| *** pcrews has joined #openstack-ironic | 00:03 | |
| *** igordcard has quit IRC | 00:10 | |
| *** Marga_ has joined #openstack-ironic | 00:11 | |
| *** Marga_ has quit IRC | 00:14 | |
| *** Marga_ has joined #openstack-ironic | 00:15 | |
| *** Marga_ has quit IRC | 00:19 | |
| *** Marga_ has joined #openstack-ironic | 00:21 | |
| *** Marga_ has quit IRC | 00:22 | |
| *** Marga_ has joined #openstack-ironic | 00:22 | |
| *** eghobo has joined #openstack-ironic | 00:24 | |
| *** Marga_ has quit IRC | 00:27 | |
| *** Haomeng|2 has joined #openstack-ironic | 00:28 | |
| *** Haomeng has quit IRC | 00:29 | |
| *** Marga_ has joined #openstack-ironic | 00:32 | |
| *** eghobo has quit IRC | 00:45 | |
| mrda | hi naohirot | 00:50 |
|---|---|---|
| naohirot | mrda: hi good morning :) | 00:50 |
| jroll | \o mrda and naohirot :) | 00:53 |
| naohirot | jroll: good evening | 00:53 |
| *** Marga_ has quit IRC | 00:55 | |
| *** Marga_ has joined #openstack-ironic | 00:56 | |
| *** chenleji has quit IRC | 00:56 | |
| *** chenleji has joined #openstack-ironic | 00:56 | |
| *** eghobo has joined #openstack-ironic | 00:56 | |
| *** eghobo has quit IRC | 00:58 | |
| mrda | hey jroll | 01:06 |
| jroll | hiya mrda, how goes? | 01:06 |
| mrda | good - sorry about the slow progress on logical names. Was on leave last week (before the kids went back to school) | 01:06 |
| mrda | And I had the LCA conference before that. But will get the ironic portion finished off today and a new patch to review. | 01:07 |
| mrda | Then the python-ironicclient | 01:07 |
| jroll | it's all good, I just hadn't heard about it for a while, I asked before I actually looked at the last update :P | 01:07 |
| mrda | np, it's in good shape, just a few niggles | 01:07 |
| jroll | sweet | 01:08 |
| * jroll will review when it's his monday | 01:08 | |
| mrda | :) | 01:08 |
| *** zhenzanz has joined #openstack-ironic | 01:38 | |
| *** Marga_ has quit IRC | 01:41 | |
| *** jerryz has joined #openstack-ironic | 01:53 | |
| *** chenglch has joined #openstack-ironic | 02:09 | |
| *** jiangfei has joined #openstack-ironic | 02:17 | |
| openstackgerrit | Merged openstack/ironic: Drop deprecated namespace for oslo.rootwrap https://review.openstack.org/151851 | 02:17 |
| lintan__ | Hi jiangfei | 02:19 |
| jiangfei | lintan_: hi, we add --instance option on node-show, why not no node-list, it has some reason? | 02:20 |
| lintan__ | jiangfei: Hmm, I am not familiar with this :( You can ask others | 02:23 |
| jiangfei | lintan__: I saw that you submit commit :). | 02:24 |
| jiangfei | lintan_:sorry, it's linggao. | 02:26 |
| lintan__ | jiangfei, :) | 02:26 |
| jiangfei | lintan_: :) | 02:27 |
| *** ramineni has joined #openstack-ironic | 02:43 | |
| *** achanda has joined #openstack-ironic | 02:48 | |
| *** achanda has quit IRC | 02:50 | |
| *** achanda has joined #openstack-ironic | 02:52 | |
| *** achanda has quit IRC | 02:54 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add Node.driver_internal_info https://review.openstack.org/150784 | 03:02 |
| *** achanda has joined #openstack-ironic | 03:05 | |
| *** achanda has quit IRC | 03:09 | |
| *** achanda has joined #openstack-ironic | 03:09 | |
| *** achanda has quit IRC | 03:28 | |
| *** achanda has joined #openstack-ironic | 03:35 | |
| *** achanda has quit IRC | 03:38 | |
| *** pcrews has quit IRC | 03:40 | |
| *** achanda has joined #openstack-ironic | 03:43 | |
| *** eghobo has joined #openstack-ironic | 03:45 | |
| *** eghobo has quit IRC | 03:58 | |
| *** achanda has quit IRC | 04:12 | |
| *** rameshg87 has joined #openstack-ironic | 04:20 | |
| *** eghobo has joined #openstack-ironic | 04:23 | |
| *** achanda has joined #openstack-ironic | 04:25 | |
| *** achanda has quit IRC | 04:27 | |
| *** eghobo has quit IRC | 04:32 | |
| *** spandhe has joined #openstack-ironic | 04:33 | |
| *** eghobo has joined #openstack-ironic | 04:34 | |
| *** anderbubble has joined #openstack-ironic | 04:46 | |
| *** ramineni1 has joined #openstack-ironic | 05:00 | |
| *** ramineni has quit IRC | 05:02 | |
| *** SpamapS_ is now known as SpamapS | 05:09 | |
| *** SpamapS has quit IRC | 05:09 | |
| *** SpamapS has joined #openstack-ironic | 05:09 | |
| *** pensu has joined #openstack-ironic | 05:14 | |
| *** Marga_ has joined #openstack-ironic | 05:26 | |
| *** Marga_ has quit IRC | 05:37 | |
| *** Marga__ has joined #openstack-ironic | 05:37 | |
| *** lazy_prince has quit IRC | 05:46 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add Node.driver_internal_info https://review.openstack.org/150784 | 05:46 |
| openstackgerrit | Tan Lin proposed openstack/ironic: Clean driver_internal_info when changes nodes' driver. https://review.openstack.org/151990 | 05:46 |
| *** rakesh_hs has joined #openstack-ironic | 05:51 | |
| *** killer_prince has joined #openstack-ironic | 05:51 | |
| *** killer_prince is now known as lazy_prince | 05:51 | |
| *** anderbubble has quit IRC | 06:00 | |
| openstackgerrit | jiangfei proposed openstack/python-ironicclient: node-list support get-node-by-instance_id https://review.openstack.org/151864 | 06:00 |
| openstackgerrit | Tan Lin proposed openstack/ironic: Clean driver_internal_info when changes nodes' driver https://review.openstack.org/151990 | 06:01 |
| *** Marga__ has quit IRC | 06:22 | |
| *** Marga_ has joined #openstack-ironic | 06:22 | |
| *** lazy_prince has quit IRC | 06:26 | |
| *** Marga_ has quit IRC | 06:26 | |
| *** ramineni1 has quit IRC | 06:44 | |
| *** killer_prince has joined #openstack-ironic | 06:50 | |
| *** killer_prince is now known as lazy_prince | 06:50 | |
| *** achanda has joined #openstack-ironic | 06:52 | |
| *** Marga_ has joined #openstack-ironic | 06:53 | |
| *** ramineni has joined #openstack-ironic | 06:56 | |
| *** achanda has quit IRC | 06:56 | |
| *** yog_ has joined #openstack-ironic | 06:56 | |
| *** Marga_ has quit IRC | 06:57 | |
| *** jcoufal has joined #openstack-ironic | 07:00 | |
| *** jiangfei has quit IRC | 07:02 | |
| *** jiangfei has joined #openstack-ironic | 07:02 | |
| *** lazy_prince has quit IRC | 07:04 | |
| openstackgerrit | Michael Davies proposed openstack/ironic: Add logical name support to Ironic https://review.openstack.org/141737 | 07:07 |
| *** achanda has joined #openstack-ironic | 07:07 | |
| *** chuckC__ has joined #openstack-ironic | 07:11 | |
| *** jmanko has joined #openstack-ironic | 07:11 | |
| *** chuckC_ has quit IRC | 07:14 | |
| *** chuckC__ is now known as chuckC_ | 07:14 | |
| *** jmank has quit IRC | 07:14 | |
| *** mrda is now known as mrda-away | 07:14 | |
| *** Marga_ has joined #openstack-ironic | 07:23 | |
| *** Marga_ has quit IRC | 07:27 | |
| *** eghobo has quit IRC | 07:31 | |
| *** killer_prince has joined #openstack-ironic | 07:32 | |
| *** killer_prince has quit IRC | 07:32 | |
| *** lazy_prince has joined #openstack-ironic | 07:33 | |
| *** ijw has joined #openstack-ironic | 07:43 | |
| *** lazy_prince has quit IRC | 07:45 | |
| *** jcoufal_ has joined #openstack-ironic | 07:49 | |
| *** jcoufal has quit IRC | 07:52 | |
| *** lazy_prince has joined #openstack-ironic | 07:58 | |
| *** lazy_prince has quit IRC | 07:58 | |
| *** lazy_prince has joined #openstack-ironic | 07:59 | |
| *** chlong has quit IRC | 08:00 | |
| *** mkerrin has quit IRC | 08:00 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add iRMC Virtual Media Deploy module for iRMC Driver https://review.openstack.org/151958 | 08:01 |
| *** dtantsur|afk is now known as dtantsur | 08:05 | |
| *** spandhe has quit IRC | 08:06 | |
| dtantsur | Morning Ironic | 08:06 |
| naohirot | dtantsur: good morning :) | 08:07 |
| dtantsur | naohirot, g'afternoon (right?) | 08:07 |
| naohirot | dtantsur: Yeah, getting dark, 5PM | 08:08 |
| *** stendulker has joined #openstack-ironic | 08:14 | |
| *** achanda has quit IRC | 08:16 | |
| *** ndipanov has joined #openstack-ironic | 08:17 | |
| lintan__ | hi dtantsur | 08:17 |
| dtantsur | o/ | 08:18 |
| openstackgerrit | jiangfei proposed openstack/python-ironicclient: node-list support get nodes from chassis_uuid https://review.openstack.org/151859 | 08:18 |
| lintan__ | I try updates['driver_internal_info'] = {} at first, but node.driver_internal_info doesn't change immediately | 08:19 |
| lintan__ | dtantsur, so I change node.driver_internal_info and get updates again | 08:21 |
| dtantsur | lintan__, ok, looks valid | 08:22 |
| * dtantsur changes vote | 08:22 | |
| *** Marga_ has joined #openstack-ironic | 08:24 | |
| *** Marga_ has quit IRC | 08:26 | |
| *** Marga_ has joined #openstack-ironic | 08:26 | |
| *** achanda has joined #openstack-ironic | 08:30 | |
| *** Marga_ has quit IRC | 08:31 | |
| *** bauwser is now known as bauzas | 08:32 | |
| openstackgerrit | jiangfei proposed openstack/python-ironicclient: port-list support get-port-from-mac https://review.openstack.org/151868 | 08:42 |
| *** achanda has quit IRC | 08:44 | |
| *** rwsu has joined #openstack-ironic | 08:49 | |
| *** rwsu is now known as rwsu-afk | 08:50 | |
| *** jistr has joined #openstack-ironic | 08:58 | |
| *** erwan_taf has joined #openstack-ironic | 08:59 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Add support for driver-specific periodic tasks https://review.openstack.org/147523 | 09:06 |
| openstackgerrit | jiangfei proposed openstack/python-ironicclient: node-list support get nodes from chassis_uuid https://review.openstack.org/151859 | 09:12 |
