| rloo_ | devananda: hi. | 00:04 |
|---|---|---|
| rloo_ | devananda: ok, we / I will start -2'ing patches that I encounter that implement unapproved specs. | 00:05 |
| jroll | gah, I'm going to end up with so many -2's :P | 00:05 |
| rloo_ | jroll: speak up now. say you disagree! | 00:05 |
| devananda | rloo_: i think it will get confusing if more than one (or at most two) people are doing that | 00:05 |
| jroll | well, to be fair, I think it's fine | 00:06 |
| rloo_ | jroll: ha ha. ok then. | 00:06 |
| jroll | I just tend to push code early and often | 00:06 |
| rloo_ | devananda: so you want me to be the bad cop? I can do it ;) | 00:06 |
| devananda | HP is going to have a lot of -2's as well, eg. for the windows disk images, ipmi double briding, and ilo driver | 00:06 |
| devananda | rloo_: that'd be great | 00:06 |
| devananda | rloo_: not you beign the bad cop - i mean, the help keeping track of what to -2 / when to un-2 stuff | 00:07 |
| jroll | serious question, is it ok for people to review code even if the spec is not approved? | 00:07 |
| devananda | jroll: yes | 00:07 |
| jroll | review, not merge | 00:07 |
| jroll | ok | 00:07 |
| jroll | great :) | 00:07 |
| devananda | jroll: i think it's super helpful, and also sometimes a spec is hard to talk about without seeing *some* code at least | 00:07 |
| rloo_ | devananda: jroll: so if a review is -2, it can still be reviewed? can new versions be pushed? | 00:07 |
| jroll | +1 | 00:07 |
| devananda | rloo_: yep. new versions can be pushed. it can be reviewed. all the things can happen EXCEPT it can't merge. | 00:08 |
| jroll | totally cool with me | 00:08 |
| devananda | for long patch chains | 00:08 |
| rloo_ | devananda: great. I feel better about it then. I thought a -2 meant it was frozen. | 00:08 |
| devananda | we can also -2 a placeholder at the head of the dependency tree | 00:08 |
| devananda | jroll: that ^ might be really nice for you guys | 00:08 |
| devananda | rloo_: nope :) just blocked from merging until such time as we all agree it *should* merge | 00:09 |
| jroll | devananda: indeed | 00:09 |
| devananda | rloo_: this prevents miscommunication among a large and timezone distributed team resulting in folks merging code that wasn't approved | 00:09 |
| devananda | that's basically all | 00:09 |
| devananda | right now, my sense is we should only do this for largish features | 00:10 |
| rloo_ | devananda: great! I'll add something about it to https://wiki.openstack.org/wiki/Ironic/Specs_Process. | 00:10 |
| devananda | things that we actively dont want to spend a lot of review cycles on until the design is proven out | 00:10 |
| devananda | rloo_: thanks! | 00:10 |
| rloo_ | devananda: so basically, a feature that requires some design. if it is cookie cutter stuff, then it is ok. | 00:11 |
| devananda | right | 00:11 |
| devananda | or a bug fix - also fine | 00:11 |
| rloo_ | I think dtantsur|afk is doing a good job of finding bugs that should be associated with BPs ;) | 00:12 |
| devananda | yep | 00:12 |
| devananda | a recent example of changes that should (and now do) have a spec | 00:12 |
| devananda | lucas' management interface patch series | 00:12 |
| devananda | i mean, it's half a dozen patches, building a new internal API *and* a new endpoint in the REST API | 00:13 |
| devananda | even though I like 80% of the code (all that's not WIP at this point) I don't think we should actually *land* it until we agree onthe API changes and the architecture. | 00:13 |
| devananda | the same can be said for the HP iLO driver | 00:14 |
| devananda | (except i haven't reviewed that code recently) | 00:14 |
| devananda | rloo_: so, the logging improement spec | 00:14 |
| devananda | rloo_: i actually dont think taht needs a spec any more. it's just a bunch of internal minor enhancements | 00:14 |
| devananda | rloo_: the more we iterated on that spec, the less I feel like that's appropriate for a spec. | 00:15 |
| rloo_ | devananda: yeah. I agree. That's why it was easy for me to review it :D | 00:15 |
| devananda | haha | 00:15 |
| devananda | ok then | 00:15 |
| rloo_ | devananda: I think a few bugs to cover it would be fine. | 00:15 |
| jroll | devananda: some would argue that bad logging is a bug :P | 00:15 |
| rloo_ | devananda: but I was wondering about AUDIT? | 00:15 |
| devananda | i'm going to abandon it unless someone objects now | 00:15 |
| jroll | +1 | 00:15 |
| rloo_ | devananda: but not everyone is here now? :D | 00:16 |
| rloo_ | devananda: the one useful thing there, is that i DO NOT want to see some XYZ person push up a review that changes the logging to use the _LX() macros. | 00:16 |
| lifeless | mrda: submitted a pyconau paper? | 00:17 |
| mrda | lifeless: Unfortunately I have a conflict that week - won't be there | 00:18 |
| mrda | but otherwise i would | 00:18 |
| lifeless | mrda: nuts :) | 00:19 |
| devananda | rloo_: i dislike the driveby "clean up" that is really wasteful of rewviwe time too -- but do you have another reason? | 00:19 |
| rloo_ | devananda: I disagree with those macros. It may make translators lifes easier but it dirties the code and doesn't make sense. That's all. | 00:20 |
| devananda | rloo_: :) | 00:20 |
| mrda | lifeless: it's a pretty tight schedule too - pretty much leave PDX on Wedn night from midcycle and arrive in BNE on Friday morning for pyconau | 00:20 |
| devananda | rloo_: i suggest taking that up on the i18n and oslo lists then | 00:20 |
| rloo_ | (bad enough we have to remember not to use _() for DEBUG). What is code for, if we have to special case these things. | 00:20 |
| rloo_ | devananda: I thought about it, but it would mean having to do some background research to see why they did it etc. I'm hoping that some other sensible folks will take up that challenge ;) | 00:21 |
| devananda | mrda: oh, btw, we *should* be using _() for exceptions | 00:21 |
| mrda | devananda: +1 | 00:22 |
| mrda | sorry | 00:22 |
| mrda | I disagree | 00:22 |
| mrda | :) | 00:22 |
| mrda | I'm not sure translating exceptions make sense | 00:22 |
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic: Cleanup virt driver tests and verify final spawn https://review.openstack.org/99518 | 00:23 |
| mrda | It's just to report where a problem occured back to this team. I'm not sure the extra burden on translators provides any real benefit. | 00:23 |
| jroll | exceptions are not always bugs | 00:24 |
| lifeless | mrda: exceptions are used by operators | 00:24 |
| jroll | ^ | 00:24 |
| mrda | but I'm happy to fall into line if, once you've heard me, disagree :) | 00:24 |
| JayF | step 1 of troubleshooting Ironic in our lab: grep TRACE /var/log/ironic-conductor.log :) | 00:24 |
| JayF | haha | 00:24 |
| jroll | I may be an ironic developer, but I'm also an ironic deployer/operator, and readable exceptions | 00:24 |
| jroll | help | 00:24 |
| jroll | hahaha +1 JayF | 00:25 |
| mrda | lifeless: really? Exceptions are exceptional situations. I would argue that if an operator should be looking at something, that exceptions should be turned into a (translated) log | 00:25 |
| devananda | being fluent in english, working with mostly software written in english, being easily able to google for errors in said software, I *think* it's actually helpful for others if our exceptions are not translated away from english | 00:25 |
| jroll | mrda: unfortunately that's not always true :| | 00:26 |
| devananda | that said, it's a totally ethnocentric view and I know that I dont have any idea what it's like to run english software for folks who speak very little english | 00:26 |
