 14:43 | |
openstackgerrit | Markos Chandras proposed openstack/bifrost: scripts: test-bifrost.sh: Use 'unsafe' disk cache for test VMs https://review.openstack.org/427744 | 14:44 |
TheJulia | jroll: I think https://review.openstack.org/#/c/350177/ is what I was thinking of | 14:45 |
patchbot | patch 350177 - ironic - Add test to ensure policy is always authorized (MERGED) | 14:45 |
jroll | TheJulia: aha, I was looking for files matching test.*policy | 14:45 |
jroll | thanks | 14:46 |
TheJulia | no problem :) | 14:47 |
*** baoli has joined #openstack-ironic | 14:47 | |
*** rbudden has joined #openstack-ironic | 14:52 | |
rloo | so to summarize? i think a reno is useful, and if we know what/if there is another policy rule that replaces it, we can mention in the reno too. | 15:00 |
rloo | and the good news is that i didn't need to ask the question (although useful for future) cuz it turns out that rule is still being used, just elsewhere. (this is where better commit msgs would help) :) | 15:01 |
openstackgerrit | Andrey Shestakov proposed openstack/bifrost: Allow enroll-dynamic on remote host https://review.openstack.org/411299 | 15:01 |
openstackgerrit | Andrey Shestakov proposed openstack/bifrost: Allow deploy-dynamic on remote host https://review.openstack.org/411304 | 15:01 |
openstackgerrit | Alexander Chuzhoy proposed openstack/ironic: Adding clean-steps via json string examples https://review.openstack.org/427753 | 15:03 |
*** salmankhan has joined #openstack-ironic | 15:05 | |
*** salmankhan has quit IRC | 15:05 | |
*** salmankhan has joined #openstack-ironic | 15:07 | |
openstackgerrit | Andrey Shestakov proposed openstack/bifrost: Allow deploy-dynamic on remote host https://review.openstack.org/411304 | 15:08 |
openstackgerrit | Andrey Shestakov proposed openstack/bifrost: Add support of collecting logs from baremetal nodes https://review.openstack.org/415503 | 15:11 |
vdrok | good morning rloo | 15:14 |
*** udesale has joined #openstack-ironic | 15:14 | |
openstackgerrit | Galyna Zholtkevych proposed openstack/ironic: ETAGs maintenance for all api node's requests https://review.openstack.org/392213 | 15:14 |
vsaienk0 | jroll, vdrok just mentioned that multinode job is broken since today morning http://logs.openstack.org/75/427675/2/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/3ff2401/console.html#_2017-02-01_14_55_05_875428 | 15:14 |
rloo | hi vdrok | 15:15 |
vdrok | vsaienk0: https://bugs.launchpad.net/ironic/+bug/1661014 | 15:16 |
openstack | Launchpad bug 1661014 in Ironic "Multinode job fails with "Compute host X not found"" [Undecided,New] | 15:16 |
*** salmankhan has quit IRC | 15:16 | |
jroll | vsaienk0: vdrok: this seems related to what dansmith was talking about with discover_hosts, maybe? | 15:16 |
vdrok | jroll: my thought is, it has something to do with cells | 15:16 |
jroll | yeah | 15:16 |
sambetts | I'm hitting a devstack failure locally with regards to cells | 15:17 |
* jroll in a meeting right now | 15:17 | |
jlvillal | Good morning Ironic | 15:18 |
dtantsur | morning jlvillal | 15:18 |
jlvillal | I kind of feel sorry for the Gitlab.com folks today: https://docs.google.com/document/d/1GCK53YDcBWQveod9kfzW-VCxIABGiryG7_z_6jHdVik/pub | 15:18 |
vdrok | jroll: the compute node this X refers to is deleted tho, seems one of our nodes is unusable for some reason. but still, this should not lead to failure | 15:18 |
*** makowals_ has quit IRC | 15:18 | |
*** cyberik has joined #openstack-ironic | 15:18 | |
openstackgerrit | Alexander Chuzhoy proposed openstack/ironic: Adding clean-steps via json string examples https://review.openstack.org/427753 | 15:18 |
jroll | vdrok: I can't think about this until after this meeting, sorry | 15:19 |
vdrok | morning jlvillal | 15:19 |
*** makowals has joined #openstack-ironic | 15:19 | |
vdrok | sure, np | 15:19 |
jlvillal | :) | 15:19 |
*** salmankhan has joined #openstack-ironic | 15:20 | |
*** owalsh is now known as owalsh-brb | 15:22 | |
dtantsur | vsaienk0, you may be interested (or may be not): https://review.openstack.org/427741 | 15:23 |
patchbot | patch 427741 - puppet-ironic - [WIP] Support new driver configuration options | 15:23 |
rloo | dtantsur: wrt https://review.openstack.org/#/c/427643/. Is this something that just borked now, or is this Sem-Ver thing something that should be done when, at the start of a new cycle? | 15:25 |
patchbot | patch 427643 - ironic - Add sem-ver flag so pbr generates correct version | 15:25 |
dtantsur | rloo, probably should have done, but we never cared too much | 15:26 |
dtantsur | rloo, ideally we should mark all commits as such, but it's quite annoying.. | 15:26 |
rloo | dtantsur: worth adding something to http://docs.openstack.org/developer/ironic/dev/releasing.html? | 15:26 |
rloo | dtantsur: what? add to all patches? | 15:26 |
dtantsur | *ideally* | 15:26 |
dtantsur | pbr view on this thingy | 15:26 |
dtantsur | rloo, I planned to update it, but still can't find time. | 15:27 |
*** owalsh-brb has quit IRC | 15:27 | |
rloo | dtantsur: i didn't/don't actually want to read the pbr doc :) ok. thx for the future-update to the releasing docs. | 15:27 |
rloo | dtantsur: i don't (want to) grok it; otherwise i'd update our doc myself. | 15:28 |
*** chedda has joined #openstack-ironic | 15:28 | |
sambetts | vdrok: I'm very confused whats going on with the cells thing, devstack failed with an error regarding "No compute hosts found for cell" or something like that, all the services seems to be running ok, but I try to nova boot and get a failure "<compute host hostname> not mapping to any cell" :/ | 15:30 |
*** lucasagomes is now known as lucas-brb | 15:31 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: [Devstack] fix waiting resources on subnode https://review.openstack.org/427675 | 15:32 |
*** chedda has quit IRC | 15:33 | |
*** owalsh-brb has joined #openstack-ironic | 15:33 | |
*** owalsh-brb is now known as owalsh | 15:33 | |
*** baoli has quit IRC | 15:34 | |
vdrok | sambetts: yeah, seems like cells are turned on by default, you need to run discover_hosts devstack script now. That ^^ "Compute host X not found" seem to happen after that script was run | 15:34 |
*** jaganathan has quit IRC | 15:41 | |
jroll | came up with this super helpful thing: git diff --summary origin/stable/newton...HEAD releasenotes/notes/ | sed -e 's^ create mode [0-9]* ^http://git.openstack.org/cgit/openstack/ironic/tree/^g' | 15:44 |
*** baoli has joined #openstack-ironic | 15:45 | |
jroll | vdrok: I recall something about compute services needing at least one node to be discovered correctly, (which I assume means a node with resources, meaning AVAILABLE state) | 15:45 |
* jroll brb then can help dig in | 15:45 | |
*** sayantani01 has quit IRC | 15:46 | |
*** dsneddon_afk is now known as dsneddon | 15:47 | |
vdrok | jroll: a bit of log of custom multitenant multinode grenade that does upgrade from master to master :) http://logs.openstack.org/26/397126/38/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/5ab459b/logs/grenade.sh.txt.gz#_2017-02-01_13_43_27_495 there it complains about compute node 7 not found | 15:48 |
vdrok | and compute_node with id 7 in that output has been deleted | 15:49 |
vdrok | deleted_at is set | 15:49 |
sambetts | vdrok: is it because Ironic restarts the nova compute service? | 15:49 |
*** rcernin has quit IRC | 15:51 | |
vdrok | sambetts: no, I think n-cpu is working at that moment, it happens only if the node is unusable by nova | 15:51 |
*** Sukhdev has joined #openstack-ironic | 15:51 | |
rloo | jroll, dtantsur: about the release notes, is 6.0.0 on the newton branch? | 15:51 |
vdrok | anyway, that should not be happening I think, if one node is unusable, we need to schedule on other ones, not complain about No. 7 :( | 15:52 |
jroll | rloo: afaik, yes | 15:52 |
* dtantsur does not remember | 15:52 | |
sambetts | vdrok: +1 | 15:52 |
jroll | rloo: "git show 6.0.0" has lots of info :) | 15:52 |
openstackgerrit | Merged openstack/ironic: Add sem-ver flag so pbr generates correct version https://review.openstack.org/427643 | 15:52 |