| *** derekh has joined #openstack-ironic | 09:12 | |
| *** zhenzanz has quit IRC | 09:13 | |
| *** pensu has quit IRC | 09:13 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Update agent driver with new field driver_internal_info https://review.openstack.org/152055 | 09:17 |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: Create a new python entrypoint for testing drivers https://review.openstack.org/152056 | 09:18 |
| *** MattMan has joined #openstack-ironic | 09:19 | |
| *** andreykurilin_ has joined #openstack-ironic | 09:21 | |
| *** GheRiver1 is now known as GheRivero | 09:22 | |
| *** athomas has joined #openstack-ironic | 09:25 | |
| *** Marga_ has joined #openstack-ironic | 09:27 | |
| *** Marga__ has joined #openstack-ironic | 09:29 | |
| openstackgerrit | sandhya proposed openstack/ironic-specs: Chassis Level Discovery Tool https://review.openstack.org/134866 | 09:29 |
| *** subscope has joined #openstack-ironic | 09:29 | |
| openstackgerrit | sandhya proposed openstack/ironic-specs: Chassis Level Discovery Tool https://review.openstack.org/134866 | 09:31 |
| *** Marga_ has quit IRC | 09:32 | |
| dtantsur | jroll, please have a look at https://review.openstack.org/#/c/152055/ (agent driver change) once you're here | 09:32 |
| *** Marga__ has quit IRC | 09:33 | |
| *** pensu has joined #openstack-ironic | 09:33 | |
| *** ijw has quit IRC | 09:37 | |
| *** andreykurilin_ has quit IRC | 09:39 | |
| *** andreykurilin_ has joined #openstack-ironic | 09:39 | |
| *** yog_ has quit IRC | 09:40 | |
| *** pensu has quit IRC | 09:49 | |
| *** mkerrin has joined #openstack-ironic | 09:54 | |
| *** yog_ has joined #openstack-ironic | 09:56 | |
| *** pensu has joined #openstack-ironic | 10:00 | |
| *** pensu has quit IRC | 10:04 | |
| rameshg87 | dtantsur, hi | 10:14 |
| *** yog_ has quit IRC | 10:15 | |
| rameshg87 | dtantsur, regarding https://review.openstack.org/#/c/151851/1 | 10:15 |
| rameshg87 | dtantsur, shouldn't we actually change requirements.txt to oslo.rootwrap>=1.5.0 ? | 10:15 |
| rameshg87 | dtantsur, oslo_rootwrap seem to be available only from oslo.rootwrap==1.5.0 | 10:16 |
| *** pelix has joined #openstack-ironic | 10:16 | |
| dtantsur | rameshg87, we probably should, but we can't change this file directly | 10:17 |
| dtantsur | rameshg87, it should go to global-requirements | 10:17 |
| dtantsur | https://github.com/openstack/requirements/blob/master/global-requirements.txt | 10:18 |
| *** yog_ has joined #openstack-ironic | 10:18 | |
| dtantsur | but yeah, maybe we landed that patch a bit too early... | 10:18 |
| openstackgerrit | sandhya proposed openstack/ironic-specs: Chassis Level Discovery Tool https://review.openstack.org/134866 | 10:18 |
| rameshg87 | dtantsur, yeah | 10:19 |
| rameshg87 | dtantsur, if someone does devstack now, i guess it won't pick up the latest oslo.rootwrap if they had oslo.rootwrap=1.3.0 installed | 10:20 |
| rameshg87 | dtantsur, and starting conductor might fail | 10:20 |
| dtantsur | we might want to check with infra about bumping global-requirements... and I don't if we should revert this one | 10:21 |
| dtantsur | sorry, brb | 10:21 |
| rameshg87 | dtantsur, have added a comment in that review | 10:22 |
| rameshg87 | dtantsur, infra is still at 1.3.0: https://github.com/openstack/requirements/blob/master/global-requirements.txt#L66 | 10:23 |
| rameshg87 | dtantsur, might need to change that first | 10:23 |
| *** yog_ has quit IRC | 10:23 | |
| *** andreykurilin_ has quit IRC | 10:24 | |
| openstackgerrit | sandhya proposed openstack/ironic-specs: Chassis Level Discovery Tool https://review.openstack.org/134866 | 10:26 |
| *** Marga_ has joined #openstack-ironic | 10:29 | |
| *** Marga_ has quit IRC | 10:35 | |
| *** yog_ has joined #openstack-ironic | 10:36 | |
| *** dtantsur is now known as dtantsur|brb | 10:48 | |
| *** dlpartain has joined #openstack-ironic | 11:01 | |
| *** dlpartain has left #openstack-ironic | 11:01 | |
| *** ramineni has quit IRC | 11:05 | |
| openstackgerrit | Michael Davies proposed openstack/ironic: Add logical name support to Ironic https://review.openstack.org/141737 | 11:10 |
| * mrda-away says to himself, "This time for sure!" | 11:12 | |
| *** kbyrne has joined #openstack-ironic | 11:14 | |
| *** rameshg87 has quit IRC | 11:22 | |
| *** stendulker has quit IRC | 11:30 | |
| *** Marga_ has joined #openstack-ironic | 11:30 | |
| *** Marga_ has quit IRC | 11:35 | |
| *** dtantsur|brb is now known as dtantsur | 12:00 | |
| *** Haomeng has joined #openstack-ironic | 12:01 | |
| *** Haomeng|2 has quit IRC | 12:03 | |
| *** mrmoje has joined #openstack-ironic | 12:11 | |
| mrmoje | join #openstack-oslo | 12:12 |
| *** pensu has joined #openstack-ironic | 12:17 | |
| *** subscope has quit IRC | 12:21 | |
| *** Marga_ has joined #openstack-ironic | 12:23 | |
| *** Marga_ has quit IRC | 12:24 | |
| *** Marga_ has joined #openstack-ironic | 12:24 | |
| *** subscope has joined #openstack-ironic | 12:31 | |
| *** mlsoule has joined #openstack-ironic | 12:46 | |
| *** dprince has joined #openstack-ironic | 13:01 | |
| *** Marga_ has quit IRC | 13:04 | |
| *** Marga_ has joined #openstack-ironic | 13:05 | |
| *** rakesh_hs has quit IRC | 13:22 | |
| *** mjturek has joined #openstack-ironic | 13:32 | |
| *** rloo has joined #openstack-ironic | 13:43 | |
| *** stendulker has joined #openstack-ironic | 13:51 | |
| *** jjohnson2 has joined #openstack-ironic | 14:00 | |
| jroll | morning all :) | 14:02 |
| Shrews | jroll: morning | 14:02 |
| jroll | nice catch by ramesh... dtantsur do we want to revert that? I see global reqs got bumped | 14:03 |
| jroll | hiya Shrews :) | 14:03 |
| dtantsur | Shrews, jroll, o/ | 14:04 |
| dtantsur | if global-reqs are ok, we should just wait for the bot to come IMO | 14:04 |
| Shrews | hey dtantsur | 14:05 |
| jroll | yeah, agree | 14:05 |
| Shrews | jroll: isn't it, like, 6am for you??? | 14:05 |
| jroll | ... yes | 14:05 |
| * jroll is a morning person :P | 14:05 | |
| Shrews | jroll: morning people are weird | 14:05 |
| jroll | dtantsur: that agent patch has $heartbeat_period seconds of backwards incompatibility :( | 14:06 |
| jroll | Shrews: indeed | 14:06 |
| derekh | Hi all, tripleo overcloud on F20 is broken , I think its an ironic regression https://bugs.launchpad.net/tripleo/+bug/1417026 | 14:06 |
| jroll | derekh: :/ | 14:07 |
| jroll | derekh: you have more info? | 14:07 |
| jroll | one thing we realized, is we landed the change to move oslo.rootwrap to oslo_rootwrap before we picked up the new version | 14:07 |
| jroll | it's in global reqs at this point but I could propose a bump before the bot does to see what happens | 14:08 |
| derekh | jroll: screen show of what I'm seeing when trying to boot a f20 image https://launchpadlibrarian.net/196381334/Screenshot%20from%202015-02-02%2011%3A31%3A25.png | 14:08 |
| derekh | jroll: that made me try a test revert of https://review.openstack.org/#/c/141148/6 | 14:09 |
| *** chenglch has quit IRC | 14:09 | |
| stendulker | Hi jroll | 14:09 |
| jroll | morning stendulker | 14:09 |
| derekh | jrist: which passed, the first F20 overcloud pass we have had since friday | 14:09 |
| stendulker | morning jroll ! | 14:09 |
| stendulker | jroll: Its evening for me. | 14:10 |
| jroll | derekh: strange to me that this commit only broke f20 :/ | 14:10 |
| jroll | stendulker: oops, good evening :) | 14:10 |
| stendulker | jroll: For ILOAgent driver, for secure boot, I need to enable secure boot once deploy is complete. | 14:10 |
| derekh | jroll: yup, I thought so to but that seems to be whats happening, see http://goodsquishy.com/downloads/tripleo-jobs.html | 14:11 |
| jroll | derekh: does the f20 environment have dhcp? | 14:11 |
| stendulker | jroll: For this one option is to override heartbeat method and check for deploy_is_done() and perform the task | 14:11 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Bump oslo.rootwrap to 1.5.0 https://review.openstack.org/152127 | 14:12 |
| jroll | dtantsur: ^ in case we want to be proactive | 14:13 |
| stendulker | jroll: Would it be better to have a interface in AgentVendorInterface which could be redefined by drivers to perform things upon deploy complete or in vendor passthrough method but not to get called regularly like heartbeat() | 14:13 |
| jroll | derekh: hrm, it must have dhcp because pxe | 14:13 |
| derekh | jroll: I thought so but maybe not | 14:13 |
| jroll | stendulker: you could override _reboot_to_instance() no? | 14:14 |
| stendulker | jroll: Any suggestion on this? | 14:15 |
| jroll | stendulker: _reboot_to_instance is called when the deploy is done, just override that method instead https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/agent.py#L406-425 | 14:16 |
| stendulker | jroll: this sounds good | 14:17 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Revert "Do not pass PXE net config from bootloader to ramdisk" https://review.openstack.org/152129 | 14:17 |
| jroll | derekh: I'm not sure if this is the fix we should land, but ^ | 14:18 |
| jroll | derekh: I'd rather fix that environment, might be ok with a revert in the meantime though | 14:18 |
| rloo | Happy Monday ironickers | 14:18 |
| jroll | derekh: what's different about that env? the undercloud runs on fedora 20? | 14:18 |