| JayF | devananda: that's an interesting point | 00:26 |
| devananda | and if THEY want the exceptions in their local language -- graet | 00:26 |
| JayF | I don't care if they are translated or not, I just think having run a lot of software as an op, that exceptions are absolutely consumed by ops. all the time. | 00:26 |
| devananda | i think we should enable that | 00:26 |
| mrda | if someone sends me a stack trace in Estonian, it's of no use to me. (apologies to any native Estonian speakers) | 00:26 |
| jroll | so, the software can be run in whichever language the deployer chooses, correct? why not enable that? | 00:26 |
| devananda | mrda: it'll still have file names and line numbers | 00:27 |
| jroll | only the actual exception message is translated, not the code in the stack trace | 00:27 |
| mrda | because i18n is a huge task, and we should minimise the amount of work required there, especially if there isn't much benefit | 00:27 |
| devananda | right | 00:27 |
| devananda | jroll: right | 00:27 |
| lifeless | mrda: that is a strawman | 00:27 |
| mrda | lifeless: disagree :) | 00:28 |
| devananda | mrda: "isn't much benefit" is an opinion | 00:28 |
| lifeless | mrda: exceptions are flow control. They are not exceptional - if they were, they wouldn't be encouraged as a flow control construct. | 00:28 |
| lifeless | mrda: its non-local return | 00:28 |
| mrda | if something is of benefit to an operator, I'd suggest it should be logged | 00:28 |
| devananda | mrda: BUT, fwiw, the _L*() macros were recently added to oslo.i18n for *just that purpose* :) | 00:28 |
| lifeless | mrda: other languages have different definitions for this | 00:28 |
| lifeless | mrda: I'm not against it being logged for operators. | 00:28 |
| devananda | so we can now do LOG.info(_LI('message')) | 00:28 |
| jroll | mrda: if i18n bandwidth is a thing, I'll personally petition people at rax to hire translators to work on openstack :P | 00:28 |
| devananda | and the translations team can then choose whether they prioritize translating INFO messagse over EXCEPTION messages | 00:29 |
| devananda | for example | 00:29 |
| devananda | so we are again enabling the team for whom this matters to make their own decisions | 00:29 |
| devananda | rather than forcing our opinion on them | 00:29 |
| devananda | given that I know I DONT KNOW what matters to a chinese operator of Ironic, I think this is good | 00:29 |
| JayF | This seems very much like a mailing list discussion more than an IRC one? Like it impacts much more than Ironic and I think if we made a different decision than other projects it'd be the silliest decision possilbe :) | 00:29 |
| devananda | totally | 00:30 |
| devananda | except it was :) | 00:30 |
| mrda | So there was a discussion about this on a nova review | 00:30 |
| devananda | i'm just recapping it for ya'll | 00:30 |
| mrda | and a few people came down on either side | 00:30 |
| devananda | and there's now a vote at the TC about setting policy on this | 00:30 |
| JayF | ah, okay | 00:30 |
| JayF | and you're on the TC | 00:30 |
| JayF | so this is deva-lobbying-time in #openstack-ironic | 00:30 |
| devananda | and I -1'd it because there hasn't been a concensus in the i18n team /as far as I know/ | 00:30 |
| mrda | devananda: cool - happy to follow whatever policy is ;) | 00:31 |
| devananda | hah | 00:31 |
| openstackgerrit | Alex Weeks proposed a change to openstack/ironic-specs: Add metric reporting spec. https://review.openstack.org/100729 | 00:31 |
| JayF | devananda: 301 redirect to tha i18n team seems like a wise choice | 00:31 |
| jroll | aweeks: welcome, new contributor! | 00:31 |
| jroll | (I love that message) | 00:31 |
| JayF | I didn't get that message | 00:32 |
| JayF | we need a 'thanks, reliable contributer' message poster | 00:32 |
| JayF | that comes with a free get-out-of-'-2' card, monopoly style | 00:32 |
| * jroll makes JayF a cardboard sign | 00:32 | |
| devananda | right - i need to get ready for evening plans | 00:33 |
| mrda | devananda: so I think that log spec has served a purpose already - it's making us confirm what we're going to do in our project :) | 00:33 |
| devananda | by which i mean, apartment shopping | 00:33 |
| devananda | mrda: indeed :) | 00:33 |
| mrda | :) | 00:33 |
| jroll | devananda: right on, good luck :) | 00:33 |
| mrda | devananda: have fun appartment shopping! | 00:33 |
| aweeks | jroll: :) | 00:37 |
| *** Penick has quit IRC | 00:38 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Updated from global requirements https://review.openstack.org/96228 | 00:42 |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/88722 | 00:42 |
| *** bcaldwell has quit IRC | 00:46 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/python-ironicclient: Updated from global requirements https://review.openstack.org/96263 | 00:47 |
| *** rloo_ has quit IRC | 00:54 | |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Factor out TFTPImageCache https://review.openstack.org/100734 | 00:57 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Factor out deploy info from PXE driver https://review.openstack.org/100735 | 00:57 |
| *** ellenh has quit IRC | 00:57 | |
| *** rwsu has quit IRC | 00:58 | |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Nova Ironic Driver to not get deploy r&k from flavor https://review.openstack.org/95551 | 01:03 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Add migration script to instance_info https://review.openstack.org/96136 | 01:03 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Move PXE instance level parameters to instance_info https://review.openstack.org/94855 | 01:03 |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic: Drivers determine acceptable power states https://review.openstack.org/86744 | 01:04 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Adding swift temp url support https://review.openstack.org/81391 | 01:05 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Factor out TFTPImageCache https://review.openstack.org/100734 | 01:05 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Factor out deploy info from PXE driver https://review.openstack.org/100735 | 01:05 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Add methods to ipmitool driver https://review.openstack.org/100364 | 01:05 |
| *** matsuhashi has joined #openstack-ironic | 01:17 | |
| *** theDavidAikenToo has joined #openstack-ironic | 01:27 | |
| *** eghobo has quit IRC | 01:28 | |
| *** theDavidAiken has quit IRC | 01:30 | |
| *** nosnos has joined #openstack-ironic | 01:31 | |
| *** theDavidAikenToo has quit IRC | 01:33 | |
| *** yjiang5 is now known as yjiang5_away | 01:43 | |
| *** dwalleck has joined #openstack-ironic | 02:06 | |
| *** dwalleck has quit IRC | 02:08 | |
| *** dwalleck has joined #openstack-ironic | 02:08 | |
| *** coolsvap|afk is now known as coolsvap | 02:25 | |
| *** matsuhashi has quit IRC | 02:29 | |
| *** matsuhashi has joined #openstack-ironic | 02:32 | |
| *** harlowja is now known as harlowja_away | 02:48 | |
| *** theDavidAikenToo has joined #openstack-ironic | 02:54 | |
| *** matsuhashi has quit IRC | 03:00 | |
| *** vinbs has joined #openstack-ironic | 03:02 | |
| *** ramineni has joined #openstack-ironic | 03:19 | |
| *** nosnos has quit IRC | 03:27 | |
| *** jrist has quit IRC | 03:49 | |
| *** eghobo has joined #openstack-ironic | 03:51 | |
| *** matsuhashi has joined #openstack-ironic | 03:53 | |