rloo | dhellmann: maybe you and jroll can discuss... | 15:52 |
openstackgerrit | Merged openstack/ironic: Remove support for driver object periodic tasks https://review.openstack.org/427493 | 15:52 |
rloo | or jroll, we're discussing in #openstack-release | 15:53 |
sambetts | vdrok: I think there is a registration issue somewhere, I've tried running discover_hosts on my local devstack now Ironic has nodes enrolled and availiable and its not fixing it | 15:53 |
jroll | rloo: yes I see now | 15:53 |
rloo | jroll: thx :) | 15:53 |
sambetts | vdrok: but the nova error I'm getting isn't telling me about the right compute, if I do hypervisor-list | 15:54 |
sambetts | vdrok: my compute host has a UUID for a name | 15:54 |
*** mtanino has joined #openstack-ironic | 15:54 | |
*** tuan_luong has joined #openstack-ironic | 15:54 | |
sambetts | vdrok: my compute host has a UUID for a name instead of the host name which is what the error is outputing | 15:54 |
jroll | so I assume we delete compute nodes that no longer have resources, right | 15:55 |
*** trown is now known as trown|brb | 15:55 | |
vdrok | sambetts: hm, so it's different, what I see in gates reports the actual id in the database | 15:55 |
*** Nisha_Agarwal has joined #openstack-ironic | 15:57 | |
*** athomas has quit IRC | 15:58 | |
*** trown|brb is now known as trown | 15:58 | |
*** athomas has joined #openstack-ironic | 15:59 | |
sambetts | vdrok: how are you checking? actually looking at the DB? BTW I just ran nova-manage cell_v2 map_cell_and_hosts and it seems to resolved the issue that I was seeing and my ironic node is deploying but I'm not sure whats happing with my nova-list calls because they showed the instances for a minute and then it went away... | 16:01 |
sambetts | vdrok: and now its back... | 16:03 |
sambetts | vdrok: and now its gone... I think if I hit nova list too many times it kills it... | 16:03 |
*** udesale has quit IRC | 16:03 | |
sambetts | sigh... | 16:03 |
vdrok | sambetts: yeah, that output was just selecting everything from compute_nodes. I don't have this locally, as it's multinode. do you hit this locally on afresh devstack? | 16:03 |
mariojv | JayF: i'm presuming we should add the rescue_interface field to the node in the api/conductor patch right? | 16:03 |
mariojv | JayF: to allow compatibility with dynamic drivers | 16:04 |
sambetts | vdrok: yeah fresh devstack plumbed into real hardware | 16:04 |
JayF | mariojv: yeah, I think so. I honestly havent' thought about all the changes we have to make for that to work | 16:04 |
mariojv | it won't be too hard | 16:04 |
*** dprince has joined #openstack-ironic | 16:04 | |
mariojv | i'm surprised it wasn't included in previous driver composition patches, since the "no-rescue" interface existed | 16:05 |
mariojv | i guess there's no point in including it until an implementation's ready | 16:05 |
vdrok | sambetts: hm, devstack should be doing this one already https://github.com/openstack-dev/devstack/blob/master/lib/nova#L952 | 16:06 |
rloo | mariojv: yup, that's why it wasn't included | 16:06 |
mariojv | that's fair | 16:07 |
jroll | vdrok: I think cells are default now, is the thing | 16:08 |
sambetts | vdrok: it is this call that I just did manually and it seems to have repaired my environment, https://github.com/openstack-dev/devstack/blob/master/lib/nova#L954 | 16:08 |
openstackgerrit | Galyna Zholtkevych proposed openstack/ironic: ETAGs maintenance for all api node's requests https://review.openstack.org/392213 | 16:08 |
sambetts | vdrok: after calling that I did a nova-manage cell_v2 list_cells and a new cell had appeared | 16:09 |
jroll | sambetts: vdrok: I'm wondering if we need to do that again after we bounce n-cpu | 16:09 |
jroll | dan was asking me things that seem relevant yesterday | 16:09 |
jroll | 17:50:57 dansmith | jroll: sdague: so, we really need to be running discover_hosts after compute nodes are in the database, which for ironic happens pretty late compared to other virt drivers | 16:09 |
jroll | 17:51:17 dansmith | jroll: sdague: should I try to wedge in a discover_hosts call into the late stages of ironic devstack stuff? | 16:09 |
jroll | 17:52:28 dansmith | jroll: sdague: or maybe it's futile if people don't provide ironic nodes at the time of a devstack run? I'm not actually sure how that works | 16:09 |
dansmith | jroll: we're already on it, waiting for a run to confirm | 16:10 |
jroll | dansmith: oh, awesome, got a link? | 16:10 |
sambetts | jroll: in my case because real hardware I'm not registering anything in Ironic until post-devstack so that isn't really going to help, is it really not possible to have a nova-compute with no resources registered any more#/ | 16:11 |
*** sayantan_ has joined #openstack-ironic | 16:11 | |
sambetts | ? | 16:11 |
dansmith | jroll: https://review.openstack.org/#/c/427675/3 | 16:11 |
patchbot | patch 427675 - ironic - [Devstack] fix waiting resources on subnode | 16:11 |
*** rpioso has joined #openstack-ironic | 16:11 | |
dansmith | jroll: this depends on a hack patch in devstack, but we'll roll the meat back into this one if it works | 16:11 |
dansmith | I didn't realize you guys were plugin only | 16:11 |
jroll | sambetts: I think you'd just have to run discover_hosts after enrolling these | 16:11 |
dansmith | correct | 16:11 |
jroll | sambetts: we should fix devstack to allow you to enroll at the usual time | 16:12 |
jroll | dansmith: indeed, devstack kicked us out :| | 16:12 |
jroll | we used to be in tree | 16:12 |
dansmith | jroll: tbh, that seems better to me :) | 16:12 |
jroll | eh, it has its ups and downs | 16:12 |
sambetts | seems super glitchy if a compute host can't exist without resources... what happens if I have a network outage and a number of my nodes go into maintenance, will i have to re-run that command everytime? | 16:13 |
jroll | vsaienk0: please let us know when you come up with fixes like that, we've been sitting here trying to figure this out when you all already have :) | 16:14 |
jroll | sambetts: yeah dunno, we should have a conversation about this | 16:15 |
JayF | sambetts: I was worried about the same | 16:15 |
JayF | and even in normal ironic operation | 16:15 |
JayF | I provision all nodes into instances, delete them all, all are cleaning | 16:15 |
JayF | does the compute node unregister itself? | 16:15 |
sambetts | +1 | 16:15 |
sambetts | that seems bad | 16:16 |
JayF | This seems like a ball of bugs for us | 16:16 |
*** Nisha_Agarwal has quit IRC | 16:16 | |
*** Nisha_Agarwal has joined #openstack-ironic | 16:16 | |
JayF | I don't know how much control we have over it, but if ^ that scenario is possible, there's no way we can release Ocata with that bug, right? | 16:16 |
jroll | there's so many things that play into this | 16:17 |
jroll | I will figure it out | 16:17 |
* jroll talking with dan and reading code | 16:17 | |
*** ccamacho has quit IRC | 16:18 | |
vsaienk0 | jroll: I'm just checking if fixes/hacks proposed by dansmith works :) | 16:21 |
jroll | vsaienk0: right, but you knew a fix was being worked on :) | 16:21 |
dansmith | vsaienk0: see my comment on your patch? if this works, that's what you need to add | 16:21 |
*** mjturek has joined #openstack-ironic | 16:22 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: [Devstack] fix waiting resources on subnode https://review.openstack.org/427675 | 16:23 |
*** lucas-brb is now known as lucasagomes | 16:27 | |
*** Syed__ has joined #openstack-ironic | 16:27 | |
vsaienk0 | jroll, dansmith: so what will be with our api tests, where we enroll fake nodes on the fly and do not call hosts discover, it will break nova? | 16:29 |
dansmith | vsaienk0: you enroll fake nodes and try to nova boot on them? | 16:30 |
vsaienk0 | danms seems we are lucky that we do not try to boot anything in nova, and we cleanup all fake nodes before run baremetal_basic_ops... | 16:31 |