| jroll | morning rloo, "happy" is a fun word :P | 14:18 |
| rloo | evening stendulker, I was looking at your comments for https://review.openstack.org/#/c/135845/ | 14:19 |
| stendulker | thanks jroll | 14:19 |
| derekh | jroll: the difference between that env for the F20 ci job when compared to the ubuntu one ? | 14:19 |
| rloo | jroll: heh, am still waking up. hoping it'll be a happy week ;) | 14:19 |
| jroll | derekh: correct | 14:19 |
| jroll | rloo: indeed :) | 14:19 |
| derekh | jroll: its a F20 undercloud booting a F20 image , the passing job is a ununtu undercloud booting ubunutu image | 14:20 |
| rloo | stendulker: where in https://review.openstack.org/#/c/135228/3/specs/kilo/uefi-secure-boot.rst is there mention of set_secure_boot_state? (oh, maybe I should do a grep, i was just skimming) | 14:20 |
| jroll | derekh: hrm, everything else is the same? ramdisks are the same? | 14:20 |
| stendulker | rloo: Its in comments | 14:21 |
| rloo | stendulker: oh, let me see | 14:21 |
| derekh | jroll: ramdisk would be built with F20 aswell (as opposed to Ubuntu) | 14:21 |
| jroll | derekh: aha | 14:22 |
| rloo | stendulker: ok, I see the comments now. They were just comments; ie, not proposed. I don't think that those settings are 'states'. | 14:22 |
| rloo | stendulker: they don't reflect the current state of secure_boot, do they? | 14:22 |
| rloo | stendulker: i'll comment in the spec | 14:23 |
| stendulker | rloo: secure boot can have 2 states. viz enabled or disabled | 14:23 |
| rloo | stendulker: i'm objecting to the word 'state'. Can't we use something like 'mode'? | 14:24 |
| rloo | stendulker: just cuz we already use state in provision_state and power_state. | 14:25 |
| jroll | derekh: I'm thinking the f20 ramdisk needs to be fixed somehow, maybe there's a different code path compared to https://review.openstack.org/#/c/140870/ | 14:28 |
| jroll | derekh: I'm fine with reverting that for now, might like to hear from others, adam_g etc | 14:29 |
| *** Haomeng has quit IRC | 14:29 | |
| *** stendulker_ has joined #openstack-ironic | 14:30 | |
| derekh | jroll: ok, makes sense, in the meantime I'll revert it in tripleo-ci | 14:30 |
| derekh | jroll: thanks | 14:30 |
| stendulker_ | rloo: If naming could ause confusion, we can change the names of APIs | 14:30 |
| *** BadCub_ has joined #openstack-ironic | 14:31 | |
| NobodyCam | good morning Ironic :) | 14:31 |
| jroll | derekh: hm, wondering if we should revert in ironic proper, there could be other people using the f20 ramdisk | 14:31 |
| jroll | hiya NobodyCam :) | 14:31 |
| BadCub_ | Morning Iroinic | 14:31 |
| BadCub_ | Ironic even | 14:31 |
| *** zigo has quit IRC | 14:31 | |
| jroll | mornin BadCub_ :) | 14:31 |
| NobodyCam | morning jroll and BadCub_ | 14:31 |
| NobodyCam | :) | 14:31 |
| naohirot | NobodyCam: jroll: BadCub_: good morning | 14:32 |
| GheRivero | morning all | 14:32 |
| naohirot | NobodyCam: I'd like to know one thing before going to bed :) | 14:32 |
| derekh | jroll: yup, my suggestion of reverting in tripleo-ci is only a temp measure to allow our ci to keep moving, while a permanent solution is settled on | 14:32 |
| rloo | stendulker_: that's my opinion. What do you/others think? I've commented in the spec. thx. | 14:33 |
| stendulker_ | rloo: By having get_secure_boot_mode(), one should not assume its yet boot mode in addition to (bios, uefi) | 14:33 |
| *** stendulker has quit IRC | 14:33 | |
| NobodyCam | morning GheRivero naohirot rloo and derekh | 14:33 |
| jroll | derekh: right, ok, thanks | 14:33 |
| jroll | \o GheRivero | 14:33 |
| naohirot | NobodyCam: do we have IRC meeting tomorrow? | 14:33 |
| rloo | morning NobodyCam | 14:33 |
| NobodyCam | I think we do | 14:33 |
| rloo | GheRivero: hi! Have you been away? | 14:33 |
| naohirot | NobodyCam: or there is no IRC meeting due to mid cycle? | 14:33 |
| *** xxj has quit IRC | 14:34 | |
| naohirot | NobodyCam: Okay | 14:34 |
| jroll | I think we should still have a meeting but idk what the actual plan is | 14:34 |
| NobodyCam | naohirot: :) | 14:34 |
| jroll | this week is the tuesday time right? | 14:34 |
| jroll | (tuesday utc) | 14:34 |
| naohirot | jroll: yes, i think So | 14:34 |
| *** zigo has joined #openstack-ironic | 14:35 | |
| GheRivero | rloo: kind of. Focusing on other things | 14:35 |
| rloo | GheRivero: welcome back ;) | 14:35 |
| naohirot | NobodyCam: jroll: all: see you good night! | 14:35 |
| *** naohirot has quit IRC | 14:35 | |
| dtantsur | NobodyCam, rloo, morning | 14:35 |
| NobodyCam | morning dtantsur | 14:36 |
| rloo | stendulker_: maybe some other word than 'mode'? I didn't think much beyond not liking 'state'. | 14:36 |
| stendulker_ | rloo: How about naming them as get_uefi_secure_boot_mode() and set_uefi_Secure_boot_mode()? This would avoid confusion and also associate it with the uefi boot mode | 14:36 |
| *** vipul has quit IRC | 14:37 | |
| rloo | stendulker_: this is in the management interface right? I don't think we want 'uefi' there if setting/getting this secure boot mode could be used non-uefi. | 14:38 |
| stendulker_ | rloo: secure boot is applicable only to UEFI boot mode | 14:38 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Move oslo.config references to oslo_config https://review.openstack.org/151957 | 14:39 |
| *** jjohnson2 has quit IRC | 14:40 | |
| *** Haomeng has joined #openstack-ironic | 14:40 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Move oslo.db references to oslo_db https://review.openstack.org/151955 | 14:40 |
| *** vipul has joined #openstack-ironic | 14:41 | |
| *** Marga_ has quit IRC | 14:42 | |
| rloo | stendulker_: how about instead of true/false, the value is none or 'uefi'? | 14:42 |
| stendulker_ | rloo: UEFI can still work without secure boot | 14:43 |
| rloo | stendulker_: right, so in that case, secure_boot_mode would be none | 14:44 |
| stendulker_ | rloo: In UEFI boot mode, secure boot can be in enabled or disabled state. | 14:44 |
| rloo | stendulker_: do we have any other properties that are only related to uefi or is this the first one? (other than boot mode) | 14:45 |
| dtantsur | so, today http://ironic-bugs.divius.net/ is broken due to unicode problems... one more reason to have Py3k ASAP >_< | 14:45 |
| stendulker_ | rloo: do you mean true=uefi and false=none for secure boot state? | 14:45 |
| *** xxj has joined #openstack-ironic | 14:45 | |
| stendulker_ | rloo: this is the first one that differs from bios | 14:46 |
| rloo | stendulker_: yes. I mean, set_secure_boot_mode() (or whatever we call it), can have values None or 'uefi'. | 14:46 |
| rloo | stendulker_: if that makes sense. I don't have a good idea right now, of how we want to do the api so it is generalized cuz I'm not sure what the future holds. I don't want us to end up with eg get_uefi_secure_boot_mode, get_uefi_propertyZ_mode, get_uefi_properyY_mode... | 14:47 |
| stendulker_ | rloo: I feel, such values (none, uefi) do not give correct interpretation of secure boot mode. | 14:47 |
| rloo | stendulker_: and I don't know if we'll ever have secure_boot in the context of something outside uefi. if we do, I don't know that I want 'uefi' in the name of the method. | 14:48 |
| stendulker_ | rloo: We can drop uefi from the name of the API and it cane be get/set_secure_boot_mode() | 14:50 |
| rloo | stendulker_: that's fine with me but I thought you were concerned about it. | 14:51 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Move oslo.config references to oslo_config https://review.openstack.org/151957 | 14:51 |
| jroll | holy cow, that's a lot of oslo.config imports | 14:51 |
| stendulker_ | rloo: i was trying to be more explicit, but its ok with me not to have uefi | 14:52 |
| *** openstackgerrit has quit IRC | 14:52 | |
| stendulker_ | rloo: Will change the API names and upload the new patch for same. | 14:52 |
| rloo | stendulker_: I'd rather not have uefi in the name. You're going to hate this, but I'm wondering why we even want a method in the management interface for this. | 14:52 |
| jroll | stendulker_: rloo I like the (none, uefi) thing, that would give us explicit-ness | 14:52 |
| *** openstackgerrit has joined #openstack-ironic | 14:52 | |
| jroll | rloo: where would you put it? boot interface? | 14:53 |
| rloo | jroll, NobodyCam, anyone. How are we matching/mapping information in 'capabilities' (that eg nova driver sets) to our nodes? | 14:53 |
| jroll | rloo: it's in nova, let me find it | 14:54 |
| rloo | jroll: looking at the uefi secure boot spec, it seems to indicate that the flavor will have capabilities:secure_boot:True | 14:54 |
| jroll | rloo: I think it's this https://github.com/openstack/nova/blob/master/nova/scheduler/filters/compute_capabilities_filter.py | 14:55 |
| jroll | oh, flavor extra specs, mmm | 14:55 |
| jroll | yeah, that's the flavor filter | 14:56 |
| rloo | jroll: so for each capability that can be set, are we going to have code that sez: if node.capability['secure_boot'] and driver.management.get_secure_boot()... | 14:56 |
| *** jerryz has quit IRC | 14:56 | |
| jroll | rloo: and the mapping here https://github.com/openstack/nova/blob/master/nova/virt/ironic/driver.py#L223-236 | 14:57 |
| jroll | mmm, I see | 14:57 |
| jroll | I think it depends on the capability | 14:58 |