| *** nosnos has joined #openstack-ironic | 04:07 | |
| *** matsuhashi has quit IRC | 04:12 | |
| *** matsuhashi has joined #openstack-ironic | 04:12 | |
| *** matsuhashi has quit IRC | 04:17 | |
| *** jgrimm has joined #openstack-ironic | 04:19 | |
| *** matsuhashi has joined #openstack-ironic | 04:21 | |
| *** rakesh_hs2 has joined #openstack-ironic | 04:28 | |
| *** nikunj2512 has joined #openstack-ironic | 04:30 | |
| *** matsuhashi has quit IRC | 04:36 | |
| *** matsuhashi has joined #openstack-ironic | 04:36 | |
| *** matsuhashi has quit IRC | 04:39 | |
| *** eghobo has quit IRC | 04:39 | |
| *** matsuhashi has joined #openstack-ironic | 04:39 | |
| *** eghobo has joined #openstack-ironic | 04:40 | |
| *** Poornima|mtg has joined #openstack-ironic | 04:56 | |
| *** rameshg87 has joined #openstack-ironic | 04:58 | |
| *** matsuhashi has quit IRC | 05:11 | |
| *** matsuhashi has joined #openstack-ironic | 05:11 | |
| *** radsy has quit IRC | 05:11 | |
| *** matsuhashi has quit IRC | 05:16 | |
| *** theDavidAikenToo has quit IRC | 05:16 | |
| *** matsuhashi has joined #openstack-ironic | 05:17 | |
| *** lazy_prince has joined #openstack-ironic | 05:17 | |
| *** sabah has joined #openstack-ironic | 05:23 | |
| *** matsuhashi has quit IRC | 05:24 | |
| *** matsuhashi has joined #openstack-ironic | 05:24 | |
| *** dwalleck has quit IRC | 05:36 | |
| openstackgerrit | ryo kurahashi proposed a change to openstack/ironic: Add 'context' parameter from the get_console_output() methods. https://review.openstack.org/100768 | 05:43 |
| vinbs | Morning Ironic! :) | 05:49 |
| *** sseago_ has quit IRC | 05:59 | |
| *** subah has joined #openstack-ironic | 06:01 | |
| *** sabah has quit IRC | 06:01 | |
| *** max_lobur has joined #openstack-ironic | 06:01 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Imported Translations from Transifex https://review.openstack.org/96063 | 06:02 |
| *** sysexit has joined #openstack-ironic | 06:03 | |
| *** bmahalakshmi has joined #openstack-ironic | 06:04 | |
| *** wendar has quit IRC | 06:04 | |
| *** wendar_ has joined #openstack-ironic | 06:04 | |
| mrda | Hi vinbs | 06:07 |
| *** dwalleck has joined #openstack-ironic | 06:08 | |
| *** k4n0 has joined #openstack-ironic | 06:11 | |
| *** dwalleck has quit IRC | 06:14 | |
| *** k4n0 has quit IRC | 06:19 | |
| *** nikunj2512 has quit IRC | 06:19 | |
| *** vinbs has quit IRC | 06:19 | |
| *** pcrews has quit IRC | 06:19 | |
| *** openstackgerrit has quit IRC | 06:19 | |
| *** foexle has joined #openstack-ironic | 06:23 | |
| *** k4n0 has joined #openstack-ironic | 06:23 | |
| *** nikunj2512 has joined #openstack-ironic | 06:23 | |
| *** vinbs has joined #openstack-ironic | 06:23 | |
| *** pcrews has joined #openstack-ironic | 06:23 | |
| *** openstackgerrit has joined #openstack-ironic | 06:23 | |
| *** k4n0 has quit IRC | 06:24 | |
| *** k4n0 has joined #openstack-ironic | 06:24 | |
| *** romcheg has joined #openstack-ironic | 06:35 | |
| *** jcoufal has joined #openstack-ironic | 06:37 | |
| *** Manishanker has quit IRC | 06:43 | |
| openstackgerrit | Sandhya Balakrishnan proposed a change to openstack/ironic: Updates Ironic Guide with deployment information This patch updates Ironic deployment information Adds 3 deployment diagrams Describes deployment process https://review.openstack.org/100787 | 06:48 |
| *** romcheg has quit IRC | 06:48 | |
| *** Poornima|mtg is now known as Poornima | 06:56 | |
| *** dwalleck has joined #openstack-ironic | 07:00 | |
| *** dwalleck has quit IRC | 07:05 | |
| *** matsuhashi has quit IRC | 07:08 | |
| *** matsuhas_ has joined #openstack-ironic | 07:08 | |
| *** eguz has joined #openstack-ironic | 07:08 | |
| *** eguz has quit IRC | 07:11 | |
| *** jcoufal has quit IRC | 07:12 | |
| *** eghobo has quit IRC | 07:12 | |
| *** jcoufal has joined #openstack-ironic | 07:13 | |
| *** bmahalakshmi has quit IRC | 07:13 | |
| *** bmahalakshmi has joined #openstack-ironic | 07:17 | |
| *** takadayuiko has joined #openstack-ironic | 07:22 | |
| *** athomas has joined #openstack-ironic | 07:27 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 07:30 | |
| Haomeng | vinbs: morning:) | 07:33 |
| *** romcheg has joined #openstack-ironic | 07:35 | |
| *** rameshg87 has quit IRC | 07:43 | |
| *** rameshg87 has joined #openstack-ironic | 07:45 | |
| openstackgerrit | Sandhya Balakrishnan proposed a change to openstack/ironic: Updates Ironic Guide with deployment information https://review.openstack.org/100787 | 07:47 |
| mrda | 'night Ironic | 07:53 |
| *** mrda is now known as mrda-away | 07:53 | |
| *** coolsvap is now known as coolsvap|afk | 07:55 | |
| *** max_lobur has quit IRC | 07:56 | |
| openstackgerrit | Imre Farkas proposed a change to openstack/ironic: Fix workaround for the "device is busy" problem https://review.openstack.org/90126 | 07:59 |
| *** jistr has joined #openstack-ironic | 08:03 | |
| openstackgerrit | A change was merged to openstack/ironic: Improving unit tests for _do_sync_power_state https://review.openstack.org/99271 | 08:25 |
| *** viktors|afk is now known as viktors | 08:28 | |
| *** lucasagomes has joined #openstack-ironic | 08:28 | |
| openstackgerrit | ryo kurahashi proposed a change to openstack/ironic: Add 'context' parameter from get_console_output() https://review.openstack.org/100768 | 08:35 |
| openstackgerrit | Ramakrishnan G proposed a change to openstack/ironic: Add IloDriver and its IloPower module https://review.openstack.org/89500 | 08:40 |
| *** takadayuiko has quit IRC | 08:42 | |
| *** subah has quit IRC | 08:43 | |
| *** sabah has joined #openstack-ironic | 08:44 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Nova Ironic Driver to not get deploy ramdisk and kernel from flavor https://review.openstack.org/95551 | 08:49 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add migration script to instance_info https://review.openstack.org/96136 | 08:49 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Move PXE instance level parameters to instance_info https://review.openstack.org/94855 | 08:49 |
| *** pelix has joined #openstack-ironic | 08:50 | |
| *** max_lobur has joined #openstack-ironic | 08:55 | |
| *** dtantsur|afk is now known as dtantsur | 08:56 | |
| dtantsur | Morning, Ironic! | 08:56 |
| *** matsuhas_ has quit IRC | 09:01 | |
| *** dwalleck has joined #openstack-ironic | 09:03 | |
| *** dwalleck has quit IRC | 09:08 | |
| *** matsuhashi has joined #openstack-ironic | 09:09 | |
| *** martyntaylor has joined #openstack-ironic | 09:09 | |
| yuriyz | morning Ironic, dtantsur | 09:09 |
| romcheg | Morning guys! | 09:09 |
| *** mkerrin has quit IRC | 09:22 | |
| *** loki184 has joined #openstack-ironic | 09:22 | |
| dtantsur | morning, yuriyz, romcheg | 09:23 |
| *** Mikhail_D_ltp has quit IRC | 09:32 | |
| *** Faizan has joined #openstack-ironic | 09:34 | |
| *** matsuhashi has quit IRC | 09:38 | |
| *** igordcard has joined #openstack-ironic | 09:41 | |
| openstackgerrit | Ghe Rivero proposed a change to openstack/ironic: Migration to oslo.db https://review.openstack.org/92138 | 09:41 |
| romcheg | viktors, GheRivero I can see both of you guys have patches that do the same | 09:46 |
| openstackgerrit | Victor Sergeyev proposed a change to openstack/ironic: Use oslo.db library https://review.openstack.org/42159 | 09:46 |
| viktors | romcheg: lol | 09:47 |
| viktors | GheRivero: hi! | 09:48 |
| GheRivero | viktors: hi! | 09:49 |
| *** ramineni has quit IRC | 09:49 | |
| viktors | as romcheg told, looks like, that we are doing the same thing | 09:49 |