*** absubram has joined #openstack-ironic | 16:31 | |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add the timeout parameter to relevant methods in the fake power interface https://review.openstack.org/427736 | 16:31 |
dansmith | vsaienk0: okay | 16:31 |
vsaienk0 | jroll, so what if we will increase concurrency to 2 someday? we definitely need to run api tests in separate job than? | 16:32 |
*** owalsh is now known as owalsh-brb | 16:32 | |
jroll | vsaienk0: yes, that's always been a known issue | 16:33 |
jroll | vsaienk0: or make those api tests work on nodes that are not in available | 16:33 |
vsaienk0 | jroll: but we have api tests that do create nodes... | 16:34 |
jroll | vsaienk0: right, any nodes we create in available state in api tests, will conflict with scenario tests | 16:34 |
vsaienk0 | I mean test to test if we are able to create a node :) | 16:34 |
*** amarao has quit IRC | 16:35 | |
*** absubram has quit IRC | 16:36 | |
jroll | vsaienk0: yeah, another option would be to give the nodes for API tests 1MB of RAM or something that the scheduler won't like | 16:38 |
vsaienk0 | jroll, we have two snmp job at the moment agent and pxe. Since snmp is a power driver what do you think we we replace pxe_snmp by ironic standalone with API tests only and disable api tests on all other jobs? | 16:40 |
jroll | JayF: sambetts: I'm pretty sure we're fine, we don't delete compute node records when we present 0 resources for a thing | 16:41 |
jroll | s/pretty/99%/ | 16:41 |
JayF | jroll: so do we need to update install guide to reflect a new step for deployers or something? | 16:41 |
jroll | there's some edge cases with very small numbers of servers and multiple compute hosts, but /shrug | 16:42 |
jroll | JayF: maybe, seems like a nova install guide patch to me | 16:42 |
jroll | JayF: in kvm land this would be "run any time you add hypervisors", so maybe we can add "or baremetal machines" to that as the safest option | 16:42 |
* jroll doesn't know how docs get written for nova | 16:43 | |
jroll | nor when | 16:43 |
sambetts | yeah, normally I have nova + ironic up and running before I start adding machines to Ironic, so I think we might need to add it somewhere in our guide post-enroll | 16:44 |
jroll | that's fine with me, too | 16:44 |
jroll | redundancy in docs never hurt anyone | 16:44 |
jroll | dansmith: cells v2 is default/required for ocata, yes | 16:45 |
jroll | ? | 16:45 |
dansmith | yes | 16:45 |
sambetts | i think people might be very confused if they install Nova Compute + Ironic and run that command and it rejects the nova compute | 16:45 |
jroll | ok cool, so we can just doc to always do it | 16:45 |
*** fxpester has quit IRC | 16:45 | |
sambetts | need a massive warning that says "AND HAS RESOURCES" | 16:46 |
jroll | well, ops dgaf what "resources" means to nova | 16:46 |
jroll | needs a thing like "if using with ironic, after enrolling your baremetals" | 16:46 |
dansmith | jroll: right now, what's the delay between adding something to ironic and having nova's computes notice, digest, and expose it/ | 16:47 |
jroll | dansmith: $resource_tracker_interval | 16:48 |
dansmith | jroll: 60s by default? | 16:48 |
jroll | dansmith: dunno, I'm not a nova dev :P | 16:48 |
dansmith | jroll: doesn't the hash ring slow that further? | 16:48 |
lucasagomes | sambetts, quick q, do you have a quick link for the ml2 drivers for cisco switches ? | 16:48 |
jroll | I think so yeah | 16:48 |
jroll | dansmith: should be fairly minimal | 16:48 |
jroll | dansmith: though, for large deployments, I've seen up to 5 minutes for resource tracker interval | 16:49 |
sambetts | lucasagomes: our repo is here: https://github.com/openstack/networking-cisco we have several different drivers for different equipment | 16:49 |
dansmith | jroll: so if you have 10 computes and 1000 ironic nodes, each compute draws the full list of ironic nodes every minute? | 16:49 |
jroll | dansmith: ya | 16:49 |
* JayF notes that loop used to do one call per node, back years ago. It was super slow. | 16:49 | |
jroll | dansmith: at that point I'd recommend more like 5 minutes (for a single compute), hash ring should actually speed it up I think, because less db writes | 16:50 |
dansmith | jroll: so in that case, use my scheduler poll loop patch and set it to whatever you want and you're fine right? | 16:50 |
dansmith | jroll: it's doing the same thing effectively | 16:50 |
jroll | dansmith: ya that would work | 16:50 |
*** Marx314 has joined #openstack-ironic | 16:51 | |
lucasagomes | sambetts, cheers! | 16:51 |
*** rama_y has joined #openstack-ironic | 16:52 | |
openstackgerrit | Mario Villaplana proposed openstack/ironic: Agent driver rescue implementation https://review.openstack.org/400437 | 16:54 |
openstackgerrit | Mario Villaplana proposed openstack/ironic: Add API and conductor methods for [un]rescue https://review.openstack.org/350831 | 16:54 |
*** tesseract has quit IRC | 16:56 | |
*** athomas has quit IRC | 16:56 | |
mgould | hi everyone! Is the format of IRONIC_HWINFO_FILE for devstack documented anywhere? I can't find anything :-( | 16:57 |
sambetts | mgould: its in the ironic plugin.sh file | 16:57 |
NobodyCam | Good Morning Ironic'ers :) | 16:58 |
sambetts | mgould: https://github.com/openstack/ironic/blob/master/devstack/lib/ironic#L70 | 16:58 |
vdrok | morning mgould and NobodyCam | 16:58 |
mgould | morning vdrok | 16:58 |
mgould | sambetts: excellent, thanks | 16:58 |
NobodyCam | Morning vdrok :) | 16:59 |
dtantsur | morning NobodyCam | 16:59 |
mgould | sambetts: but you're not using that to enroll hardware in your devstack+BM setup? | 17:00 |
NobodyCam | Good morning dtantsur :) | 17:00 |
tuan_luong | morning Ironic | 17:00 |
NobodyCam | morning tuan_luong | 17:00 |
*** kjw3 has quit IRC | 17:00 | |
mgould | morning NobodyCam tuan_luong | 17:00 |
NobodyCam | morning mgould :) | 17:00 |
tuan_luong | Morning mgould NobodyCam dtantsur sambetts vdrok | 17:01 |
vdrok | morning tuan_luong | 17:01 |
lucasagomes | NobodyCam, hi there | 17:01 |
sambetts | mgould: no, I'm using auto discovery to enroll my hardware | 17:01 |
NobodyCam | Morning lucasagomes :) | 17:01 |
mgould | OK, thanks | 17:01 |
*** yolanda has quit IRC | 17:01 | |
* mgould is currently trying to get a devstack+BM setup working | 17:01 | |
sambetts | mgould: /me is lazy and doesn't want have to find out alllllll the mac addresses etc to put in the file ;) | 17:02 |
dtantsur | sambetts, oh, this is interesting to know | 17:02 |
dtantsur | sambetts, how do you deal with credentials? | 17:02 |
jroll | root:calvin, duh | 17:03 |
sambetts | hahaha | 17:03 |
jroll | :) | 17:03 |
dtantsur | jroll, one of the primary use cases for auto discovery, yes :) | 17:03 |
dtantsur | but dunno, maybe some CMDB integration | 17:03 |
sambetts | dtantsur: currently remote powering them on in order, and as they appear in Ironic updating their creds, as this is a small test bed I can do that | 17:05 |
dtantsur | ah, so manually. I see. | 17:05 |
* dtantsur still remembers the idea (I think by NobodyCam) to use IPMI's cypher 0 for auto-discovery :) | 17:06 | |
mgould | how do you do auto-discovery? | 17:07 |
sambetts | dtantsur: yeah, I've been doing some thinking about how to use the rules etc to populate that info, if I could discover the mgnt address from inside the node it would be easy, but that doesn't get exposed on my hardware :/ | 17:07 |
dtantsur | mgould, you mean, in general? | 17:07 |
dtantsur | sambetts, oh, so no IPMI address field for you? | 17:08 |
sambetts | dtantsur: nope :( | 17:08 |
*** e0ne has quit IRC | 17:08 | |
dtantsur | mmmm | 17:08 |
NobodyCam | hehehehe :p | 17:08 |
sambetts | dtantsur: I have to rely on mac address for Ironic Node to Inspected node mapping :( | 17:08 |
mgould | dtantsur: I guess? | 17:09 |
dtantsur | :( | 17:09 |
dtantsur | mgould, http://docs.openstack.org/developer/ironic-inspector/usage.html#discovery | 17:09 |
mgould | thanks! | 17:09 |
* jroll lunches | 17:09 | |
*** tuan_luong has quit IRC | 17:10 | |
mgould | OK, so I *could* create an ironic_hwinfo_file with all the mac addresses I care about, *or* I could set up autodiscovery | 17:10 |