| stendulker_ | jroll: There is a spec that will pass the capabilities as part of instance_info | 14:58 |
| jroll | for example we might have capabilities['awesome_video_card'] | 14:58 |
| stendulker_ | jroll: https://review.openstack.org/#/c/136104/ | 14:58 |
| rloo | what do you do when it is -10 and storming outside and your cat comes in with a (dead I think) bird... go back to work... | 14:58 |
| jroll | which is just a filtering thing | 14:58 |
| jroll | rloo: lollllll | 14:58 |
| jroll | here's the actual patch passing capabilities along https://review.openstack.org/#/c/141012/ | 14:59 |
| rloo | jroll: right, so it will be in the node's instance info. then what? | 15:00 |
| jroll | idk, just narrating :P | 15:01 |
| jroll | I mean, we'll need code to do things with it | 15:01 |
| rloo | jroll: we treat them all individually/specially? I think now, that I didn't agree with a spec that wanted to add some method to mgt interface. | 15:01 |
| jroll | but it's going to depend what the capability is | 15:01 |
| NobodyCam | nice HipChat reset all there passwords :( | 15:01 |
| rloo | jroll: ah yes, get/set boot mode to mgt interface (https://review.openstack.org/#/c/129529/) | 15:01 |
| jroll | NobodyCam: /me wonders what other data was compromised | 15:02 |
| rloo | jroll: yeah, at some point it is going to depend on the capability. | 15:02 |
| NobodyCam | jroll: yep! | 15:02 |
| rloo | jroll: Are we OK with adding a new set of methods get/set_capabilityX() to mgt interface for each capabililty (worst case I guess) | 15:02 |
| jroll | rloo: I'm ok with that for now. if it gets to be too many we can pretty easily make that generic | 15:03 |
| jroll | or make it generic now, idk | 15:03 |
| jroll | set_capability(capability) | 15:03 |
| rloo | jroll: or is it more extensible to make it get/set_capability(capability name, value) | 15:03 |
| NobodyCam | rloo: ++ | 15:03 |
| rloo | I think we should make it generic now. | 15:03 |
| jroll | are we confident there will be many of them? | 15:04 |
| jroll | it's easy to make that generic, set_capability_uefi = functools.partial(set_capability, 'uefi') | 15:05 |
| rloo | jroll: didn't people mention at the summit sessions, that they wanted to pass down capabilities from nova/flavour? | 15:05 |
| jroll | (for backwards compat) | 15:05 |
| jroll | rloo: yeah, this is exactly that | 15:05 |
| rloo | but why bother doing the backwards compatibility if we can do the generic one now. | 15:05 |
| rloo | is there any reason not to have a generic now, and to add set_boot_mode(), set_secure_boot_mode() now? | 15:05 |
| *** yog_ has quit IRC | 15:06 | |
| rloo | the fewer backwards compat stuff we have the better. | 15:06 |
| jroll | yeah, I suppose | 15:10 |
| *** jerryz has joined #openstack-ironic | 15:12 | |
| jroll | rloo: did you see devananda's comments here? https://review.openstack.org/#/c/129529/ | 15:12 |
| victor_lowther | Good morning, Ironic | 15:12 |
| jroll | heya victor_lowther :) | 15:12 |
| *** erwan_taf has quit IRC | 15:13 | |
| rloo | jroll: thx, didn't see that (haven't gone through all my email) | 15:13 |
| jroll | np | 15:14 |
| rloo | jroll: that's fine then. although my comment was dec 8, and it seems a shame that the spec was held up for 2 months+... | 15:15 |
| jroll | rloo: yeah, your comments could have been addressed in that time :P | 15:15 |
| jroll | I think the hold up is actually people reviewing it | 15:16 |
| rloo | jroll: 'people', not reviewing it. that's us :-( | 15:16 |
| jroll | I know | 15:16 |
| jroll | :/ | 15:16 |
| jroll | reviewing is hard | 15:16 |
| rloo | it is hard and to be honest, i have no idea what to comment on any more. | 15:17 |
| jroll | I need to start using that dashboard | 15:17 |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic-specs: Ironic Management Interfaces to support UEFI Secure Boot https://review.openstack.org/135845 | 15:17 |
| stendulker_ | rloo: Have updated the spec by changing the API names as get_secure_boot_mode() and set_secure_boot_mode() | 15:22 |
| rloo | thx stendulker_. will look at it in a few minutes. | 15:22 |
| jroll | devananda: want to unblock https://review.openstack.org/#/c/150784/4 ? | 15:23 |
| stendulker_ | rloo: Thank you for all your comments. They are really valuable. | 15:23 |
| rloo | stendulker_: yw. | 15:23 |
| stendulker_ | Hi NobodyCam | 15:24 |
| stendulker_ | NobodyCam: Good morning :) | 15:25 |
| jroll | dtantsur: https://review.openstack.org/#/c/151686/1/ironic/api/controllers/base.py did you want to actually move MIN_VER and MAX_VER or just the config stuff? | 15:25 |
| dtantsur | jroll, both | 15:26 |
| jroll | dtantsur: it's min/max for v1, seems like it should go there so we don't need to move stuff when we do a v2 | 15:26 |
| dtantsur | so that you don't have circular imports | 15:26 |
| jroll | yeah | 15:26 |
| dtantsur | jroll, not it's not for v1 :) | 15:26 |
| jroll | it is | 15:26 |
| dtantsur | it is in base class and applies to everything | 15:26 |
| jroll | those are min/max minor versions | 15:26 |
| dtantsur | jroll, if it's for v1 it should not go to base.py | 15:27 |
| dtantsur | the way it's implemented now is that it applies to all (1 of 1) version :) | 15:27 |
| jroll | well | 15:27 |
| *** erwan_taf has joined #openstack-ironic | 15:27 | |
| jroll | so base.py handles version things | 15:27 |
| jroll | we could theoretically have a min of 1.3 and max of 2.5 | 15:27 |
| jroll | we need to parse version headers and whatnot whether it's v1 or something else | 15:28 |
| jroll | so that bit *should* be in base.py | 15:28 |
| jroll | idk, it's weird | 15:28 |
| dtantsur | jroll, ok, then let's move MIN_VER to base.py | 15:28 |
| dtantsur | especially if you envision things like "a min of 1.3 and max of 2.5" | 15:29 |
| jroll | it's really MIN_V1_VER | 15:29 |
| jroll | because it's just the minor version that MIN_VER specifies | 15:29 |
| jroll | idk, maybe we should make it total version | 15:29 |
| jroll | yeah, let's do that | 15:29 |
| *** achanda has joined #openstack-ironic | 15:30 | |
| NobodyCam | morning stendulker_ :) | 15:30 |
| *** jerryz has quit IRC | 15:31 | |
| adam_g | derekh, does tripleo ci build a new deploy ramdisk each run? if so, is it using the most recent DIB release? | 15:31 |
| derekh | adam_g: yes and yes | 15:31 |
| jroll | adam_g: just blame fedora :P | 15:31 |
| derekh | adam_g: or at least if it isn't we have a problem | 15:32 |
| stendulker_ | NobodyCam : Fedora element on DIB seems broken with Fedora stopped hosting cloud images at http://cloud.fedoraproject.org | 15:32 |
| adam_g | jroll, it shouldn't matter | 15:32 |
| jroll | adam_g: I know, I was joking | 15:32 |
| jroll | X-OpenStack-Ironic-API-Minimum-Version: 0 | 15:32 |
| derekh | this is a serious channel | 15:32 |
| jroll | X-OpenStack-Ironic-API-Maximum-Version: 1 | 15:32 |
| jroll | dtantsur: ^ I don't love this | 15:32 |
| NobodyCam | stendulker_: humm I'd check in the #tripleo channel | 15:32 |
| stendulker_ | NobodyCam: They are hosting at http://download.fedoraproject.org and image type is raw instead of qcow2 | 15:32 |
| NobodyCam | there may be a patch already for that | 15:33 |
| dtantsur | stendulker_, I'm constantly using fedora element for building ramdisk, what do you mean by broken? | 15:33 |
| stendulker_ | NobodyCam: Ok. I will sheck at #tripleo channel | 15:33 |
| stendulker_ | dtantsur: When I was trying to build new agent ramdisk, it complained. If you try to open http://cloud.fedoraproject.org you get redirected to http://download.fedoraproject.org | 15:34 |
| dtantsur | jroll, well.. honestly I'm a bit afraid that this micro-version stuff is going to complicate things a lot | 15:34 |
| jroll | dtantsur: why? | 15:35 |
| stendulker_ | dtantsur: The URL and image types have changed | 15:35 |
| dtantsur | jroll, too many combinations to keep in mind and test | 15:35 |
| adam_g | derekh, cant seem to find what DIB version is being used here: http://logs.openstack.org/28/137028/11/check-tripleo/check-tripleo-ironic-overcloud-f20-nonha/c82decb/ | 15:35 |
| adam_g | pip-freeze.txt is empty | 15:35 |
| dtantsur | stendulker_, curl/wget should handle the redirect... when did you last try? | 15:35 |
| stendulker_ | dtantsur: Some 6hrs back. | 15:36 |
| jroll | dtantsur: yeah... I think we should move functional API testing stuff to our tree as well, I don't think it will be too bad, just need more context in the head | 15:36 |
| dtantsur | stendulker_, hmmm... strange, I've build a ramdisk with 'fedora ironic-discoverd-ramdisk' like 4hrs ago... | 15:37 |
| adam_g | derekh, i dont suppose you have some way of pulling a ramdisk from a failed run before its torn down? | 15:37 |
| derekh | adam_g: no, but I can run it locally and get you the ramdisk | 15:38 |
| derekh | adam_g: could take a little time though | 15:38 |
| adam_g | derekh, cool. that would be helpful. also, if you are running with libvirt, and you can install and use sgabios, you can capture console logs too | 15:38 |
| derekh | adam_g: ok, will give it a go | 15:39 |
| adam_g | derekh, thanks! | 15:39 |
| derekh | adam_g: I havn't use sgabios before is there anything I need to run (besides install it) | 15:40 |
| stendulker_ | dtantsur: Its building now. I think curl was hanging 6hrs back and was not getting redirected. | 15:41 |
| stendulker_ | dtantsur: Thank you :) | 15:41 |