| GheRivero | viktors: let me see the delta between our patches | 09:49 |
| viktors | GheRivero: ok ) | 09:49 |
| *** ifarkas_ has quit IRC | 09:51 | |
| viktors | GheRivero: they looks very similar | 09:51 |
| GheRivero | yeah. almost the same. that means we are both completely wrong :) | 09:52 |
| viktors | :) | 09:53 |
| viktors | GheRivero: so how should we solve this situation? | 09:55 |
| GheRivero | the only 'important; difference I see are: | 09:55 |
| *** rakesh_hs2 has quit IRC | 09:56 | |
| GheRivero | the oslo.db requirement, I have the license | 09:56 |
| GheRivero | minor changes in the ironic.conf (don;t knwo who is uptodate) | 09:56 |
| GheRivero | and the fucking timeutils mock in the tests :) same approach, different way | 09:57 |
| *** Mikhail_D_ltp has joined #openstack-ironic | 09:57 | |
| *** rakesh_hs2 has joined #openstack-ironic | 09:57 | |
| GheRivero | the best could be to merge both patches and add a co-authored in the commit msg | 09:57 |
| viktors | GheRivero: I tried to use an another approach in the previous patch. See https://review.openstack.org/#/c/42159/10/ironic/db/sqlalchemy/models.py | 09:58 |
| GheRivero | yeah I saw that. But i prefer our way now :) | 09:58 |
| *** jcoufal has quit IRC | 09:59 | |
| *** ramineni has joined #openstack-ironic | 10:00 | |
| *** matsuhashi has joined #openstack-ironic | 10:01 | |
| GheRivero | I'm going to do a rebase of both patches to eliminate the base patch delta | 10:02 |
| *** Haomeng has quit IRC | 10:03 | |
| viktors | GheRivero: ok, but I suppose, that we will have a minimum delta between these patches. | 10:03 |
| openstackgerrit | Anusha Ramineni proposed a change to openstack/ironic-specs: Out-Of-Band firmware update for Baremetal Servers https://review.openstack.org/100842 | 10:03 |
| GheRivero | it's just to avoid extra noise, and confirm the up-to-date status of the ironic.conf | 10:04 |
| *** matsuhashi has quit IRC | 10:06 | |
| *** matsuhashi has joined #openstack-ironic | 10:08 | |
| GheRivero | it's weird. we also dissagre in several 1/2 blank lines but pep8 didnot complaint to any of us | 10:09 |
| *** Manishanker has joined #openstack-ironic | 10:14 | |
| *** Manishanker has quit IRC | 10:14 | |
| *** ifarkas has joined #openstack-ironic | 10:15 | |
| *** matsuhashi has quit IRC | 10:17 | |
| *** matsuhashi has joined #openstack-ironic | 10:18 | |
| openstackgerrit | Anusha Ramineni proposed a change to openstack/ironic-specs: Out-Of-Band firmware update for Baremetal Servers https://review.openstack.org/100842 | 10:24 |
| *** ifarkas has quit IRC | 10:26 | |
| *** mkerrin has joined #openstack-ironic | 10:27 | |
| *** ifarkas has joined #openstack-ironic | 10:29 | |
| viktors | GheRivero: If you'll have any questions, I'll be back in a 2 hours | 10:29 |
| *** viktors is now known as viktors|afk | 10:30 | |
| *** sabah has quit IRC | 10:33 | |
| *** sabah has joined #openstack-ironic | 10:34 | |
| openstackgerrit | Anusha Ramineni proposed a change to openstack/ironic-specs: Out-Of-Band firmware update for Baremetal Servers https://review.openstack.org/100842 | 10:36 |
| *** jcoufal has joined #openstack-ironic | 10:43 | |
| *** bmahalakshmi has quit IRC | 11:02 | |
| *** ramineni has left #openstack-ironic | 11:04 | |
| *** Haomeng has joined #openstack-ironic | 11:09 | |
| lucasagomes | dtantsur, any reason why 97447 wasn't approved yet? | 11:10 |
| openstackgerrit | Martin Geisler proposed a change to openstack/ironic: Add UTF-8 coding lines to all Python files https://review.openstack.org/97027 | 11:10 |
| lucasagomes | code looks good and it's passing the tests | 11:10 |
| openstackgerrit | A change was merged to openstack/ironic: Update IronicHostManager tests to better match how code works https://review.openstack.org/99026 | 11:11 |
| *** nikunj2512 has quit IRC | 11:13 | |
| *** dshulyak has quit IRC | 11:14 | |
| *** romcheg has quit IRC | 11:15 | |
| *** Mikhail_D_ltp has quit IRC | 11:15 | |
| *** Haomeng has quit IRC | 11:16 | |
| *** yuriyz has quit IRC | 11:16 | |
| *** max_lobur has quit IRC | 11:16 | |
| *** Mikhail_D_wk has quit IRC | 11:17 | |
| *** viktors|afk has quit IRC | 11:17 | |
| *** sysexit has quit IRC | 11:17 | |
| dtantsur | lucasagomes, a bit afraid to approve it :) I don't now this code very well | 11:18 |
| lucasagomes | dtantsur, a sure, yeah np... seems to work fine (looking at the tests). Will try to run it locally here | 11:21 |
| *** romcheg has joined #openstack-ironic | 11:25 | |
| *** max_lobur has joined #openstack-ironic | 11:25 | |
| *** yuriyz has joined #openstack-ironic | 11:25 | |
| openstackgerrit | A change was merged to openstack/ironic: Add missing stats to IronicNodeState https://review.openstack.org/99027 | 11:26 |
| *** dshulyak has joined #openstack-ironic | 11:26 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 11:26 | |
| *** viktors has joined #openstack-ironic | 11:27 | |
| *** Mikhail_D_wk has joined #openstack-ironic | 11:27 | |
| *** romcheg1 has joined #openstack-ironic | 11:28 | |
| *** romcheg2 has joined #openstack-ironic | 11:28 | |
| *** saripurigopi has joined #openstack-ironic | 11:32 | |
| *** romcheg has quit IRC | 11:32 | |
| *** romcheg1 has quit IRC | 11:33 | |
| *** martyntaylor has quit IRC | 11:35 | |
| saripurigopi | Hi | 11:36 |
| saripurigopi | I'm trying to test vendor_passthru functions..., | 11:37 |
| saripurigopi | This is the query i'm sending. | 11:37 |
| saripurigopi | curl -H "X-Auth-Token:${token}" -X POST -d '' "http://10.105.214.179:6385/v1/nodes/2d70d135-85b5-4f75-b741-0ead90a42b29/vendor_passthru?method=get_firmware_info&data=" | 11:37 |
| saripurigopi | Its a simple get function, I'm not sure what data I need to pass. | 11:38 |
| saripurigopi | it is failing with the below error | 11:38 |
| saripurigopi | {"error_message": "{\"debuginfo\":\"Traceback (most recent call last):\\n\\n File \\\"/usr/local/lib/python2.7/dist-packages/wsmeext/pecan.py\\\", line 82, in callfunction\\n result = f(self, *args, **kwargs)\\n\\n File \\\"/opt/stack/ironic/ironic/api/controllers/v1/node.py\\\", line 439, in post\\n pecan.request.context, node_uuid, method, data, topic)\\n\\n File \\\"/opt/stack/ironic/ironic/conductor/rp | 11:38 |
| saripurigopi | {"error_message": "{\"debuginfo\":\"Traceback (most recent call last):\\n\\n File \\\"/usr/local/lib/python2.7/dist-packages/wsmeext/pecan.py\\\", line 82, in callfunction\\n result = f(self, *args, **kwargs)\\n\\n File \\\"/opt/stack/ironic/ironic/api/controllers/v1/node.py\\\", line 439, in post\\n pecan.request.context, node_uuid, method, data, topic)\\n\\n File \\\"/opt/stack/ironic/ironic/conductor/rp | 11:38 |
| saripurigopi | File \\\"/opt/stack/ironic/ironic/openstack/common/excutils.py\\\", line 70, in __exit__\\n six.reraise(self.type_, self.value, self.tb)\\n\\n File \\\"/opt/stack/ironic/ironic/conductor/manager.py\\\", line 287, in vendor_passthru\\n **info)\\n\\nTypeError: validate() argument after ** must be a mapping, not unicode\\n\\n\",\"faultcode\":\"Server\",\"faultstring\":\"validate() argument after ** must be a mapp | 11:39 |
| *** sabah has quit IRC | 11:39 | |
| saripurigopi | can some one help me how to solve this... | 11:39 |
| dtantsur | saripurigopi, try data={} | 11:40 |
| dtantsur | or maybe you need to shell-escape it like \{\} I don't remember | 11:40 |
| yuriyz | data should be in json body | 11:40 |
| *** martyntaylor has joined #openstack-ironic | 11:42 | |