sambetts | mgould: +1 | 17:10 |
mgould | since n = 2, the ironic_hwinfo_file route seems easy enough | 17:10 |
dtantsur | yep | 17:10 |
mgould | awesome | 17:11 |
* mgould shall do that | 17:11 | |
openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Pop chassis_uuid field when it is not specified in node show https://review.openstack.org/427815 | 17:11 |
dtantsur | I wonder if we should have an example introspection rules file shipped for this root:calvin case :) | 17:11 |
mgould | dtantsur: +1 | 17:12 |
dtantsur | mgould, I'll do it when I get a chance to play without Dell lab | 17:12 |
* mgould will try to relinquish it soon :-) | 17:13 | |
*** vsaienko has joined #openstack-ironic | 17:13 | |
dtantsur | no hurry, this is not critical | 17:16 |
vsaienko | dansmith is it ok found 0 cells? http://logs.openstack.org/75/427675/4/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/23ad084/logs/devstacklog.txt.gz#_2017-02-01_16_48_46_016 | 17:16 |
dansmith | vsaienko: no | 17:17 |
openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: Find a node by multiple attributes https://review.openstack.org/421285 | 17:19 |
*** owalsh-brb is now known as owalsh | 17:20 | |
*** aNuposic has joined #openstack-ironic | 17:20 | |
milan | dtantsur, seems it's possible to workaround missing naming constraints w/ foreign keys ^ ; see https://review.openstack.org/#/c/421285/8/ironic_inspector/migrations/versions/882b2d84cb1b_attribute_constraints_relaxing.py | 17:20 |
patchbot | patch 421285 - ironic-inspector - Find a node by multiple attributes | 17:20 |
dtantsur | ok, will check | 17:21 |
*** athomas has joined #openstack-ironic | 17:22 | |
*** athomas has quit IRC | 17:25 | |
*** trown is now known as trown|lunch | 17:29 | |
*** chedda has joined #openstack-ironic | 17:29 | |
*** chedda has quit IRC | 17:34 | |
* milan relocates bbl | 17:34 | |
*** milan has quit IRC | 17:34 | |
vsaienko | dansmith how this is possible? also on subnode seems some parameters are missed in config http://logs.openstack.org/75/427675/4/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/23ad084/logs/subnode-2/devstacklog.txt.gz#_2017-02-01_16_56_45_674 | 17:36 |
openstackgerrit | Ruby Loo proposed openstack/ironic: Use delay configoption for ssh.SSHPower drivers https://review.openstack.org/427829 | 17:37 |
*** mgoddard_ has quit IRC | 17:37 | |
*** rpioso has quit IRC | 17:37 | |
dansmith | vsaienko: it means you're not running simple_cell_setup for whatever reason | 17:38 |
vsaienko | dansmith n-cell is not enabled on any ironic job, it explains why discover hosts return 0 cells | 17:40 |
dansmith | vsaienko: n-cell has nothing to do with it | 17:40 |
dansmith | vsaienko: n-cell is cellsv1, not cellsv2 | 17:40 |
*** Sukhdev has quit IRC | 17:41 | |
dtantsur | vsaienko, what was the reason to set https://bugs.launchpad.net/ironic/+bug/1618754 to Fix Committed? We don't use this status. | 17:42 |
openstack | Launchpad bug 1618754 in Ironic "[RFE] Add portgroups support to Ironic" [Wishlist,Fix committed] - Assigned to Vasyl Saienko (vsaienko) | 17:42 |
*** jkilpatr_ has quit IRC | 17:43 | |
vsaienko | dtantsur: sorry I didn't know that, changed to fix released | 17:44 |
dtantsur | thanks! | 17:44 |
rloo | vsaienko: how can that be fixed? i thought there were docs needed. and this: https://review.openstack.org/#/c/382476/ | 17:47 |
patchbot | patch 382476 - ironic - Tempest tests for portgroups | 17:47 |
vsaienko | dansmith, so should we need to call simple_cell_setup on each node where n-cpu is enabled, on only at the primary node? | 17:48 |
*** dprince has quit IRC | 17:49 | |
dansmith | vsaienko: only primary, and it can be before nodes are registered, as long as you call discover later | 17:49 |
dansmith | vsaienko: but, base devstack should be doing it for you when nova runs | 17:49 |
dansmith | vsaienko: is your ironic plugin running before or after nova? | 17:49 |
openstackgerrit | Merged openstack/bifrost: Remove local actions for ssh keys https://review.openstack.org/426188 | 17:51 |
vsaienko | dansmith, oh seems found we call discover_hosts prior to do simple_cells_setup... | 17:51 |
dansmith | vsaienko: well, then that's yer problem :) | 17:52 |
*** cyberik has quit IRC | 17:52 | |
*** jkilpatr has joined #openstack-ironic | 17:53 | |
vsaienko | dansmith, if we change the order, will it fix http://logs.openstack.org/75/427675/4/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/23ad084/logs/subnode-2/devstacklog.txt.gz#_2017-02-01_16_56_45_674 discover_hosts call on subnode? | 17:53 |
dansmith | vsaienko: you do not need to call discover_hosts on the subnode | 17:53 |
vsaienko | dansmith, but we adding more ironic nodes on subnode | 17:54 |
dansmith | vsaienko: it still doesn't need to run on the subnode | 17:54 |
dansmith | unless you just want it there for sequencing or something | 17:54 |
dansmith | but that would be kinda weird, and last I checked, the subnode doesn't have nova db creds | 17:54 |
dansmith | unless yours is different somehow | 17:54 |
vsaienko | dansmith, as I understood we should run discover_hosts each time we added a new node no? | 17:54 |
dansmith | vsaienko: yes, but it doesn't need to be run on the host that has the node or anything | 17:55 |
dansmith | vsaienko: also, you can just run it at the end of adding all the nodes, unless you need the nodes discoverable the minute you add them | 17:55 |
jlvillal | rloo, I would vote no, if it wasn't used and had no effect. | 17:55 |
vsaienko | dansmith, so we should do discover_hosts call in devstack-gate when both primary and subnode are ready | 17:56 |
vsaienko | dansmith, adding it to our devstack plugin will not help | 17:56 |
dansmith | vsaienko: d-g deals with the high level nodes, not the subnodes as they're being setup.. it will be too early from d-g, unless you mean at the very end | 17:57 |
*** aarefiev is now known as aarefiev_afk | 17:57 | |
jlvillal | rloo, I would vote no, if it wasn't used and had no effect. Because it sounds like it has no impact on the operator/user. | 17:58 |
*** owalsh is now known as owalsh-afk | 17:58 | |
jlvillal | rloo, Sorry, I was reading back-scroll and didn't realize how far I was when I replied :) | 17:59 |
vsaienko | dansmith, it seems discover_hosts already done https://github.com/openstack-infra/devstack-gate/blob/f7c2d925f98377ba5c762b9255dc052289cd60ad/devstack-vm-gate.sh#L703-L713 | 17:59 |
*** derekh has quit IRC | 18:01 | |
*** sdake_ has joined #openstack-ironic | 18:01 | |
vdrok | so maybe we can just wait for hypervisors and that would be enough | 18:01 |
vdrok | on the subnode | 18:01 |
vsaienko | dansmith, so for nova multinode jobs you have to run discover_hosts when both primary and subnode with n-cpu are ready, and looks like the place mentioned above in devstack gate | 18:02 |
openstackgerrit | Merged openstack/ironic: Remove deprecated [deploy]erase_devices_iterations https://review.openstack.org/427500 | 18:02 |
vsaienko | vdrok no it doesn't help | 18:02 |
*** chedda has joined #openstack-ironic | 18:02 | |
vdrok | vsaienko: do you have a log of that? | 18:02 |
*** sdake__ has joined #openstack-ironic | 18:03 | |
*** ralonsoh has quit IRC | 18:04 | |
vsaienko | vdrok http://logs.openstack.org/75/427675/2/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/3ff2401/logs/subnode-2/devstacklog.txt.gz#_2017-02-01_14_32_00_011 | 18:04 |
*** sdake has quit IRC | 18:04 | |
vdrok | thx | 18:04 |
vsaienko | vdrok so we checked that all 14 nodes are in nova (7 on primary, 7 on subnode) but the job still failed | 18:05 |
*** harlowja_ has joined #openstack-ironic | 18:05 | |
*** sdake_ has quit IRC | 18:06 | |
vsaienko | vdrok, dansmith discover_hosts was called when all 14 nodes were registered and nova hypervisor statistic is up to date http://logs.openstack.org/75/427675/2/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/3ff2401/console.html#_2017-02-01_14_33_09_601690 | 18:06 |