| adam_g | derekh, install it, and add <bios useserial='yes'/> to the <os> section of the libvirt domain | 15:42 |
| adam_g | derekh, oh wait, nvm thats not needed to capture the ramdisk output | 15:42 |
| adam_g | derekh, you can just set the vm to log console to a file | 15:43 |
| *** achanda has quit IRC | 15:43 | |
| adam_g | derekh, for example http://paste.ubuntu.com/10017379/ | 15:44 |
| derekh | adam_g: ok, will give it a go, have to hack this into devtest | 15:46 |
| adam_g | derekh, FYI this is the related DIB change: https://review.openstack.org/#/c/140870/ | 15:48 |
| adam_g | which was released in 0.1.39 | 15:49 |
| adam_g | but without that, i'd expect the error your seeing (BOOT_SERVER not set) | 15:49 |
| derekh | adam_g: ack | 15:50 |
| marios | dtantsur: hey man, is this a known thing ? (from running devtest_undercloud waiting for the undercloud instance to come up, devtest_undercloud.sh. ) ironic conductor errors like http://paste.openstack.org/show/165639/ | 15:52 |
| marios | (anyone else) | 15:52 |
| * dtantsur is looking | 15:52 | |
| dtantsur | derekh, does it ^^^ look similar? | 15:53 |
| jroll | that would come from the current tripleo issues, yes | 15:54 |
| marios | dtantsur: thanks gents. ftr it ran to completion on friday. i just reprovisioned the box from scratch so may be something that landed recently | 15:54 |
| jroll | marios: is that the fedora job? | 15:54 |
| marios | jroll: no on my lab box | 15:54 |
| jroll | marios: oh, make sure your ramdisk is build with DIB 0.1.39 | 15:54 |
| marios | jroll: http://docs.openstack.org/developer/tripleo-incubator/devtest.html | 15:54 |
| dtantsur | but still Fedora, I suspect? | 15:54 |
| marios | dtantsur: yes | 15:54 |
| dtantsur | aha | 15:54 |
| dtantsur | I guess you're broken by the same thing | 15:54 |
| dtantsur | marios, can you attach to the vm and see what's going on there? | 15:55 |
| derekh | Yup, looks like https://bugs.launchpad.net/ironic/+bug/1417026 | 15:55 |
| jroll | any cores want to approve an early requirements bump? :) https://review.openstack.org/#/c/152127/ | 15:55 |
| jroll | NobodyCam: ^^ | 15:55 |
| marios | dtantsur: yeah i am. the log is from the seed /var/log/messages | 15:56 |
| adam_g | jroll, dtantsur im fine with the revert at https://review.openstack.org/#/c/152129/ | 15:56 |
| marios | derekh: yes thanks, indeed it does | 15:56 |
| *** jjohnson2 has joined #openstack-ironic | 15:56 | |
| jroll | adam_g: then +1 it? :P | 15:57 |
| adam_g | doh | 15:57 |
| dtantsur | jroll, adam_g, check-tripleo didn't pass on https://review.openstack.org/#/c/152129/ | 15:57 |
| dtantsur | should we recheck? | 15:58 |
| adam_g | dtantsur, is it the same issue? im having trouble finding any useful debug info in those jobs logs | 15:58 |
| jroll | dtantsur: yeah, let's recheck | 15:58 |
| jroll | it looks like setup scripts failed | 15:58 |
| dtantsur | adam_g, I don't think so, that's why I suggest rechecking | 15:59 |
| *** ndipanov has quit IRC | 15:59 | |
| adam_g | that one looks like some devtest error with bootstrapping the seed node | 15:59 |
| jroll | just submitted a recheck there | 15:59 |
| jroll | derekh: any news on if the revert worked for f20? | 15:59 |
| adam_g | i forgot my euro power adapter and need to go find one. back in a bit | 16:00 |
| *** beekneemech is now known as bnemec | 16:00 | |
| *** ndipanov has joined #openstack-ironic | 16:01 | |
| derekh | jroll: yup it did, added comment to bug | 16:01 |
| jroll | sweet | 16:01 |
| jroll | let's get 152129 done then | 16:02 |
| jroll | it's rechecking now | 16:02 |
| *** Marga_ has joined #openstack-ironic | 16:02 | |
| tteggel_ | hey ironiciers | 16:04 |
| *** pcrews has joined #openstack-ironic | 16:04 | |
| *** tteggel_ is now known as tteggel | 16:04 | |
| jroll | heya tteggel :) | 16:04 |
| jroll | oh that reminds me, you need juno patches approved yes? | 16:05 |
| tteggel | I have a stable patch that previously had a +W but needed a rebase, yeah | 16:05 |
| tteggel | just 1 noe | 16:05 |
| *** achanda has joined #openstack-ironic | 16:05 | |
| tteggel | which is https://review.openstack.org/#/c/149876/ | 16:05 |
| jroll | devananda: adam_g ^ | 16:05 |
| jroll | tteggel: those are all 2 of our stable cores :P | 16:06 |
| tteggel | that's good to know | 16:06 |
| NobodyCam | there was a reason we waited before landing that I seem to recall | 16:06 |
| jroll | oh? | 16:06 |
| jroll | huh | 16:06 |
| *** jjohnson2 has quit IRC | 16:07 | |
| tteggel | adam_g: thanks :) | 16:07 |
| *** anderbubble has joined #openstack-ironic | 16:10 | |
| *** achanda has quit IRC | 16:10 | |
| tteggel | NobodyCam: one of the backports was a bit of a mess. but i cleaned up after myself | 16:11 |
| *** zz_jgrimm is now known as jgrimm | 16:12 | |
| NobodyCam | tteggel: ahh :) | 16:16 |
| NobodyCam | tteggel: Thank you :) | 16:16 |
| *** david-lyle_afk is now known as david-lyle | 16:19 | |
| NobodyCam | brb | 16:23 |
| rloo | hey jroll, wrt https://review.openstack.org/152127, the line in global_requirements has '# Apache-2.0' | 16:25 |
| rloo | jroll: does it matter? will it confuse the bot or whatever that generates those patches? | 16:25 |
| *** lazy_prince is now known as killer_prince | 16:31 | |
| *** rameshg87 has joined #openstack-ironic | 16:36 | |
| *** rwsu-afk is now known as rwsu | 16:40 | |
| *** anderbubble has quit IRC | 16:40 | |
| jroll | rloo: it should be fine but I can add it if you want | 16:43 |
| rloo | jroll: if it is fine, then I'm ok. dtantsur said it was fine so | 16:44 |
| jroll | rloo: ok, cool | 16:44 |
| rloo | jroll: done... | 16:44 |
| jroll | woot | 16:45 |
| jroll | thanks! | 16:45 |
| *** vdrok_ has joined #openstack-ironic | 16:45 | |
| *** rameshg87 has quit IRC | 16:46 | |
| derekh | adam_g: my attempt to log the console resulted in a blank file, but there is the randisk and some screen shots http://goodsquishy.com/downloads/image-bug1417026/ | 16:47 |
| *** anderbubble has joined #openstack-ironic | 16:47 | |
| jroll | one more patch for cores to +A, fixes tripleo f20 jobs: https://review.openstack.org/#/c/152129/ | 16:47 |
| jroll | it's rechecking because tripleo failed, but derekh confirmed it works | 16:48 |
| Shrews | jroll: hm, interesting | 16:51 |
| jroll | Shrews: what's interesting about it? | 16:52 |
| Shrews | jroll: that it only broke fedora | 16:52 |
| jroll | Shrews: yeah... | 16:52 |
| jroll | weird | 16:53 |
| * Shrews would be interested in seeing an analysis of "why" | 16:53 | |
| jroll | I think that's what derekh is working on right now :P | 16:53 |
| Shrews | i guess bug 1401298 needs reopened now | 16:54 |
| derekh | jroll: I havn't looked into why yet, but have it reproduced locally, I can dig into it later if needed but at the moment have some other tripleo things to deal with | 16:55 |
| *** killer_prince has quit IRC | 16:55 | |
| *** killer_prince has joined #openstack-ironic | 16:56 | |
| *** killer_prince is now known as lazy_prince | 16:56 | |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Bump randomly set retry count and delay https://review.openstack.org/152179 | 16:57 |
| *** lazy_prince has quit IRC | 16:58 | |
| *** killer_prince has joined #openstack-ironic | 17:01 | |
| *** killer_prince is now known as lazy_prince | 17:01 | |
| *** Marga_ has quit IRC | 17:02 | |
| Shrews | we don't have a meeting now? | 17:02 |
| *** mjturek has quit IRC | 17:03 | |
| * Shrews is off by a week, methinks | 17:03 | |
| *** krtaylor has quit IRC | 17:04 | |
| jroll | you are :P | 17:04 |
| jroll | it's tonight | 17:04 |
| *** ndipanov has quit IRC | 17:07 | |
| NobodyCam | brb...bbt | 17:07 |
| jroll | adam_g: I re-opened https://bugs.launchpad.net/ironic/+bug/1401298 and left it assigned to you, hope that's ok | 17:08 |
| *** eghobo has joined #openstack-ironic | 17:08 | |
| *** coolsvap is now known as coolsvap|afk | 17:12 | |
| *** chuckC_ has quit IRC | 17:12 | |
| *** krtaylor has joined #openstack-ironic | 17:12 | |
| *** eghobo has quit IRC | 17:13 | |
| *** ChuckC_ has joined #openstack-ironic | 17:14 | |
| *** eghobo has joined #openstack-ironic | 17:15 | |
| *** ChuckC has quit IRC | 17:18 | |
| *** EmilienM is now known as EmilienM|afk | 17:18 | |
| *** dtantsur is now known as dtantsur|afk | 17:23 | |
| openstackgerrit | Merged stackforge/ironic-discoverd: Bump randomly set retry count and delay https://review.openstack.org/152179 | 17:29 |
| *** jistr has quit IRC | 17:39 | |
| *** jcoufal_ has quit IRC | 17:46 | |
| *** jlvillal has joined #openstack-ironic | 17:46 | |
| *** coolsvap|afk is now known as coolsvap | 17:52 | |
| *** derekh has quit IRC | 17:53 | |
| *** Haomeng has quit IRC | 17:56 | |
| *** Haomeng has joined #openstack-ironic | 17:57 | |
| *** davideagnello has joined #openstack-ironic | 17:58 | |
| *** spandhe has joined #openstack-ironic | 17:59 | |
| *** Marga_ has joined #openstack-ironic | 17:59 | |
| *** eghobo has quit IRC | 17:59 | |
| *** anderbubble has quit IRC | 18:00 | |
| *** EmilienM|afk is now known as EmilienM | 18:00 | |
| *** stendulker_ has quit IRC | 18:01 | |
| *** anderbubble has joined #openstack-ironic | 18:05 | |
| *** ijw has joined #openstack-ironic | 18:06 | |
| *** harlowja has joined #openstack-ironic | 18:11 | |
| *** athomas has quit IRC | 18:11 | |
| *** rlr0329 has joined #openstack-ironic | 18:15 | |