| yuriyz | an example https://github.com/openstack/diskimage-builder/blob/master/elements/deploy-ironic/init.d/80-deploy-ironic#L39 | 11:42 |
| *** matsuhashi has quit IRC | 11:48 | |
| *** sseago has joined #openstack-ironic | 11:50 | |
| *** matsuhashi has joined #openstack-ironic | 11:51 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Nova Ironic Driver to not get deploy ramdisk and kernel from flavor https://review.openstack.org/95551 | 11:51 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add migration script to instance_info https://review.openstack.org/96136 | 11:51 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Move PXE instance level parameters to instance_info https://review.openstack.org/94855 | 11:51 |
| saripurigopi | dtantsur, passing data={}, doesn't have any effect, it is still failing with same error. | 11:52 |
| dtantsur | saripurigopi, please see yuriyz' comment, it seems to be more correct | 11:53 |
| saripurigopi | I'm trying that. | 11:53 |
| *** matsuhashi has quit IRC | 11:56 | |
| *** matsuhashi has joined #openstack-ironic | 11:56 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add migration script to instance_info https://review.openstack.org/96136 | 11:57 |
| *** sysexit has joined #openstack-ironic | 11:58 | |
| *** lucasagomes is now known as lucas-hungry | 11:59 | |
| *** matsuhashi has quit IRC | 12:01 | |
| *** jistr is now known as jistr|english | 12:02 | |
| *** romcheg has joined #openstack-ironic | 12:02 | |
| *** max_lobur1 has joined #openstack-ironic | 12:02 | |
| *** yuriyz has quit IRC | 12:03 | |
| *** max_lobur has quit IRC | 12:03 | |
| *** yuriyz has joined #openstack-ironic | 12:03 | |
| *** romcheg2 has quit IRC | 12:04 | |
| *** romcheg has quit IRC | 12:07 | |
| *** dshulyak has quit IRC | 12:08 | |
| *** Mikhail_D_wk has quit IRC | 12:08 | |
| *** max_lobur1 has quit IRC | 12:08 | |
| *** yuriyz has quit IRC | 12:08 | |
| *** viktors has quit IRC | 12:08 | |
| *** Mikhail_D_ltp has quit IRC | 12:08 | |
| *** romcheg has joined #openstack-ironic | 12:11 | |
| *** yuriyz has joined #openstack-ironic | 12:11 | |
| *** max_lobur has joined #openstack-ironic | 12:11 | |
| *** viktors has joined #openstack-ironic | 12:12 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 12:13 | |
| *** dshulyak has joined #openstack-ironic | 12:13 | |
| *** Mikhail_D_wk has joined #openstack-ironic | 12:13 | |
| *** loki184 has quit IRC | 12:16 | |
| *** saripurigopi has quit IRC | 12:23 | |
| *** jdob has joined #openstack-ironic | 12:27 | |
| dtantsur | lucas-hungry and folks: could you review https://review.openstack.org/#/c/94371/ please? I was a good boy and reviewed your patches :D | 12:34 |
| dtantsur | (has 3x +1, needs attention from cores) | 12:35 |
| dtantsur | Also https://review.openstack.org/#/c/94620/ | 12:35 |
| *** lucas-hungry is now known as lucasagomes | 12:46 | |
| *** linggao has joined #openstack-ironic | 12:47 | |
| *** viktors1 has joined #openstack-ironic | 12:50 | |
| lucasagomes | dtantsur, sure thing will take a look | 12:50 |
| *** vinbs has quit IRC | 12:51 | |
| *** vinbs has joined #openstack-ironic | 12:52 | |
| *** rakesh_hs2 has quit IRC | 12:54 | |
| lucasagomes | dtantsur, one thing about that 2* | 12:55 |
| dtantsur | ? | 12:56 |
| lucasagomes | dtantsur, https://review.openstack.org/#/c/94371/8/ironic/drivers/modules/pxe.py | 12:56 |
| lucasagomes | dtantsur, multiplying by 2 | 12:56 |
| dtantsur | yeah, I remember | 12:56 |
| lucasagomes | right, there's anyway to check whether it's a raw image or not? | 12:56 |
| lucasagomes | like a metadata at glance saying it's a raw image so that you don't need to multiply? | 12:57 |
| lucasagomes | otherwise ur freeing twice more than you need if the image is already raw | 12:57 |
| dtantsur | lucasagomes, accessing Glance, I guess. I'm not sure what is worse: accessing Glance or freeing a bit more space | 12:57 |
| dtantsur | all these calculations are not precise anyway | 12:57 |
| dtantsur | and anyway I'd prefer to have more space available | 12:58 |
| dtantsur | so don't know | 12:58 |
| lucasagomes | dtantsur, sure... but when u get the download_size ur already accessing it | 12:58 |
| lucasagomes | I mean ur asking the image service what's the download size | 12:58 |
| lucasagomes | dtantsur, right that's fine | 12:58 |
| dtantsur | yeah, you're right, but it may make sense to have some room for whatever reasons. Not *2, but who knows :) | 12:59 |
| *** rloo has joined #openstack-ironic | 13:01 | |
| *** Poornima|mtg has joined #openstack-ironic | 13:13 | |
| lucasagomes | dtantsur, +1 + a question inline | 13:15 |
| *** Poornima has quit IRC | 13:17 | |
| *** k4n0 has quit IRC | 13:17 | |
| *** k4n0 has joined #openstack-ironic | 13:18 | |
| dtantsur | lucasagomes, hmm... what is the question finally? Do you suggest to change the comment? | 13:18 |
| *** Poornima|afk has joined #openstack-ironic | 13:19 | |
| lucasagomes | dtantsur, what would happen if multiplying by 2 is not enough? | 13:19 |
| *** k4n0 has quit IRC | 13:22 | |
| *** Poornima|mtg has quit IRC | 13:22 | |
| *** k4n0 has joined #openstack-ironic | 13:23 | |
| dtantsur | lucasagomes, left a reply | 13:23 |
| lucasagomes | dtantsur, thanks, will take a look | 13:24 |
| lucasagomes | reviewing the other one | 13:24 |
| dtantsur | one more point is that I want to stop improving these caching tasks at some moment, because it seems infinite :) | 13:25 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Adding swift temp url support https://review.openstack.org/81391 | 13:28 |
| *** romcheg has quit IRC | 13:30 | |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Factor out TFTPImageCache https://review.openstack.org/100734 | 13:30 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Factor out deploy info from PXE driver https://review.openstack.org/100735 | 13:30 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Add methods to ipmitool driver https://review.openstack.org/100364 | 13:30 |
| *** jrist has joined #openstack-ironic | 13:35 | |
| *** vinbs has quit IRC | 13:39 | |
| *** jistr|english is now known as jistr | 13:42 | |
| openstackgerrit | Dmitry Tantsur proposed a change to openstack/ironic: Fix leaking DB details to API on error https://review.openstack.org/73121 | 13:46 |
| *** jistr has quit IRC | 13:59 | |
| Shrews | adam_g: devananda: Improvement from the dashboard from last night: http://bit.ly/1ikKVFA That tracks both tempest AND devstack changes (hopefully) | 14:00 |
| * Shrews borrows heavily from dtantsur's work | 14:00 | |
| *** jistr has joined #openstack-ironic | 14:01 | |
| dtantsur | Shrews, wow cool :) | 14:02 |
| rloo | hello Ironic'ers. | 14:02 |
| rloo | Shrews: nice! | 14:03 |
| *** nosnos has quit IRC | 14:03 | |
| *** rloo has quit IRC | 14:05 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Updated from global requirements https://review.openstack.org/96228 | 14:05 |
| *** rloo has joined #openstack-ironic | 14:05 | |
| *** theDavidAiken has joined #openstack-ironic | 14:05 | |
| dtantsur | rloo, hi | 14:06 |
| rloo | hi dtantsur :) | 14:07 |
| *** rloo has quit IRC | 14:09 | |
| NobodyCam | good morning says the man make'n coffee | 14:09 |
| *** rloo has joined #openstack-ironic | 14:09 | |
| dtantsur | NobodyCam, morning | 14:11 |
| NobodyCam | morning dtantsur | 14:12 |
| *** devananda has quit IRC | 14:14 | |
| *** devananda has joined #openstack-ironic | 14:15 | |
| lucasagomes | morning rloo NobodyCam Shrews | 14:18 |
| rloo | hiya lucasagomes | 14:19 |