*** harlowja has quit IRC | 18:06 | |
*** lucasagomes is now known as lucas-afk | 18:08 | |
*** owalsh-afk has quit IRC | 18:09 | |
openstackgerrit | Miles Gould proposed openstack/ironic: Allow setting UEFI boot devices in DRAC driver https://review.openstack.org/420107 | 18:09 |
vdrok | vsaienko: so, it seems that that one single node goes out for some time for some reason, and at the moment discover_hosts is called, it does not touch it | 18:09 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Do not change admin_state for tenant port https://review.openstack.org/427603 | 18:10 |
vdrok | vsaienko: a bit of that weird master-master grenade log - http://logs.openstack.org/26/397126/38/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/2c5b72c/logs/grenade.sh.txt.gz#_2017-02-01_17_25_24_444 | 18:10 |
vsaienko | vdrok how it is possible? | 18:10 |
vdrok | no idea yet | 18:11 |
*** owalsh-afk has joined #openstack-ironic | 18:12 | |
dtantsur | going now, see you tomorrow | 18:13 |
jlvillal | dtantsur, Bye | 18:13 |
sambetts | cya dtantsur | 18:13 |
mgould | dtantsur: o/ | 18:13 |
*** parlos has joined #openstack-ironic | 18:13 | |
*** makowals has quit IRC | 18:13 | |
*** dtantsur is now known as dtantsur|afk | 18:14 | |
*** makowals has joined #openstack-ironic | 18:15 | |
*** xavierr has quit IRC | 18:16 | |
vdrok | vsaienko: here http://logs.openstack.org/26/397126/38/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/2c5b72c/logs/grenade.sh.txt.gz#_2017-02-01_17_25_51_359 some records are deleted and then added back | 18:16 |
vdrok | see hypervisor_hostname fields | 18:16 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: [Devstack] fix waiting resources on subnode https://review.openstack.org/427675 | 18:18 |
vdrok | eg node 00d798ef-* was first registered with id 4, then with id 10 | 18:18 |
openstackgerrit | Pavlo Shchelokovskyy proposed openstack/bifrost: Use Ansible to create test VMs https://review.openstack.org/427071 | 18:19 |
*** dprince has joined #openstack-ironic | 18:19 | |
vdrok | vsaienko: so we need to 1. not run discover_hosts until all nodes created on subnode are visible in hypervisor stats, and that resource tracker does not run at the moment? :( | 18:20 |
*** vsaienko has quit IRC | 18:20 | |
*** vsaienko has joined #openstack-ironic | 18:24 | |
openstackgerrit | Chris Krelle proposed openstack/bifrost: Move misplaced releasenote https://review.openstack.org/427852 | 18:26 |
openstackgerrit | Anton Arefiev proposed openstack/ironic-specs: Torrent-based provisioning https://review.openstack.org/311091 | 18:26 |
openstackgerrit | Anton Arefiev proposed openstack/ironic-inspector: Trivial: improve logging in FSM transition https://review.openstack.org/427853 | 18:26 |
dansmith | vdrok: you can run discover_hosts whenever you want | 18:27 |
dansmith | vdrok: you just have to run it at least once after all your hosts show up | 18:27 |
*** trown|lunch is now known as trown | 18:28 | |
vdrok | dansmith: yeah, but it will not map a deleted compute node to a cell? | 18:28 |
*** pcaruana has quit IRC | 18:28 | |
rloo | jlvillal: was that a vote wrt the policy rule I asked about? i agree, no if it has no impact, but that rule was (and as it turns out, still is) being used. So moot point for now, but we may have that occurring in the future. | 18:28 |
dansmith | vdrok: when and where do you end up with a deleted compute node? | 18:28 |
openstackgerrit | Pavlo Shchelokovskyy proposed openstack/bifrost: Consolidate release notes https://review.openstack.org/427854 | 18:29 |
vdrok | dansmith: I think when ironic node becomes unusable, it gets deleted from compute_nodes table, then when it's usuble again, it gets added back. and in the period of time when it is deleted, discover_hosts get run | 18:29 |
dansmith | vdrok: no, jroll and I tracked that down earlier.. they don't get deleted unless they stop being returned from get_available_nodes() in your ironic driver, | 18:30 |
sambetts | vdrok: in my devstack I brought add nodes to Ironic, and then made sure they were avaible in nova, and then ran discover hosts and it didn't fix the issue :/ | 18:30 |
dansmith | vdrok: which should only happen if they go away from ironic, right? | 18:30 |
dansmith | sambetts: did you use --verbose and see if they were found? | 18:31 |
dansmith | sambetts: and had simple_cell_setup been run before that? devstack runs it, but it might not be getting done with your config | 18:31 |
sambetts | dansmith: the output said "One cell mapping found. Skipping cell0 because it has no hosts in it" | 18:32 |
dansmith | sambetts: that means it didn't do the thing | 18:32 |
*** krtaylor has quit IRC | 18:33 | |
*** sayantan_ has quit IRC | 18:34 | |
dansmith | sambetts: how did you validate that the resources were visible in nova before you ran it? | 18:37 |
sambetts | dansmith: nova hypervisor-stats showed a count of 1 and resources matching my enrolled Ironic nodes avaible | 18:38 |
dansmith | sambetts: but it should show >1 for >1 ironic nodes right? | 18:38 |
sambetts | dansmith: I only have one node enrolled right now, I should have s/nodes/node | 18:39 |
dansmith | sambetts: oh okay, and you didn't see discover_hosts "check" or "create" a record for that one? | 18:39 |
*** yolanda has joined #openstack-ironic | 18:39 | |
sambetts | nope | 18:40 |
*** mgould is now known as mgould|afk | 18:40 | |
dansmith | actually, you should've seen another line for the next cell, even before that | 18:40 |
dansmith | sambetts: did simple_cell_setup get run? | 18:40 |
sambetts | I'm not sure :/ I've not activly done anything in my local.conf to disable it | 18:41 |
dansmith | yeah, but ironic isn't in-tree and so I don't know that anyone has validated that it's getting run | 18:41 |
dansmith | sambetts: nova-manage cell_v2 list_cells --verbose | 18:42 |
*** xavierr has joined #openstack-ironic | 18:43 | |
sambetts | dansmith: when I did that there was only cell0 listed | 18:43 |
dansmith | sambetts: then run nova-manage cell_v2 simple_cell_setup and run it again | 18:44 |
sambetts | dansmith: then I started experiementing and ran cells_v2 map_hosts_to_cells and it seems to have fixed it, but nova is being glitchy now, e.g. every other nova list returns empty | 18:45 |
dansmith | map_cell_and_hosts ? | 18:45 |
sambetts | dansmith: yeah | 18:46 |
*** vsaienko has quit IRC | 18:50 | |
*** priteau has quit IRC | 18:52 | |
openstackgerrit | Pavlo Shchelokovskyy proposed openstack/bifrost: Use Ansible to create test VMs https://review.openstack.org/427071 | 18:54 |
*** vsaienko has joined #openstack-ironic | 18:57 | |
openstackgerrit | Nisha Agarwal proposed openstack/ironic: SNMP agent support for OOB inspection for iLO Drivers https://review.openstack.org/372428 | 18:57 |
*** mjura has quit IRC | 18:58 | |
*** yuanying has quit IRC | 19:03 | |
*** sdake__ is now known as sdkae | 19:03 | |
*** jkilpatr_ has joined #openstack-ironic | 19:03 | |
*** sdkae is now known as sdake | 19:04 | |
*** chedda has quit IRC | 19:04 | |
*** vsaienko has quit IRC | 19:04 | |
*** Nisha_Agarwal has quit IRC | 19:05 | |
*** vsaienko has joined #openstack-ironic | 19:05 | |
*** jaosorior has joined #openstack-ironic | 19:06 | |
*** vsaienko has quit IRC | 19:07 | |
*** jkilpatr has quit IRC | 19:07 | |
*** vsaienko has joined #openstack-ironic | 19:09 | |
*** vsaienko has quit IRC | 19:11 | |
*** vsaienko has joined #openstack-ironic | 19:12 | |
*** krtaylor has joined #openstack-ironic | 19:13 | |
*** sayantani01 has joined #openstack-ironic | 19:15 | |
*** vsaienko has quit IRC | 19:17 | |
openstackgerrit | Sam Betts proposed openstack/ironic: Reorganise Cisco UCSM driver code to match its name https://review.openstack.org/426814 | 19:17 |
*** harlowja_ has quit IRC | 19:19 | |
*** salmankhan has quit IRC | 19:20 | |
*** jcoufal_ has joined #openstack-ironic | 19:23 | |
openstackgerrit | Merged openstack/ironic: Fix a few docstring warnings https://review.openstack.org/427502 | 19:24 |
*** krtaylor has quit IRC | 19:24 | |
sambetts | night all | 19:25 |