| rlr0329 | We have successfully deployed a 35-node site using Juno release of ironic with IPA. Now looking to productionize the ironic services. Any tips or best practices? | 18:15 |
| NobodyCam | rlr0329: awesome!!! | 18:17 |
| openstackgerrit | Merged openstack/ironic: Simplify policy.json https://review.openstack.org/148381 | 18:17 |
| JayF | great job rlr0329 | 18:17 |
| *** ijw has quit IRC | 18:17 | |
| rlr0329 | Thanks :-) | 18:18 |
| openstackgerrit | Merged openstack/python-ironicclient: Updated from global requirements https://review.openstack.org/149978 | 18:18 |
| *** ijw has joined #openstack-ironic | 18:18 | |
| anderbubble | Forgive me, as I expect this is meant to be clear; but what does “IPMI terminal port not supplied to IPMI driver.” mean I should do? | 18:19 |
| anderbubble | (in node-validate, for console) | 18:19 |
| NobodyCam | anderbubble: sounds like the port is not set | 18:19 |
| anderbubble | I found the bug at https://bugs.launchpad.net/ironic/+bug/1379911 requesting added clarity; but I don’t know where or how to set the port. | 18:20 |
| *** Haomeng|2 has joined #openstack-ironic | 18:20 | |
| *** ChuckC has joined #openstack-ironic | 18:21 | |
| *** Haomeng has quit IRC | 18:21 | |
| anderbubble | Heh: looks like this error message also appears in xcat… how could that code have possibly gotten in there? ;) | 18:22 |
| *** achanda has joined #openstack-ironic | 18:23 | |
| *** ChuckC_ has quit IRC | 18:24 | |
| anderbubble | NobodyCam, so I see that the node apparently needs a driver_info/port; but what port is that? Like COM1? How do I determine the correct port to put there? (Sorry: I haven’t ever really grok’d SoL) | 18:25 |
| *** jjohnson2 has joined #openstack-ironic | 18:25 | |
| NobodyCam | anderbubble: what driver? | 18:27 |
| anderbubble | pxe_ipmitool | 18:27 |
| NobodyCam | anderbubble: yep: https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/ipmitool.py#L1002 | 18:28 |
| NobodyCam | that will be the shellinabox port | 18:28 |
| *** ChuckC has quit IRC | 18:28 | |
| anderbubble | Oh; so this is a tcp port? | 18:29 |
| NobodyCam | anderbubble: https://help.ubuntu.com/community/shellinabox | 18:29 |
| anderbubble | That is *not at all* what I thought it was asking for. :p | 18:29 |
| jjohnson2 | shellinabox is crappy | 18:29 |
| NobodyCam | lol morning jjohnson2 | 18:30 |
| NobodyCam | :) | 18:30 |
| jjohnson2 | I'm my usual polite self | 18:30 |
| anderbubble | NobodyCam, jjohnson2, So, presumably each node needs its own port, then, if you want to be able to access nodes simultaneously? | 18:30 |
| jjohnson2 | NobodyCam, morning | 18:30 |
| jjohnson2 | anderbubble, so I needto make a recording... | 18:30 |
| anderbubble | And I’m my usual “please help me I can’t get Ironic working” self ;) | 18:30 |
| *** Marga_ has quit IRC | 18:30 | |
| jjohnson2 | anderbubble, and I might be distracting from ironic specifically, though I think ironic would benefit through the thing I'm about to record | 18:31 |
| jjohnson2 | I came in late to the conversation, but I'm guessing we are talking about console access | 18:32 |
| anderbubble | For the record, though: I actually got a node to almost completely provision at the very end of the day on Friday; though the setup didn’t work (it didn’t get a hostname, for example) | 18:32 |
| jjohnson2 | web based or cli based or both? | 18:32 |
| anderbubble | so I’m trying to get serial console working so I don’t have to keep walking to the DC ;) | 18:32 |
| anderbubble | jjohnson2, cli | 18:32 |
| NobodyCam | anderbubble: that sounds like a cloud-init thing | 18:32 |
| jjohnson2 | so I got a present for you then | 18:32 |
| jjohnson2 | oh, you know about my present | 18:32 |
| jjohnson2 | the thing I mentioned the other day | 18:32 |
| anderbubble | I’ll try horizon once I get it working reliably from the console | 18:33 |
| anderbubble | jjohnson2, yeah, I haven’t looked at it yet; but it did make it into org-mode ;) | 18:33 |
| *** eghobo has joined #openstack-ironic | 18:33 | |
| jjohnson2 | yeah, install confluent rpm | 18:33 |
| jjohnson2 | then start confluent service | 18:33 |
| openstackgerrit | Merged openstack/ironic: Bump oslo.rootwrap to 1.5.0 https://review.openstack.org/152127 | 18:33 |
| jjohnson2 | then you can confetty create /nodes/thisnodeisbob secret.hardwaremanagementuser=bigguy secret.hardwaremanagementpassword=supersecret hardwaremanagement.manager=my.bmc.here | 18:34 |
| jjohnson2 | oh and console.method=ipmi | 18:34 |
| jjohnson2 | then you can 'confetty thisnodeisbob' | 18:34 |
| jjohnson2 | and you hav econsole | 18:34 |
| openstackgerrit | Merged openstack/ironic: Revert "Do not pass PXE net config from bootloader to ramdisk" https://review.openstack.org/152129 | 18:35 |
| anderbubble | NobodyCam, so is this a server port that I want shellinabox to use; or is it a client port that I need to have shellinabox already running on to connect to? | 18:36 |
| *** wanyen has joined #openstack-ironic | 18:36 | |
| anderbubble | NobodyCam, that is, am I defining a new port here, or referencing an existing port (and need to go set up shellinabox out-of-band)? | 18:36 |
| *** Marga_ has joined #openstack-ironic | 18:38 | |
| wanyen | NobodyCam, Just want to confirm that we wil have ironic weekly meeting today at 0500 UTC? The ironic wiki https://wiki.openstack.org/wiki/Meetings/Ironic has not been updated yet. It still says next meeting is 01/26. | 18:38 |
| NobodyCam | wanyen: gah... yes I beleieve we are having a meeting | 18:39 |
| wanyen | NobodyCam, okay. I just want o make sure because I invited Jim to attend. It wil be 10pm his time. | 18:40 |
| *** rlr0329 has quit IRC | 18:43 | |
| *** ChuckC has joined #openstack-ironic | 18:47 | |
| anderbubble | jjohnson2, I see that the ipminative.py module is already in the ironic that I have installed; so can it be used with pyghmi already in Ironic? | 18:50 |
| *** achanda has quit IRC | 18:53 | |
| *** mrmoje has quit IRC | 18:53 | |
| anderbubble | jjohnson2, also, why isn’t pxe_ipminative being “tested in the upstream gate” ? | 18:54 |
| jjohnson2 | anderbubble, yeah, current pyghmi should be good for both | 18:55 |
| jjohnson2 | I don't know about pxe_ipminative testing at the moment, I will confess to not being up to date on how pyghmi is being consumed | 18:55 |
| *** mrmoje has joined #openstack-ironic | 18:55 | |
| jjohnson2 | 0.6.23 was released to address some issues in someones CI efforts... | 18:56 |
| *** achanda has joined #openstack-ironic | 18:57 | |
| *** ChuckC has quit IRC | 19:01 | |
| *** ChuckC has joined #openstack-ironic | 19:02 | |
| *** ijw has quit IRC | 19:02 | |
| *** chuckC_ has joined #openstack-ironic | 19:04 | |
| *** dprince has quit IRC | 19:06 | |
| jroll | anderbubble: does 'ipmitool... sol activate' not work? | 19:07 |
| *** wanyen has quit IRC | 19:07 | |
| jroll | anderbubble: pyghmi isn't tested in the upstream gate because the upstream gate doesn't have real hardware | 19:07 |
| * NobodyCam need to make the time to play with jjohnson2's ipmi listening stuff | 19:08 | |
| *** ijw has joined #openstack-ironic | 19:08 | |
| *** pelix has quit IRC | 19:09 | |
| anderbubble | jroll, it connected; but no echo | 19:10 |
| jroll | hrm | 19:10 |
| anderbubble | not sure why; I’m going to try reprovisioning the node and see if maybe the provisioned image just doesn’t report on the serial console | 19:10 |
| jroll | might need to set the console device differently in your kernel command | 19:11 |
| jroll | or the baud rate | 19:12 |
| jroll | e.g. console=ttyS4,115200n8 | 19:12 |
| anderbubble | jroll, where can I pass kernel parameters? | 19:12 |
| * anderbubble should just google that | 19:12 | |
| jroll | there's a config option for extra parameters | 19:12 |
| jroll | because ironic does it | 19:12 |
| jjohnson2 | NobodyCam, yeah, I'm trying to figure out how to record my desktop sanely | 19:13 |
| jroll | anderbubble: https://github.com/openstack/ironic/blob/master/etc/ironic/ironic.conf.sample#L1189-1191 | 19:13 |
| jjohnson2 | last time I was in Linux and it was straightforward, now that I'm on windows, it's harder | 19:13 |
| anderbubble | jroll, thanks | 19:13 |
| jroll | np | 19:13 |
| openstackgerrit | Merged openstack/python-ironicclient: Check if --config-drive is only used with provision state "active" https://review.openstack.org/151270 | 19:16 |
| *** andreykurilin_ has joined #openstack-ironic | 19:19 | |
| *** harmw has joined #openstack-ironic | 19:20 | |
| harmw | is there a dummy power driver for when I want to boot a machine my manually pressing the pwr button? :) | 19:21 |
| harmw | or do I rly need to build this myself | 19:21 |
| openstackgerrit | Merged openstack/python-ironicclient: ironicclient node-set-maintenance to accept true/false https://review.openstack.org/150692 | 19:21 |
| *** ChuckC has quit IRC | 19:23 | |
| *** EmilienM is now known as EmilienM|afk | 19:25 | |
| *** andreykurilin_ has quit IRC | 19:26 | |
| jroll | harmw: I don't believe there is in tree, maybe the fake power driver | 19:30 |
| jroll | has anyone thought about suspend/resume yet? | 19:30 |
| jroll | I know that's VM terminology, but would be nice to support it somehow (even if it's just power off/on) | 19:30 |
| *** dprince has joined #openstack-ironic | 19:32 | |