| viktors1 | GheRivero: around? | 14:22 |
| NobodyCam | morning lucasagomes rloo & Shrews :) | 14:23 |
| Shrews | adam_g: devananda: sdague asked for input on 98317. I don't understand the neutron well enough to understand the benefit of that change, but the ironic tests all failed, so i voted -1 for that fact alone. | 14:23 |
| Shrews | s/the neutron/neutron/ | 14:24 |
| Shrews | NobodyCam: morning! | 14:24 |
| NobodyCam | :) | 14:24 |
| rloo | morning NobodyCam | 14:25 |
| NobodyCam | :) | 14:26 |
| GheRivero | viktors1: pong | 14:27 |
| NobodyCam | morning GheRivero | 14:27 |
| GheRivero | morning NobodyCam | 14:27 |
| mgagne | I know this could be more or less a neutron question but does Ironic support configuration of bonding interfaces? | 14:27 |
| viktors1 | GheRivero: as for our patches - do we have any decision ? | 14:29 |
| rloo | So that people aren't surprised, if you have patches associated with blueprints, see the last paragraph: https://wiki.openstack.org/wiki/Ironic/Specs_Process#Ironic_Specs_Process | 14:29 |
| *** rwsu has joined #openstack-ironic | 14:29 | |
| rloo | linggao: ^^^ | 14:30 |
| *** dwalleck has joined #openstack-ironic | 14:36 | |
| *** dwalleck has quit IRC | 14:36 | |
| *** dwalleck has joined #openstack-ironic | 14:36 | |
| rloo | lucasagomes, devananda: wrt blueprints & specs. https://blueprints.launchpad.net/ironic/+spec/add-node-instance-info has a link to the specification. | 14:40 |
| dtantsur | lucasagomes, I was always interested what people mean by assigning to a bug, but not moving to "In Progress" :) | 14:41 |
| rloo | lucasagomes, devananda: but the spec process https://wiki.openstack.org/wiki/Ironic/Specs_Process mentions copying the spec info to the blueprint. Can we delete that? | 14:41 |
| dtantsur | lucasagomes, that's e.g. re https://bugs.launchpad.net/ironic/+bug/1331494 | 14:41 |
| lucasagomes | rloo, oh, I think we can delete and add a link there then | 14:41 |
| lucasagomes | dtantsur, ah lol | 14:42 |
| rloo | lucasagomes: +1. Will check with devananda though cuz I think he wrote that. | 14:42 |
| lucasagomes | dtantsur, it's because there's no patch in the queue right now fixing that | 14:42 |
| lucasagomes | dtantsur, I'm working on it and once I submit it will mark it as "In progress" automatically | 14:42 |
| lucasagomes | but idk if that's the right way to do it | 14:42 |
| lucasagomes | dtantsur, should I mark it as in progress manually? | 14:43 |
| dtantsur | lucasagomes, do we set "In Progress" only with patch? I'm not against, but we need a clear policy | 14:43 |
| dtantsur | because people do it differently every time | 14:43 |
| lucasagomes | dtantsur, yeah I'm a bit lost there as well | 14:43 |
| *** romcheg has joined #openstack-ironic | 14:43 | |
| lucasagomes | the way I do was to mark as "In progress" after the patch is submitted but idk whether it's right or not | 14:43 |
| lucasagomes | I see, maybe we need to come up with some policy | 14:43 |
| lucasagomes | like if ur working on it even if there's no patch just mark it as "in progress" | 14:44 |
| * lucasagomes mark the bug as in progress | 14:44 | |
| dtantsur | lucasagomes, I'm starting to think that we need to write it down, yeah. And how long we kepp patches assigned and what to do with abandoned patches and so on | 14:44 |
| *** jistr has quit IRC | 14:44 | |
| lucasagomes | dtantsur, +1 to write it down | 14:44 |
| * dtantsur makes note to come up with ideas on the next meeting | 14:45 | |
| *** viktors2 has joined #openstack-ironic | 14:47 | |
| *** viktors1 has quit IRC | 14:47 | |
| *** jistr has joined #openstack-ironic | 14:48 | |
| rameshg87 | hi jroll, are you there ? | 14:49 |
| *** rloo has quit IRC | 14:49 | |
| *** rloo has joined #openstack-ironic | 14:50 | |
| *** martyntaylor1 has joined #openstack-ironic | 14:50 | |
| *** Faizan has quit IRC | 14:51 | |
| *** martyntaylor has quit IRC | 14:52 | |
| jroll | morning rameshg87 :) | 14:52 |
| rameshg87 | jroll, good morning (actually it's evening here my time :-) ) | 14:52 |
| *** rloo has quit IRC | 14:53 | |
| rameshg87 | jroll, had a question on your patch https://review.openstack.org/#/c/81391/36 | 14:53 |
| rameshg87 | swift_temp_url support | 14:53 |
| *** rloo has joined #openstack-ironic | 14:53 | |
| jroll | rameshg87: ask away :) | 14:53 |
| *** Mikhail_D_ltp has quit IRC | 14:54 | |
| rameshg87 | jroll, i was just trying out your changes. i got a url of the form "'swift+http://10.10.1.10:5000/v2.0/glance/0dd0dd0c-b9cd-424d-9ff6-b8c03d2d334d?temp_url_sig=c6fa90fb08973acb10da68f637f567d12a3e45b8&temp_url_expires=1403123539'" | 14:54 |
| rameshg87 | jroll, i mean the method glance_service.swift_temp_url() returned me back a url of this form | 14:55 |
| jroll | rameshg87: are you using direct URLs with glance? | 14:55 |
| rameshg87 | jroll, yes | 14:55 |
| jroll | ok | 14:55 |
| jroll | there may be a bug there :) | 14:55 |
| jroll | rameshg87: can you make a note on the review? | 14:56 |
| rameshg87 | jroll, my direct_url returned from glance looks similar too: | 14:56 |
| *** martyntaylor1 has quit IRC | 14:56 | |
| rameshg87 | | direct_url | swift+http://service%3Aglance-swift:openstack@10.10.1.10:5000/v2.0/glance | | 14:56 |
| rameshg87 | | | /e053f93c-0a7a-435b-b5ac-5262c3fc33e7 | | 14:56 |
| jroll | rameshg87: although I'd be curious if you stuck that url in curl, if it would just work | 14:56 |
| jroll | right | 14:56 |
| jroll | that's where the bug would come from | 14:57 |
| openstackgerrit | Dmitry Tantsur proposed a change to openstack/ironic: Fix leaking DB details to API on error https://review.openstack.org/73121 | 14:57 |
| rameshg87 | jroll, i am using devstack. it configures swift as backing store for glance | 14:57 |
| rameshg87 | jroll, am i missing something ? | 14:57 |
| jroll | rameshg87: can you run: curl 'swift+http://10.10.1.10:5000/v2.0/glance/0dd0dd0c-b9cd-424d-9ff6-b8c03d2d334d?temp_url_sig=c6fa90fb08973acb10da68f637f567d12a3e45b8&temp_url_expires=1403123539' | 14:57 |
| jroll | rameshg87: I don't think so, I think *we* are missing something | 14:58 |
| jroll | although in our environment, using the direct url thing worked | 14:58 |
| *** rloo has quit IRC | 14:58 | |
| rameshg87 | jroll, since curl doesn't know 'swift+http', it says unsupported schemee | 14:58 |
| rameshg87 | jroll, and if i remove swift+ it doesn't work :-) | 14:59 |
| *** rloo has joined #openstack-ironic | 14:59 | |
| jroll | rameshg87: ok | 14:59 |
| jroll | oh? | 14:59 |
| rameshg87 | jroll, how does your direct_url look like ? | 14:59 |
| *** Poornima|afk has quit IRC | 14:59 | |
| jroll | rameshg87: we've since turned that off in our environment :/ | 15:00 |
| jroll | rameshg87: this could have broken since | 15:00 |
| *** foexle has quit IRC | 15:00 | |
| jroll | rameshg87: can you leave a note on the review and I will take a look later? | 15:00 |
| rameshg87 | jroll, sure .. | 15:01 |
| jroll | rameshg87: in the meantime, I would think a quick hack would be in line 190, use: 'scheme': parsed.scheme.lstrip('swift+') | 15:02 |
| rameshg87 | jroll, i tried doing that myself, but removing 'swift+' doesn't help :-( | 15:02 |
| jroll | rameshg87: and then check your temp_url_key or something | 15:02 |
| *** viktors2 is now known as viktors_ | 15:02 | |
| jroll | rameshg87: if you run 'swift stat' with admin credentials, it will give you the temp_url_key you need | 15:03 |
| rameshg87 | jroll, with swift-temp-url command-line i am able to generate tmp url's correctly | 15:03 |