*** sambetts is now known as sambetts|afk | 19:25 | |
*** jcoufal has quit IRC | 19:26 | |
*** vsaienko has joined #openstack-ironic | 19:26 | |
*** jcoufal has joined #openstack-ironic | 19:28 | |
*** e0ne has joined #openstack-ironic | 19:30 | |
*** krtaylor has joined #openstack-ironic | 19:31 | |
*** jcoufal_ has quit IRC | 19:31 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: [Devstack] fix waiting resources on subnode https://review.openstack.org/427675 | 19:33 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Do not change admin_state for tenant port https://review.openstack.org/427603 | 19:36 |
vdrok | jroll: is it possible that the node is not reported by our virt driver because it's not mapped to it because of https://github.com/openstack/nova/blob/master/nova/virt/ironic/driver.py#L571-L573 ? I just can not think of any other case a record in compute_nodes can get soft deleted and not picked up by discover_hosts :( | 19:39 |
*** salmankhan has joined #openstack-ironic | 19:39 | |
*** Sukhdev has joined #openstack-ironic | 19:44 | |
*** rpioso has joined #openstack-ironic | 19:44 | |
*** ChubYann has joined #openstack-ironic | 19:45 | |
*** parlos has quit IRC | 19:45 | |
*** vsaienko has quit IRC | 19:45 | |
openstackgerrit | Anup Navare proposed openstack/ironic: Allow logical name along with UUID in port creation https://review.openstack.org/405586 | 19:48 |
vdrok | jroll: just was looking at http://logs.openstack.org/75/427675/2/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/3ff2401/logs/screen-n-cpu.txt.gz and | 19:48 |
vdrok | http://logs.openstack.org/75/427675/2/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/3ff2401/logs/subnode-2/screen-n-cpu.txt.gz at the lines saying "Returning N available nodes" | 19:48 |
*** aNuposic has quit IRC | 19:49 | |
*** aNuposic has joined #openstack-ironic | 19:49 | |
vdrok | tho at the time discover_hosts is run (http://logs.openstack.org/75/427675/2/check/gate-tempest-dsvm-ironic-ipa-wholedisk-agent_ipmitool-tinyipa-multinode-ubuntu-xenial-nv/3ff2401/console.html#_2017-02-01_14_33_09_601690) , the total count is 14 as expected | 19:49 |
* vdrok leaves noe | 19:50 | |
vdrok | good night everyone! | 19:50 |
*** Goneri has quit IRC | 19:51 | |
*** mtanino has quit IRC | 19:51 | |
*** e0ne has quit IRC | 19:55 | |
openstackgerrit | Pavlo Shchelokovskyy proposed openstack/bifrost: Allow to create cloud.yaml with many clouds https://review.openstack.org/426717 | 19:56 |
rloo | just wondering, has the ptl election started? I haven't yet gotten email/ballot/whatever-it-is-called. | 19:57 |
openstackgerrit | Pavlo Shchelokovskyy proposed openstack/bifrost: Use Ansible to create test VMs https://review.openstack.org/427071 | 19:59 |
*** mtanino has joined #openstack-ironic | 19:59 | |
*** e0ne has joined #openstack-ironic | 19:59 | |
*** vsaienko has joined #openstack-ironic | 20:00 | |
JayF | rloo: I haven't either. It was supposed to be mailed 2/1 afaict. | 20:01 |
rloo | JayF: ok, as long as you can't vote, i'm fine. ha ha. Will wait. | 20:02 |
jroll | rloo: JayF: patience, just 9 more hours: https://governance.openstack.org/election/ | 20:05 |
rloo | jroll: isn't that cute, there's a clock there... | 20:05 |
jroll | :P | 20:06 |
*** crushil has quit IRC | 20:06 | |
*** crushil has joined #openstack-ironic | 20:06 | |
*** salmankhan1 has joined #openstack-ironic | 20:07 | |
*** salmankhan has quit IRC | 20:08 | |
*** salmankhan1 is now known as salmankhan | 20:08 | |
*** Sukhdev has quit IRC | 20:08 | |
mrda | Morning Ironic | 20:08 |
*** mtanino has quit IRC | 20:09 | |
jroll | \o mrda | 20:09 |
mrda | jroll: o/ | 20:10 |
*** mtanino has joined #openstack-ironic | 20:12 | |
*** salmankhan has quit IRC | 20:14 | |
*** salmankhan has joined #openstack-ironic | 20:16 | |
*** e0ne has quit IRC | 20:17 | |
rloo | vdrok: i know you're away, but i opened a bug and don't understand why a flat network requires a cleaning uuid. don't you just need it *if* cleaning is enabled, or does 'flat' imply that cleaning (and provisioning)? networks should exist? | 20:18 |
rloo | vdrok: https://bugs.launchpad.net/ironic/+bug/1661082 | 20:18 |
openstack | Launchpad bug 1661082 in Ironic "Raise DriverLoadError if using flat network & cleaning" [Undecided,New] | 20:18 |
*** crushil has quit IRC | 20:20 | |
*** crushil has joined #openstack-ironic | 20:20 | |
*** salmankhan has quit IRC | 20:20 | |
*** e0ne has joined #openstack-ironic | 20:21 | |
*** salmankhan has joined #openstack-ironic | 20:22 | |
*** e0ne has quit IRC | 20:28 | |
*** salmankhan has quit IRC | 20:29 | |
rloo | jroll: that release notes issue with ironic may also be an issue with ipa. i'm not that familiar with ipa but Current Series shows more versions than I see in Newton or Mitaka: http://docs.openstack.org/releasenotes/ironic-python-agent/current-series.html. I was wondering when 1.3.0 was released. | 20:31 |
jroll | rloo: well, yeah, given it's a bug in reno it could be happening anywhere | 20:31 |
jroll | rloo: not sure what you want me to do? | 20:31 |
rloo | jroll: fix it! actually, i didn't know if ipa was using semver too, so i wasn't sure. | 20:32 |
jroll | I assume that fix will get in soonish and everything will be happy | 20:32 |
jroll | rloo: everything in openstack uses semver (ish) | 20:32 |
*** e0ne has joined #openstack-ironic | 20:32 | |
rloo | jroll: the ish is the icky part i suspect. | 20:33 |
TheJulia | rloo: I think the intent was to have the cleaning network as a different ip space on the same physical flat network that the node is attached to | 20:33 |
rloo | TheJulia: even if you don't do cleaning? | 20:34 |
jroll | rloo: well it's more... other projects bump major at release boundaries, rather than at "breaking change", however those are usually the same so it's fine | 20:34 |
TheJulia | Good point, it shouldn't be required then | 20:34 |
rloo | TheJulia: ok, so maybe we don't want to do the TODO that is in the code, but we can track it via the bug i opened. thx. | 20:35 |
*** salmankhan has joined #openstack-ironic | 20:35 | |
rloo | TheJulia: I was 'just' going through to see what was deprecated and can be deleted in ocata. not so simple :) | 20:35 |
*** e0ne has quit IRC | 20:37 | |
*** sambetts|afk has quit IRC | 20:37 | |
*** sambetts_ has joined #openstack-ironic | 20:38 | |
openstackgerrit | Pavlo Shchelokovskyy proposed openstack/bifrost: Use Ansible to create test VMs https://review.openstack.org/427071 | 20:39 |
*** harlowja has joined #openstack-ironic | 20:39 | |
*** salmankhan has quit IRC | 20:40 | |
*** dyasny has quit IRC | 20:41 | |
*** e0ne has joined #openstack-ironic | 20:41 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: [Devstack] fix waiting resources on subnode https://review.openstack.org/427675 | 20:44 |
*** e0ne has quit IRC | 20:47 | |
TheJulia | rloo: simple is rarely simple :) | 20:49 |
rloo | TheJulia: :) | 20:49 |
*** dyasny has joined #openstack-ironic | 20:49 | |
*** e0ne has joined #openstack-ironic | 20:51 | |
*** e0ne has quit IRC | 20:55 | |
jlvillal | Lego of my logo! ;) | 20:56 |
rloo | dumb question. When we release ocata for ironic etc, ironic doesn't need to support ipa MItaka, right? | 20:57 |
rloo | jlvillal: ? | 20:57 |
jlvillal | (A joke probably difficult to understand without watching TV commercials for Lego waffles) | 20:57 |
jroll | rloo: correct | 20:57 |
jlvillal | rloo, Just a joke about the logo process | 20:57 |
rloo | jroll: thx | 20:57 |
rloo | jlvillal: ooooooo kaaaaaaaay :) | 20:57 |
*** aNuposic has quit IRC | 20:58 | |
*** Sukhdev has joined #openstack-ironic | 20:59 | |
*** e0ne has joined #openstack-ironic | 21:00 | |
rloo | jlvillal: i don't like the bear's hand gesture but i am afraid to say anything :) | 21:00 |
* rloo is not a heavy metal person either so... | 21:00 | |
Sukhdev | reposting following question (was disconnected from IRC, was not sure it made it or not) | 21:00 |
Sukhdev | Hi Ironic experts, I have a question. I have BM server which I am trying to boot using Ipmitool driver. I can see Ironic conductor power cycles the server, the server gets the correct IP address on the neutron's provisioning network, but, does not perform the tftp boot - instead it goes to the management interface and boots the host. How do I force it to use the image specified in boot command? | 21:00 |