| harmw | jroll: so there is atleast something I could base it off of? | 19:33 |
| jroll | harmw: https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/fake.py#L34 | 19:34 |
| harmw | yea, I'm just reading that | 19:34 |
| *** Marga_ has quit IRC | 19:34 | |
| NobodyCam | harmw: there was this awhile ago: https://review.openstack.org/#/c/98950/ | 19:34 |
| *** Marga_ has joined #openstack-ironic | 19:35 | |
| *** penick has joined #openstack-ironic | 19:35 | |
| harmw | NobodyCam: thats exactly what I'm after | 19:35 |
| NobodyCam | :p | 19:36 |
| *** coolsvap is now known as coolsvap|afk | 19:36 | |
| harmw | to bad it's abandonent | 19:37 |
| NobodyCam | oh brb... quick walkies | 19:38 |
| *** Marga_ has quit IRC | 19:38 | |
| *** Marga__ has joined #openstack-ironic | 19:38 | |
| *** Marga__ has quit IRC | 19:39 | |
| *** Marga_ has joined #openstack-ironic | 19:39 | |
| *** Marga_ has quit IRC | 19:43 | |
| *** Marga_ has joined #openstack-ironic | 19:43 | |
| *** Marga_ has quit IRC | 19:45 | |
| *** Marga_ has joined #openstack-ironic | 19:46 | |
| *** EmilienM|afk is now known as EmilienM | 19:50 | |
| *** Marga_ has quit IRC | 19:54 | |
| *** Marga_ has joined #openstack-ironic | 19:54 | |
| *** eghobo has quit IRC | 19:54 | |
| *** achanda has quit IRC | 19:57 | |
| *** Marga_ has quit IRC | 19:59 | |
| *** anderbubble has quit IRC | 20:01 | |
| *** ijw_ has joined #openstack-ironic | 20:06 | |
| *** achanda has joined #openstack-ironic | 20:06 | |
| *** ijw has quit IRC | 20:08 | |
| *** anderbubble has joined #openstack-ironic | 20:13 | |
| *** dprince has quit IRC | 20:13 | |
| harmw | lifeless: will you ever complete that pxe_logpower driver? | 20:15 |
| harmw | or is it almost identical to FakePower? | 20:16 |
| *** mrda-away is now known as mrda | 20:16 | |
| mrda | Morning Ironic | 20:16 |
| NobodyCam | harmw: I do not expect that to ever land in tree | 20:17 |
| NobodyCam | morning mrda | 20:17 |
| NobodyCam | :) | 20:17 |
| spandhe | hello! | 20:17 |
| *** ijw_ has quit IRC | 20:18 | |
| NobodyCam | hi | 20:19 |
| *** pensu has quit IRC | 20:21 | |
| harmw | NobodyCam: me neither :) how should I enable the fake driver under enabled_drivers? | 20:22 |
| *** mjturek has joined #openstack-ironic | 20:22 | |
| NobodyCam | harmw: add the fake version of the driver you want to use: https://github.com/openstack/ironic/blob/master/setup.cfg#L41-L52 | 20:23 |
| harmw | ah | 20:23 |
| NobodyCam | to enabled_drivers | 20:24 |
| harmw | should've thought of that myself | 20:24 |
| *** Marga_ has joined #openstack-ironic | 20:25 | |
| NobodyCam | :) | 20:25 |
| *** pensu has joined #openstack-ironic | 20:28 | |
| *** Marga_ has quit IRC | 20:30 | |
| harmw | hm, and how would I be able to select the Fake driver when registering a new node? | 20:30 |
| harmw | enabled_drivers = pxe_ssh,pxe_ipmitool,fake | 20:31 |
| *** ijw has joined #openstack-ironic | 20:32 | |
| NobodyCam | harmw: fake has all interfaces mocked.. If your wanting to do something like an actual deploy you want a driver like fake_ipmitool | 20:36 |
| *** dprince has joined #openstack-ironic | 20:37 | |
| NobodyCam | oh wait | 20:38 |
| NobodyCam | https://github.com/openstack/ironic/blob/master/ironic/drivers/fake.py#L70 | 20:38 |
| NobodyCam | fake_pxe | 20:38 |
| NobodyCam | :-p | 20:38 |
| *** krtaylor has quit IRC | 20:40 | |
| NobodyCam | brb | 20:42 |
| Shrews | gah | 20:43 |
| Shrews | i will never understand how infra gets devstack to run on hp instances. i can NEVER get it to work "out of the box" | 20:44 |
| harmw | ah cool | 20:46 |
| harmw | and shouldn't tuskar display it as an option? | 20:47 |
| *** achanda has quit IRC | 20:50 | |
| harmw | hm, so tuskar doesn't use the enabled_drivers from ironic? the dropdown isn't updated.. | 20:51 |
| NobodyCam | harmw: I don't use taskar. :( | 20:54 |
| NobodyCam | maybe check in the TripleO channel | 20:54 |
| *** ChuckC has joined #openstack-ironic | 20:57 | |
| *** dprince has quit IRC | 21:00 | |
| harmw | ok sure | 21:03 |
| *** eghobo has joined #openstack-ironic | 21:05 | |
| *** achanda has joined #openstack-ironic | 21:13 | |
| *** krtaylor has joined #openstack-ironic | 21:13 | |
| NobodyCam | trown: you around? | 21:14 |
| trown | indeed | 21:14 |
| NobodyCam | hey hey :) | 21:14 |
| anderbubble | any reason why the adminPass displayed by nova wouldn’t work in the baremetal instance? | 21:15 |
| NobodyCam | just checking you have a item on the agenda ... is that still open? can I remove it? | 21:15 |
| anderbubble | Does cloud-init need to have been able to use the network to set it? | 21:15 |
| NobodyCam | anderbubble: sounds like cloud-init not working | 21:15 |
| NobodyCam | yep | 21:16 |
| jroll | anderbubble: correct, it needs to be able to reach the metadata service | 21:16 |
| trown | NobodyCam: it is still open, I got sidetracked with some other stuff, I will finish it up this week though | 21:16 |
| trown | oh, the agenda item is done with | 21:16 |
| NobodyCam | ok so I should leave it on the agenda | 21:16 |
| anderbubble | network is failing to start in my baremetal instance, and I don’t know how to debug it without being able to get into the instance in its provisioned state :/ | 21:16 |
| jroll | trown: do you need it on this week's... ok | 21:16 |
| NobodyCam | ahh ok | 21:16 |
| jroll | we can remove it then | 21:16 |
| trown | I have action items, no need for discussion | 21:16 |
| NobodyCam | i'm removing | 21:16 |
| trown | cool | 21:16 |
| jjohnson2 | all right, the most boring youtube video ever is about to be live | 21:16 |
| NobodyCam | jjohnson2: not borning if its setting up the ipmi listener | 21:18 |
| jjohnson2 | NobodyCam, it's 15 minutes, that's like 14 minutes and 50 seconds longer than my attention span | 21:18 |
| jjohnson2 | people have told me to talk more slowly and elaborate more on things, so I made a very concious effort | 21:19 |
| NobodyCam | :) | 21:19 |
| jjohnson2 | NobodyCam, anderbubble: youtu.be/8NEgyH_4gfo | 21:21 |
| anderbubble | jjohnson2, I’ll take a look once I’m not in the dc ;) | 21:21 |
| *** Haomeng|2 has quit IRC | 21:26 | |
| *** Marga_ has joined #openstack-ironic | 21:26 | |
| *** Haomeng|2 has joined #openstack-ironic | 21:28 | |
| *** Marga_ has quit IRC | 21:31 | |
| openstackgerrit | Michael Davies proposed openstack/ironic: Add logical name support to Ironic https://review.openstack.org/141737 | 21:32 |
| NobodyCam | brb | 21:35 |
| *** Marga_ has joined #openstack-ironic | 21:41 | |
| *** kevinbenton_ is now known as kevinbenton | 21:41 | |
| jjohnson2 | and now someone is saying I should have made it more terse... oh well, I can't do a video to please everyone | 21:41 |
| NobodyCam | lol | 21:42 |
| *** davideagnello has quit IRC | 21:43 | |
| *** pensu has quit IRC | 21:45 | |
| *** bradjones has quit IRC | 21:57 | |
| *** bradjones has joined #openstack-ironic | 21:58 | |
| *** bradjones has quit IRC | 21:58 | |
| *** bradjones has joined #openstack-ironic | 21:58 | |
| *** penick has quit IRC | 22:01 | |
| anderbubble | jjohnson2, you should make a video that tells me what I need to know without me having to watch it ;) | 22:01 |
| jroll | devananda: ping | 22:02 |
| jjohnson2 | anderbubble, well to be fare I go off topic to point out the stuff that I think is noteworthy for not normally being noticable | 22:02 |
| jroll | devananda: wondering what you think about returning e.g. 1.0 for api minimum version, rather than just 0 | 22:02 |
| *** andreykurilin_ has joined #openstack-ironic | 22:04 | |
| NobodyCam | jroll: its late at night for him now | 22:04 |
| jroll | NobodyCam: I'm aware | 22:05 |
| jroll | he can respond as needed | 22:05 |
| NobodyCam | :) ok heheeh | 22:05 |
| anderbubble | ok, so here’s the most recent reason my ironic node won’t boot: | 22:06 |
| anderbubble | it’s trying to configure eth0; but my node doesn’t have an eth0; it has an eno1 | 22:07 |
| anderbubble | *provisioning* works; it’s only the actual boot into the provisioned os that fails, which, of course, causes cloud-init to fail | 22:07 |
| anderbubble | presumably that’s the fault of the actual boot image, yes? | 22:08 |
| NobodyCam | anderbubble: sounds like it | 22:08 |
| anderbubble | so, who’s the right person to talk to about it being broken? | 22:10 |
| anderbubble | and if I try a different image, is it likely to have the correct interface in it? or is this something I can configure in my own diskimage-builder element? | 22:10 |
| NobodyCam | anderbubble: https://github.com/coreos/bugs/issues/36 | 22:11 |
| NobodyCam | look like your issue? | 22:11 |
| anderbubble | no; my issue is that the provisioned image only has a /etc/sysconfig/network-scripts/ifcfg-eth0; but the same file should, in stead, actually be ifcfg-eno1. The device just has a different name than the provisioned image expects. | 22:12 |
| anderbubble | I don’t know where the responsibility for configuring the network in the provisioned image lies. | 22:13 |
| jroll | where did your image come from? | 22:13 |
| anderbubble | disk-image-create centos7 baremetal -o my-image | 22:13 |
| jroll | mmm | 22:14 |
| jroll | yeah, I'd poke #tripleo about it then, I think | 22:14 |
| anderbubble | will do! thanks as always! | 22:14 |
| jroll | :) | 22:14 |
| *** ijw has quit IRC | 22:15 | |
| *** anderbubble has quit IRC | 22:15 | |