| rameshg87 | jroll, but to that command i give the input as /v1/AUTH_xxxxx/container/object | 15:03 |
| jroll | rameshg87: hmm | 15:04 |
| rameshg87 | jroll, here the direct_url doesn't look that way .. it looks like "swift+http://account:username:password@host:port/version/container/object" | 15:04 |
| rameshg87 | jroll, that's why i was interested to see how your direct_url looks like .. | 15:05 |
| jroll | rameshg87: indeed, I don't have a quick way to get one right now, sorry :( | 15:05 |
| rameshg87 | jroll, i am just checking your tests | 15:06 |
| rameshg87 | jroll, https://review.openstack.org/#/c/81391/36/ironic/tests/test_glance_service.py | 15:06 |
| jroll | right, looks to be the same as yours | 15:06 |
| jroll | (fwiw, JoshNang wrote most of this code, but I'm still happy to help) | 15:06 |
| rameshg87 | jroll, so i am not missing anything here...let me have a look if i can find out what's going wrong | 15:07 |
| *** viktors_ has quit IRC | 15:07 | |
| rameshg87 | jroll, will keep you updated in the review :-) | 15:07 |
| *** dwalleck has quit IRC | 15:07 | |
| jroll | rameshg87: ok :) note that we test for a url coming out with swift+http | 15:07 |
| rameshg87 | jroll, thanks :-) | 15:08 |
| *** viktors1 has joined #openstack-ironic | 15:08 | |
| jroll | np :) | 15:08 |
| rameshg87 | dtantsur, lucasagomes, NobodyCam, can you please have a look at the design spec: https://review.openstack.org/#/c/97455/ | 15:09 |
| rameshg87 | dtantsur, lucasagomes, NobodyCam, code is almost ready in the review, but waiting for design spec to be approved :-) | 15:09 |
| dtantsur | rameshg87, I'll try to, but it's likely I won't have time today, sorry :( | 15:09 |
| *** mdorman has joined #openstack-ironic | 15:10 | |
| rameshg87 | dtantsur, thanks..will wait for your review :-) | 15:10 |
| *** romcheg has quit IRC | 15:21 | |
| openstackgerrit | Dmitry Tantsur proposed a change to openstack/ironic: Fix leaking DB details to API on error https://review.openstack.org/73121 | 15:23 |
| *** athomas has quit IRC | 15:23 | |
| *** blamar has joined #openstack-ironic | 15:25 | |
| *** romcheg has joined #openstack-ironic | 15:27 | |
| *** viktors has quit IRC | 15:29 | |
| rameshg87 | rloo, please have a look at the design spec since you had reviewed it before https://review.openstack.org/#/c/97455/ | 15:32 |
| rloo | hi rameshg87. I'm struggling with trying to do reviews, fix code, and am not sure where to fit in reviewing specs :-(. Let me see if I can do some spec reviews later today or tomorrow. | 15:34 |
| rameshg87 | rloo, thanks , will wait .. :-) | 15:35 |
| *** rameshg87 has left #openstack-ironic | 15:43 | |
| dtantsur | studying time, will try to get back in the evening | 15:44 |
| *** dtantsur is now known as dtantsur|afk | 15:44 | |
| openstackgerrit | Victor Sergeyev proposed a change to openstack/ironic: Use oslo.db library https://review.openstack.org/42159 | 15:50 |
| *** eghobo has joined #openstack-ironic | 15:52 | |
| *** viktors1 is now known as viktors|afk | 15:55 | |
| *** viktors|afk has quit IRC | 15:58 | |
| devananda | g'morning, all! | 15:59 |
| lucasagomes | morning devananda | 16:00 |
| devananda | rloo: on the specs / bp link, i copied the template from nova. I'm fine keeping the BP as just ~1 paragraph, and using the "specification URL" link in LP | 16:00 |
| jroll | morning devananda | 16:01 |
| rloo | devananda: great. Less work for people! I'll update to reflect that. (Morning!) | 16:01 |
| devananda | rloo: thanks! | 16:02 |
| rloo | devananda: btw, are the specs posted somewhere after they are approved? | 16:02 |
| *** jcoufal has quit IRC | 16:02 | |
| rloo | devananda: since they are docs, it seems odd to look at them in git. | 16:02 |
| devananda | rloo: github formats .rst nicely | 16:05 |
| rloo | devananda: yeah, but I can't eg get this page from it: http://docs-draft.openstack.org/55/97455/5/check/gate-ironic-specs-docs/641eb45/doc/build/html/ | 16:08 |
| *** dwalleck has joined #openstack-ironic | 16:09 | |
| devananda | ooh, good point | 16:10 |
| rloo | devananda: I think the specs are a good form of docn, so it would be nice to make them avail to the world, outside of github. | 16:11 |
| NobodyCam | morning devananda | 16:13 |
| NobodyCam | hummm was(/is) there a issue with em interfaces? | 16:14 |
| NobodyCam | anyone know off the tops of there head? | 16:15 |
| Shrews | em? | 16:15 |
| *** yjiang5_away is now known as yjiang5 | 16:15 | |
| *** hemna_ is now known as hemna | 16:15 | |
| NobodyCam | Shrews: em0 vs eth0 | 16:16 |
| Shrews | oh | 16:18 |
| *** eghobo has quit IRC | 16:18 | |
| *** eghobo has joined #openstack-ironic | 16:18 | |
| jroll | NobodyCam: what kind of issue? | 16:18 |
| *** ellenh has joined #openstack-ironic | 16:19 | |
| *** lazy_prince has quit IRC | 16:20 | |
| rloo | lucasagomes, devananda: isn't this done? https://blueprints.launchpad.net/python-ironicclient/+spec/simple-crud | 16:21 |
| NobodyCam | systems getting dchp address (with next-server and boot-file set) and never attempting to send out a tftp request | 16:21 |
| jroll | NobodyCam: oh, I have no idea :) | 16:21 |
| devananda | rloo: oh, i dont think we've been using BPs for the client | 16:21 |
| devananda | rloo: yes, that's done | 16:21 |
| rloo | devananda: well, here's the only other BP: https://blueprints.launchpad.net/python-ironicclient/+spec/py33-support | 16:22 |
| NobodyCam | s/send/sending/ | 16:22 |
| openstackgerrit | Nisha Agarwal proposed a change to openstack/ironic-specs: Discover node properties at node-create/node-update https://review.openstack.org/100951 | 16:28 |
| *** wendar_ is now known as wendar | 16:31 | |
| *** k4n0 has quit IRC | 16:37 | |
| *** max_lobur has quit IRC | 16:43 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 16:45 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add set_on_error_hook to TaskManager https://review.openstack.org/100957 | 16:48 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Fix nodes left in an incosistent state if no workers https://review.openstack.org/100958 | 16:48 |
| *** martyntaylor has joined #openstack-ironic | 16:50 | |
| lucasagomes | folks the instance_info work is unblocked now (devstack and tempest dependencies are merged), so if u want to help review it starts here: https://review.openstack.org/#/c/94855/ | 16:53 |
| devananda | lucasagomes: \o/ | 16:56 |
| lucasagomes | :D | 16:56 |
| openstackgerrit | Nisha Agarwal proposed a change to openstack/ironic-specs: Discover node properties at node-create/node-update https://review.openstack.org/100951 | 16:56 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add migration script to instance_info https://review.openstack.org/96136 | 16:58 |
| openstackgerrit | Nisha Agarwal proposed a change to openstack/ironic-specs: Discover node properties at node-create/node-update https://review.openstack.org/100951 | 17:01 |
| lucasagomes | jroll, had to fix some nits on the migration_script pointed by dtantsur|afk ^ (in case u want to rebase it again, sorry :( ) | 17:01 |
| jroll | heh, no worries | 17:02 |
| jroll | thanks for the heads up | 17:02 |
| jroll | lucasagomes: hopefully that will just land today and won't need to worry about it ;) | 17:03 |
| lucasagomes | jroll, +1 :) | 17:03 |
| openstackgerrit | Nisha Agarwal proposed a change to openstack/ironic-specs: Discover node properties at node-create/node-update https://review.openstack.org/100951 | 17:03 |