jlvillal | rloo, I doubt we will get a logo that everyone likes :( | 21:01 |
jlvillal | rloo, Oh and it was "Eggo" waffles. | 21:01 |
rloo | jlvillal: was wondering if we could get the bear to hold two sticks instead, then i could pretend they were chop sticks :D | 21:02 |
jlvillal | heh | 21:02 |
rloo | jlvillal: phew. cuz i had no idea what lego had to do with waffles. | 21:02 |
jlvillal | rloo, Right. It was: L'Eggo of my Eggo. | 21:03 |
rloo | jlvillal: ahh. | 21:03 |
TheJulia | Sukhdev: what do you mean by "instead it goes to the management interface and boots the host"? | 21:04 |
Sukhdev | TheJulia : yes | 21:04 |
*** Syed__ has quit IRC | 21:05 | |
Sukhdev | TheJulia : I have two nics - one connects to management n/w and other to neutron provisioning network | 21:05 |
Sukhdev | TheJulia : it gets IP address on the provisioning network correctly, but, boots the existing OS (which is ubuntu) instead of going on the provisioning network and pulling the image specified in the boot command | 21:06 |
TheJulia | Sukhdev: So, Sounds like you need to run it through cleaning to purge the existing OS. Also, sounds like the boot device preference is not being accepted, or the bios is preferring in-bios settings as opposed to what is being sent over ipmi. | 21:07 |
TheJulia | Sukhdev: You may also need to look at the driver_info parameter ipmi_force_boot_device and set it to True | 21:09 |
Sukhdev | TheJulia : ah - I have automated_clean flag (in ironic.conf) set to False. You think if I set to True might do the trick? | 21:09 |
TheJulia | Sukhdev: ideally if the other OS has already been purged, partially. | 21:10 |
*** gmmaha has quit IRC | 21:10 | |
Sukhdev | TheJulia : Here is the node information - http://paste.openstack.org/show/597273/ | 21:11 |
TheJulia | Sukhdev: try the ipmi_force_boot_device option | 21:13 |
Sukhdev | TheJulia : Let me give it a shot - thanks | 21:13 |
Sukhdev | TheJulia : what is single bridging vs. Double bridging - I do not need to mess with it, right? | 21:14 |
JayF | Sukhdev: double bridging is just a special ipmi thing where you can proxy ipmi via ipmi (a bad description, but I think you get the idea) | 21:15 |
JayF | Sukhdev: 99.99% of people don't need it | 21:15 |
Sukhdev | JayF : oh I see - thanks... | 21:16 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: [Devstack] fix waiting resources on subnode https://review.openstack.org/427675 | 21:22 |
*** vsaienko has quit IRC | 21:25 | |
*** srobert_ has joined #openstack-ironic | 21:25 | |
*** vsaienko has joined #openstack-ironic | 21:26 | |
openstackgerrit | Mario Villaplana proposed openstack/ironic: Agent driver rescue implementation https://review.openstack.org/400437 | 21:27 |
openstackgerrit | Mario Villaplana proposed openstack/ironic: Add API and conductor methods for [un]rescue https://review.openstack.org/350831 | 21:27 |
mariojv | soliosg: fixed that docs problem, nice catch ^ | 21:27 |
JayF | mariojv: I just landed a conflicting thing to rescue | 21:28 |
JayF | mariojv: I wonder if we should stop rebasing it until the release is cut | 21:28 |
JayF | "we" /me lols | 21:28 |
mariojv | :'( | 21:28 |
JayF | (mario is doing all the rebasing, to be clear, lol) | 21:28 |
mariojv | maybe | 21:28 |
mariojv | feature freeze is tomorrow right ? | 21:28 |
soliosg | mariojv: nice, it was actually Joanna's suggestion :) | 21:28 |
mariojv | oh, i didn't see the commit message note | 21:29 |
mariojv | will change if / when i have to do another patch set | 21:29 |
*** srobert has quit IRC | 21:29 | |
mariojv | more like "when" now heh | 21:29 |
*** srobert_ has quit IRC | 21:30 | |
Sukhdev | TheJulia : Did not make any difference after adding ipmi_force_boot_device flag - take a look at my node/port config - http://paste.openstack.org/show/597274/ | 21:30 |
soliosg | mariojv: oh, you meant the 1.31 that needed update to 1.32 :) | 21:30 |
joanna | mariojv, soliosg: nope, it was Solio's :) | 21:32 |
Sukhdev | TheJulia : the conductor powers on the BM, I can see on the console that it is acquiring the DHCP, and then next thing I see is OS is up - (i.e. no reboot and no tftp pull) | 21:32 |
JayF | Question: I know OpenStack, in general, doesn't use "Signed-off-by" tags. Should we not merge something just because it has this in the commit though? | 21:32 |
jlvillal | JayF, I wouldn't reject it just for that. It is unnecessary, but not sure it is not allowed. Just redundant as they have agreed to the contributor agreement. | 21:33 |
mariojv | JayF: http://docs.openstack.org/infra/manual/developers.html#using-signed-off-by | 21:33 |
Sukhdev | TheJulia: considering I have two nics connected to two networks - is there anything specific I need to specify that it must use a specific nic to boot from? something of that sort? | 21:33 |
soliosg | joanna, mariojv: yes the version thing was mine, removing the redundant sentence in commit message is Joanna's | 21:33 |
mariojv | it's allowed | 21:33 |
jlvillal | mariojv, Nice reference :) | 21:34 |
*** salmankhan has joined #openstack-ironic | 21:35 | |
JayF | jroll: is there some reason I'm not aware of we haven't landed https://review.openstack.org/#/c/407651/ yet? | 21:36 |
patchbot | patch 407651 - ironic - Allow using resource classes | 21:36 |
JayF | jroll: it has 3x+2 2x+1, and I'll +A if there's no good reason not to | 21:36 |
jroll | JayF: the only reason I'm aware of is that nobody feels like clicking +A | 21:36 |
openstackgerrit | Mario Villaplana proposed openstack/ironic: Add dynamic interfaces fields to base node notification https://review.openstack.org/427651 | 21:36 |
JayF | landing it then | 21:36 |
jroll | sweet | 21:37 |
*** Goneri has joined #openstack-ironic | 21:37 | |
jroll | thanks | 21:37 |
mariojv | JayF: i fixed your release note nit on that notification patch ^ | 21:37 |
JayF | thanks mariojv | 21:38 |
*** aNuposic has joined #openstack-ironic | 21:38 | |
*** trown is now known as trown|outtypewww | 21:39 | |
*** Sukhdev_ has joined #openstack-ironic | 21:41 | |
*** vsaienko has quit IRC | 21:42 | |
*** gmmaha has joined #openstack-ironic | 21:50 | |
openstackgerrit | Bruno Cornec proposed openstack/ironic-specs: Propose Redfish support into Ironic https://review.openstack.org/184653 | 21:51 |
*** e0ne has quit IRC | 21:53 | |
JayF | jroll: dtantsur|afk: a nice bugfix for stable, if you have a sec. already has my +2 and a +1 from lucas | 21:57 |
openstackgerrit | Ruby Loo proposed openstack/ironic: Remove iSCSI deploy support for IPA Mitaka https://review.openstack.org/427941 | 22:02 |
jroll | JayF: link? | 22:02 |
JayF | jroll: oh, lol | 22:02 |
JayF | https://review.openstack.org/#/c/425124/ | 22:03 |
patchbot | patch 425124 - ironic (stable/newton) - Fix take over for ACTIVE nodes in PXEBoot | 22:03 |
JayF | sorry for the pastefail | 22:03 |
jroll | wow, that's a non trivial backport, heh | 22:04 |
jroll | +A | 22:05 |
*** ijw has joined #openstack-ironic | 22:09 | |
*** salmankhan has quit IRC | 22:14 | |
*** ijw has quit IRC | 22:15 | |
rloo | mariojv, JayF: wrt the notifications: https://review.openstack.org/#/c/427651/, i'm not sure how folks use notifications. in the reno, do we need to mention the new versions of the payloads? | 22:15 |
patchbot | patch 427651 - ironic - Add dynamic interfaces fields to base node notific... | 22:15 |
JayF | rloo: I'm not sure, that's a good question. I'll be good w/whatever mariojv suggests since he's consumed notifications like this before | 22:16 |
*** dsneddon is now known as dsneddon_afk | 22:18 | |
TheJulia | Sukhdev: by management, do you mean out of band? | 22:19 |
Sukhdev_ | TheJulia: yes | 22:20 |
*** dsneddon_afk is now known as dsneddon | 22:21 | |
Sukhdev_ | TheJulia: I used management network so that I can get Inot the server for management purposes | 22:22 |