| *** andreykurilin|ho has joined #openstack-ironic | 22:17 | |
| *** andreykurilin_ has quit IRC | 22:17 | |
| *** anderbubble has joined #openstack-ironic | 22:22 | |
| *** andreykurilin|ho has quit IRC | 22:23 | |
| *** andreykurilin_ has joined #openstack-ironic | 22:24 | |
| *** jjohnson2 has quit IRC | 22:28 | |
| anderbubble | ironic node-list shows two of my nodes in “maintenance” mode, but ironic node-set-maintenance claims it doesn’t exist! :/ | 22:39 |
| *** davideagnello has joined #openstack-ironic | 22:42 | |
| anderbubble | (by which I mean ironic claims the command node-set-maintenance doesn’t exist) | 22:45 |
| jroll | update your client maybe? | 22:46 |
| devananda | jroll: oh. we're going to need to add microversion awareness to the client, which will be a little trickier | 22:47 |
| devananda | jroll: like auto discovering what commands, or states, or things, a given endpoint supports | 22:48 |
| jroll | devananda: right... | 22:48 |
| devananda | I mean, handling multiple endpoints in a way that doesn't absolutely make our users want to kill us ... | 22:48 |
| jroll | right | 22:49 |
| jroll | devananda: also, wdyt about returning the full versino for min/max version? | 22:49 |
| devananda | we probably need some local file cache of endpoint,min,max and decorators around functions, CLI commands, etc ... justlike the server is going to need | 22:49 |
| devananda | jroll: ++ | 22:49 |
| jroll | 1.0 rather than 0 | 22:49 |
| jroll | ok | 22:49 |
| devananda | jroll: total oversight on my part. | 22:49 |
| jroll | heh | 22:50 |
| jroll | and the 4 +2'ers :P | 22:50 |
| devananda | hehe | 22:52 |
| anderbubble | how can I clear a node reservation when the conductor has failed to clean it up? | 22:56 |
| jroll | update nodes set reservation=null where uuid='blah'; | 22:57 |
| jroll | :/ | 22:57 |
| NobodyCam | anderbubble: ^^^ | 22:57 |
| jroll | err | 22:57 |
| NobodyCam | ty jroll :) | 22:57 |
| jroll | anderbubble: was the conductor restarted? | 22:57 |
| anderbubble | jroll, yes | 22:57 |
| jroll | ok yeah | 22:57 |
| jroll | there's also a recent patch to clear on startup | 22:58 |
| anderbubble | ah | 22:58 |
| jroll | idk what version you're running | 22:58 |
| NobodyCam | anderbubble: are you trunk or packager version? | 22:58 |
| jroll | https://review.openstack.org/#/c/149038/ | 22:58 |
| anderbubble | jroll, openstack-ironic-conductor-2014.2-2.el7.centos.noarch | 22:58 |
| jroll | I think that actually got backported to stable/juno | 22:58 |
| *** yjiang5 has quit IRC | 22:58 | |
| *** alex_xu_ has quit IRC | 22:58 | |
| jroll | but idk if released yet | 22:58 |
| jroll | no idea how backports work | 22:58 |
| openstackgerrit | Dan Prince proposed openstack/ironic: Partial revert of 4606716 until we debug further https://review.openstack.org/152328 | 22:59 |
| devananda | it has not been released yet | 22:59 |
| devananda | a new rev of the stable/juno branch will get tagged this week, which will then take some time to trickle through the distros | 23:00 |
| jroll | devananda: can you remove your -2 here https://review.openstack.org/#/c/150784/ | 23:00 |
| jroll | please and thanks | 23:00 |
| * anderbubble feels dirty when he mixes SQL and openstack | 23:00 | |
| Shrews | devananda: i think my brain is suffering burnout on your MANAGEABLE state change, possibly not through any fault of yours :( | 23:00 |
| *** yjiang5 has joined #openstack-ironic | 23:01 | |
| *** alex_xu has joined #openstack-ironic | 23:01 | |
| Shrews | devananda: does this stack trace make any sense to you? (from a failed deploy using your change) http://paste.openstack.org/show/165812/ | 23:01 |
| Shrews | am i seeing a failed deploy trying to deploy again? or do i need to get away from my laptop? | 23:02 |
| devananda | jroll: done | 23:02 |
| jroll | thank you | 23:02 |
| devananda | Shrews: I'm sad that my patch is in any way hurting your brain | 23:03 |
| jroll | Shrews: that could just be a race, where the token expires between the request to ironic and the request from ironic to say glance | 23:03 |
| Shrews | devananda: it's only the surrounding testing of your patch, which i blame on setting up devstack on HP cloud instances | 23:03 |
| devananda | Shrews: my first thought was what jroll just said | 23:04 |
| jroll | Shrews: devstack works fine on rackspace cloud :P | 23:04 |
| devananda | Shrews: how long did the deploy take (time from "nova boot" until this crash) | 23:04 |
| Shrews | too long. i'm sure it hit the timeout. 30min ? | 23:04 |
| devananda | yup | 23:04 |
| devananda | sounds right | 23:04 |
| Shrews | well, on the plus side, the state changes seem to be sane | 23:07 |
| devananda | hehe | 23:08 |
| NobodyCam | oh wow just saw dan's bug and patch | 23:09 |
| devananda | anyone want to critique a talk abstract i'm about to submit to oscon? | 23:09 |
| devananda | NobodyCam: eh? | 23:10 |
| jroll | NobodyCam: hrm | 23:10 |
| NobodyCam | devananda: https://bugs.launchpad.net/ironic/+bug/1417307 / https://review.openstack.org/152328 | 23:10 |
| jroll | NobodyCam: so it wasn't the pxe thing that broke them? | 23:11 |
| jroll | or maybe both | 23:11 |
| *** chlong has joined #openstack-ironic | 23:11 | |
| NobodyCam | I don't know whats up yet | 23:12 |
| devananda | so adam_g broke something in tripleo ci recently | 23:12 |
| devananda | i wonder if this is related | 23:12 |
| Shrews | devananda: we reverted one of adam's changes this morning | 23:12 |
| jroll | devananda: we reverted adam's thing already, jfyi | 23:13 |
| devananda | Shrews: k k | 23:13 |
| jroll | but that's a good point | 23:13 |
| jroll | if networking wasn't coming up on the ramdisk, iscsi will never succeed | 23:13 |
| jroll | succeed/connect | 23:13 |
| jroll | but would it even try without getting the callback? | 23:13 |
| NobodyCam | jroll: I would hope not | 23:14 |
| NobodyCam | :-p | 23:14 |
| jroll | right, ok | 23:14 |
| jroll | so maybe both things broke their f20 jobs | 23:14 |
| jroll | but they couldn't see this one until the first was fixed | 23:14 |
| NobodyCam | yea .. | 23:15 |
| NobodyCam | that would make the most sense | 23:15 |
| NobodyCam | just strange that the sleep fixes the issue where activlty checking and forcing a re-read of the luns causes it to break | 23:16 |
| jroll | yeah, that is weird | 23:17 |
| jroll | maybe fedora has a newer iscsid or whatever? | 23:17 |
| jroll | with different options | 23:17 |
| NobodyCam | lazy_prince: you happen to have a log you can point me at | 23:17 |
| jroll | idk | 23:17 |
| NobodyCam | ya | 23:17 |
| NobodyCam | that we have seen before | 23:17 |
| *** penick has joined #openstack-ironic | 23:20 | |
| jroll | devananda: NobodyCam any opposition to me making the agent_ssh job vote on ironic patches? | 23:23 |
| jroll | opposition/objection? | 23:23 |
| devananda | jroll: yup. transitive dependencies | 23:23 |
| jroll | eh? | 23:23 |
| devananda | jroll: AFAIK, doing that today will put those jobs in the global gate queue | 23:24 |
| jroll | for other projects? | 23:24 |
| jroll | sigh | 23:24 |
| devananda | not that they'll vote on nova | 23:24 |
| devananda | no | 23:24 |
| jroll | oh | 23:24 |
| jroll | so what's the problem? | 23:24 |
| devananda | it makes the gate longer | 23:25 |
| devananda | the grand unified gate | 23:25 |
| jroll | so it does run on other projects? | 23:25 |
| devananda | no | 23:25 |
| jroll | (if it votes on ironic) | 23:25 |
| jroll | so uh | 23:25 |
| jroll | how does it make the gate that much longer? | 23:25 |
| jroll | it will increase it just by n * m, where n is number of ironic jobs and m is the time to run the agent job, right? | 23:26 |
| devananda | yup | 23:26 |
| devananda | and any instability in that job will exponentially effect all projects | 23:26 |
| jroll | is that really a reason to make it not vote, though? | 23:27 |
| devananda | tldr; i am proxying for clark and sdague right now, because i understand their concerns (or think I do) | 23:27 |
| jroll | the only instability we've ever seen in that job is general ironic instability | 23:27 |
| devananda | if they are willing to accept it, I'm fine with it voting on ironic | 23:27 |
| jroll | yeah, I mean right now we effectively aren't testing half of our deploy drivers | 23:28 |
| devananda | right | 23:28 |
| *** ijw has joined #openstack-ironic | 23:28 | |
| devananda | i maen, I want it to be voting on ironic | 23:28 |
| jroll | which is pretty lame | 23:28 |
| jroll | imho | 23:28 |
| jroll | ok | 23:28 |
| devananda | but i also dont want to put more strain on infra/qa | 23:29 |
| jroll | yeah | 23:29 |
| *** anderbubble has quit IRC | 23:33 | |
| *** mjturek has quit IRC | 23:35 | |
| *** vdrok_ has quit IRC | 23:35 | |
| *** mlsoule has quit IRC | 23:38 | |
| *** derekh has joined #openstack-ironic | 23:39 | |
| devananda | jroll: adam_g: have I been imaging it, or is check-tempest-dsvm-ironic-parallel-nv is looking fairly stable lately? | 23:45 |
| jroll | I haven't been paying a ton of attention but I feel like I see it failing somewhat regularly | 23:46 |
| devananda | :-/ | 23:47 |
| jroll | I might only notice when it fails | 23:48 |
| jroll | don't rely on my memory :x | 23:48 |
| * devananda relies more on adam's stats than his own memory, too | 23:52 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