TheJulia | Sukhdev_: if both interfaces are defined as ports, ironic should be creating tftp config for both interfaces. Maybe not all physical interfaces are enabled for network booting? | 22:22 |
mariojv | main use case i know of currently is searchlight | 22:23 |
mariojv | hmm, it might be good to note the payload versions in the reno, actually | 22:23 |
mariojv | although, that could be obvious once the consumer notices the new payload + new version | 22:24 |
JayF | mariojv: wanna update the reno again? | 22:24 |
Sukhdev_ | TheJulia: I have associated only one interface with ironic port. | 22:24 |
JayF | mariojv: and since ruby and I are still around we can land it real quick? | 22:24 |
mariojv | JayF: i'm almost done for the day, but if yuriyz wants to update it feel free | 22:24 |
*** salmankhan has joined #openstack-ironic | 22:24 | |
rloo | mariojv, JayF: i added another comment to that patch, and changed my vote to +1. See what others think. or not. | 22:25 |
mariojv | rloo: s/version/payload version/ and i'm +1 to the reno change you suggested | 22:25 |
TheJulia | Sukhdev_: so the question becomes, is that the interface that the node is attempting to dhcp boot from | 22:26 |
JayF | rloo: I'm OK with the reno you suggested as well | 22:26 |
*** ijw has joined #openstack-ironic | 22:26 | |
rloo | mariojv, JayF: add your comments to that patch then :) | 22:26 |
Sukhdev_ | TheJulia: not sure. From the behavior it seem that it is not | 22:26 |
JayF | rloo: I was going to update the patch, since the three of us agree | 22:26 |
JayF | unless you don't htink I should? | 22:27 |
Sukhdev_ | TheJulia: how can I ensure that? | 22:27 |
rloo | JayF: fine with me, although i don't know if you can +2 then. but I'm fine if you +2 it ;) | 22:27 |
rloo | JayF: 'just' a reno :) | 22:27 |
*** jheroux has quit IRC | 22:27 | |
Sukhdev_ | TheJulia: is this some kind of bios setup ? | 22:28 |
*** dsneddon is now known as dsneddon_afk | 22:29 | |
Sukhdev_ | TheJulia: alternatively, I can disable the management port to see if that makes any difference | 22:30 |
TheJulia | Sukhdev_: it would likely be in the bios, so it may be a good place to check | 22:30 |
Sukhdev_ | TheJulia: how can I verify that ironic is generating correct boot config? | 22:31 |
TheJulia | Sukhdev_: look in /tftpboot or httpboot while the node is in wait callback state | 22:32 |
Sukhdev_ | TheJulia: cool thanks | 22:33 |
openstackgerrit | Merged openstack/ironic: Do not change admin_state for tenant port https://review.openstack.org/427603 | 22:34 |
openstackgerrit | Merged openstack/ironic: Use delay configoption for ssh.SSHPower drivers https://review.openstack.org/427829 | 22:35 |
openstackgerrit | Merged openstack/ironic: Adding clean-steps via json string examples https://review.openstack.org/427753 | 22:36 |
openstackgerrit | Jay Faulkner proposed openstack/ironic: Add dynamic interfaces fields to base node notification https://review.openstack.org/427651 | 22:36 |
openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic-python-agent: Remove switch_port_descr and switch_chassis_descr https://review.openstack.org/427955 | 22:36 |
JayF | rloo: ^ | 22:36 |
openstackgerrit | Jay Faulkner proposed openstack/ironic: Add dynamic interfaces fields to base node notification https://review.openstack.org/427651 | 22:37 |
JayF | rloo: missed something, but ^ that should be good | 22:37 |
*** wrouesnel has quit IRC | 22:38 | |
rloo | JayF: good thing i didn't look at the previous version :) | 22:38 |
*** wrouesnel has joined #openstack-ironic | 22:38 | |
JayF | I always miss little stupid things until I re-review in gerrit | 22:38 |
rloo | JayF: that's fine. good that you re-review! | 22:38 |
*** crushil has quit IRC | 22:39 | |
*** wrouesnel has quit IRC | 22:40 | |
*** wrouesnel has joined #openstack-ironic | 22:40 | |
*** wrouesnel has quit IRC | 22:40 | |
*** Syed__ has joined #openstack-ironic | 22:41 | |
*** wrouesnel has joined #openstack-ironic | 22:41 | |
*** wrouesnel has quit IRC | 22:41 | |
*** wrouesnel has joined #openstack-ironic | 22:41 | |
*** wrouesnel has quit IRC | 22:42 | |
*** wrouesnel has joined #openstack-ironic | 22:42 | |
openstackgerrit | Merged openstack/bifrost: Move misplaced releasenote https://review.openstack.org/427852 | 22:45 |
*** wrouesnel has quit IRC | 22:46 | |
*** wrouesnel has joined #openstack-ironic | 22:46 | |
*** wrouesnel has quit IRC | 22:47 | |
openstackgerrit | Merged openstack/ironic: Allow using resource classes https://review.openstack.org/407651 | 22:47 |
*** wrouesnel has joined #openstack-ironic | 22:47 | |
*** wrouesnel has quit IRC | 22:49 | |
*** pc_m has quit IRC | 22:49 | |
*** wrouesnel has joined #openstack-ironic | 22:49 | |
openstackgerrit | Merged openstack/ironic-python-agent: Limit memory usage when running qemu-img convert https://review.openstack.org/427134 | 22:49 |
*** wrouesnel has quit IRC | 22:49 | |
*** salmankhan has quit IRC | 22:50 | |
*** wrouesnel has joined #openstack-ironic | 22:50 | |
*** pc_m has joined #openstack-ironic | 22:50 | |
*** jcoufal has quit IRC | 22:53 | |
*** wrouesnel has quit IRC | 22:53 | |
*** wrouesnel has joined #openstack-ironic | 22:53 | |
*** g3ek has quit IRC | 22:53 | |
openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Improve conductor driver validation at startup https://review.openstack.org/426213 | 22:54 |
openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Fail conductor startup if invalid defaults exist https://review.openstack.org/427322 | 22:54 |
jroll | whee | 22:54 |
*** wrouesnel has quit IRC | 22:54 | |
*** Goneri has quit IRC | 22:54 | |
*** wrouesnel has joined #openstack-ironic | 22:54 | |
*** baoli has quit IRC | 22:54 | |
*** Goneri has joined #openstack-ironic | 22:55 | |
*** crushil has joined #openstack-ironic | 22:57 | |
*** bfournie has quit IRC | 22:58 | |
*** wrouesnel has quit IRC | 22:58 | |
*** wrouesnel has joined #openstack-ironic | 22:59 | |
*** wrouesnel has quit IRC | 23:00 | |
*** wrouesnel has joined #openstack-ironic | 23:00 | |
openstackgerrit | Merged openstack/ironic: Configure tempest for multitenancy/flat network https://review.openstack.org/408417 | 23:01 |
*** dprince has quit IRC | 23:01 | |
*** g3ek has joined #openstack-ironic | 23:05 | |
openstackgerrit | Ruby Loo proposed openstack/ironic: [devstack] remove deprecated IRONIC_IPMIINFO_FILE https://review.openstack.org/427960 | 23:05 |
*** yuanying has joined #openstack-ironic | 23:07 | |
openstackgerrit | Merged openstack/ironic: Fix take over for ACTIVE nodes in PXEBoot https://review.openstack.org/425124 | 23:08 |
* jroll out for the night, see y'all tomorrow | 23:09 | |
JayF | o/ | 23:11 |
*** pmannidi has joined #openstack-ironic | 23:14 | |
*** jkilpatr_ has quit IRC | 23:17 | |
*** rama_y has quit IRC | 23:19 | |
openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic-inspector: Update documentation to deploy Ironic Inspector with DevStack https://review.openstack.org/410456 | 23:25 |
*** rama_y has joined #openstack-ironic | 23:26 | |
*** wrouesnel has quit IRC | 23:32 | |
*** wrouesnel has joined #openstack-ironic | 23:32 | |
*** wrouesnel has quit IRC | 23:33 | |
*** wrouesnel has joined #openstack-ironic | 23:34 | |
*** wrouesnel has quit IRC | 23:34 | |
*** wrouesnel has joined #openstack-ironic | 23:35 | |
*** zackf has quit IRC | 23:37 | |
*** zackf has joined #openstack-ironic | 23:39 | |
openstackgerrit | Merged openstack/ironic: Remove deprecated heartbeat policy check https://review.openstack.org/427654 | 23:41 |
*** dsneddon_afk is now known as dsneddon | 23:43 | |
*** wrouesnel has quit IRC | 23:45 | |
*** wrouesnel has joined #openstack-ironic | 23:45 | |
*** wrouesnel has quit IRC | 23:46 | |
*** wrouesnel has joined #openstack-ironic | 23:46 | |
*** aNuposic has quit IRC | 23:47 | |
*** wrouesnel has quit IRC | 23:48 | |
*** wrouesnel has joined #openstack-ironic | 23:48 | |
*** wrouesnel has quit IRC | 23:49 | |
*** wrouesnel has joined #openstack-ironic | 23:49 | |
*** awiddersheim has joined #openstack-ironic | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!