| lucasagomes | jroll, also, I don't think u may need to depend on the migration script, you put a dependency on the previous on (if that helps) | 17:03 |
| lucasagomes | you can put a dependency on the previous patch* | 17:04 |
| jroll | lucasagomes: that's true | 17:04 |
| jroll | I'll change it when I have some more patches done | 17:04 |
| lucasagomes | aight :) | 17:04 |
| *** harlowja_away is now known as harlowja | 17:06 | |
| jroll | dtantsur|afk: curious if you're planning on factoring out the image caching stuff in pxe.py any time soon :) | 17:10 |
| jroll | dtantsur|afk: I might end up doing it... | 17:10 |
| *** lazy_prince has joined #openstack-ironic | 17:12 | |
| *** Mikhail_D_ltp has quit IRC | 17:13 | |
| * devananda spends an hour reviewing tempest patches | 17:20 | |
| devananda | romcheg: hi! around? | 17:22 |
| romcheg | devananda: Hi, i'm having a class right how. 30 minutes pls | 17:23 |
| openstackgerrit | A change was merged to openstack/ironic: Fix host manager node detection logic https://review.openstack.org/97447 | 17:23 |
| devananda | romcheg: ack | 17:23 |
| *** jdob has quit IRC | 17:24 | |
| *** jdob has joined #openstack-ironic | 17:24 | |
| lucasagomes | time to go, have a good night everybody | 17:25 |
| *** lucasagomes is now known as lucas-dinner | 17:26 | |
| devananda | g'night, lucasagomes ! | 17:26 |
| *** pcrews has quit IRC | 17:27 | |
| *** pcrews has joined #openstack-ironic | 17:29 | |
| *** bmahalakshmi has joined #openstack-ironic | 17:33 | |
| *** jdob has quit IRC | 17:39 | |
| *** jdob has joined #openstack-ironic | 17:39 | |
| romcheg | devananda: I'm back | 17:48 |
| devananda | romcheg: hi! have a moment to update me on the progress to upgrade testing? | 17:48 |
| romcheg | sure | 17:49 |
| romcheg | The problem I found is that I forgot to move certain IPMI params to instance_info. Working on that right now | 17:50 |
| romcheg | Or you mean grenade? | 17:51 |
| devananda | both | 17:52 |
| devananda | romcheg: you should read https://review.openstack.org/#/c/95025/ | 17:53 |
| devananda | this spec describes the work we need to do | 17:53 |
| romcheg | So as I said, I'm not familiar to grenade. Probably someone should take a look at that in order to do that asap | 17:54 |
| jroll | devananda: it seems PXE driver does not power off the node on deploy failures? | 17:54 |
| jroll | devananda: is that an oversight or intentional? | 17:55 |
| romcheg | Taking a look at the spec | 17:55 |
| devananda | jroll: humm... i think that's a bug, but also, there may be some history as to why, which i'm forgetting | 17:56 |
| jroll | devananda: ok, thanks | 17:56 |
| jroll | devananda: this is just what I'm seeing in code, I haven't actually ran it :) | 17:57 |
| romcheg | devananda should be really add that to the Nova codebase? | 17:57 |
| openstackgerrit | Devananda van der Veen proposed a change to openstack/ironic: Clarify doc: API is admin only https://review.openstack.org/100974 | 17:58 |
| devananda | romcheg: add what? | 17:58 |
| romcheg | devananda: Migration tools. Right now the specification says that it should be in Nova. I thought Ironic have to provide it | 17:59 |
| devananda | romcheg: ah. interesting point. | 17:59 |
| romcheg | In general it's not a problem to move that technically | 18:00 |
| devananda | romcheg: we should at least be developing it now, regardless of where it lives | 18:00 |
| *** igordcard has quit IRC | 18:00 | |
| devananda | romcheg: and actually it may make sense to put it in a third repo somewhere, since it needs to consume BOTH noav and ironic libs | 18:00 |
| devananda | i'll poke the nova team to see what they are expecting // what they did with cinder and neutron here | 18:01 |
| romcheg | I'll continue working with in in Ironic then. At some point we will move it and resolve deps | 18:01 |
| devananda | ++ | 18:01 |
| romcheg | So please give me some time today/tomorrow. I have my wedding this Saturday so spent last two days in different bureaucratic places | 18:03 |
| romcheg | I will update the first tool which migrates bm-nodesfrom nova | 18:04 |
| jroll | devananda: totally missed a single line of code, it does power the node off | 18:08 |
| jroll | devananda: in _set_failed_state | 18:08 |
| devananda | romcheg: ack. please keep folks updated via the etherpad -- i just added a section at the bottom of the whiteboard | 18:09 |
| devananda | romcheg: we need to track progress on this closely over the next month | 18:10 |
| devananda | romcheg: also - have fun at the wedding!! | 18:10 |
| romcheg | Thank you. I'll be posting updates there | 18:10 |
| romcheg | Thanks! | 18:10 |
| jroll | romcheg: congrats, have fun :) | 18:10 |
| devananda | jroll: cool | 18:10 |
| devananda | ok, gotta run out for an hour ... bbiab | 18:11 |
| rloo | romcheg: are you getting married?! | 18:11 |
| *** dtantsur|afk is now known as dtantsur | 18:11 | |
| dtantsur | jroll, I'm not doing anything more on image caching, except for my patch under review | 18:11 |
| jroll | dtantsur: my selfish side says :( | 18:12 |
| jroll | dtantsur: but I'll take care of it :) | 18:12 |
| dtantsur | btw, hi, devananda :) | 18:12 |
| dtantsur | thanks, jroll :) | 18:12 |
| jroll | dtantsur: I'm just thinking of moving out the _cache_tftp_images, _fetch_images, etc, out | 18:12 |
| jroll | because I just copied a bunch of that code into agent.py :) | 18:13 |
| dtantsur | jroll, I'm pretty sure it should be more or less common code | 18:13 |
| jroll | indeed | 18:13 |
| romcheg | rloo: Yup | 18:25 |
| rloo | romcheg: whoo hoo. Big Congratulations!!! | 18:25 |
| romcheg | rloo: thank you! | 18:25 |
| JayF | I want ironic to /never log anything ever/ without the node uuid :x | 18:26 |
| openstackgerrit | Ruby Loo proposed a change to openstack/ironic: Add 'context' parameter to get_console_output() https://review.openstack.org/100768 | 18:30 |
| romcheg | Gah… Seems like publishing a review wipes unpublished comments in another one. | 18:37 |
| *** pelix has quit IRC | 18:39 | |
| *** ellenh has quit IRC | 18:43 | |
| *** sysexit has quit IRC | 18:49 | |
| *** igordcard has joined #openstack-ironic | 18:52 | |
| *** bmahalakshmi has quit IRC | 18:54 | |
| openstackgerrit | Nisha Agarwal proposed a change to openstack/ironic-specs: Discover node properties at node-create/node-update https://review.openstack.org/100951 | 18:54 |
| *** jistr has quit IRC | 18:55 | |
| *** blamar has quit IRC | 18:57 | |
| rloo | any core reviewer with a few minutes - I wiped out Yuriy's +2 by updating the commit msg. Easy one: https://review.openstack.org/#/c/100768/ | 19:05 |
| *** openstackgerrit has quit IRC | 19:06 | |
| *** openstackgerrit_ has joined #openstack-ironic | 19:07 | |
| *** openstackgerrit_ is now known as openstackgerrit | 19:08 | |
| *** ellenh has joined #openstack-ironic | 19:09 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Updated from global requirements https://review.openstack.org/96228 | 19:35 |
| * devananda is back | 19:40 | |
| Shrews | hi devananda | 19:41 |
| devananda | Shrews: hi! | 19:41 |
| *** blamar has joined #openstack-ironic | 19:43 | |
| devananda | adam_g: looks like https://review.openstack.org/#/c/97475/ is ready to merge to stable/icehouse | 19:45 |
| *** Chandra has joined #openstack-ironic | 19:48 | |
| Chandra | Hi, Has anyone come across this error. Error: /etc/iscsi/initiatorname.isci does not contain valid initator name | 19:50 |
| devananda | NobodyCam: ^ did you put up a patch similar to that recently? | 19:52 |
| Chandra | Devananda this is Chandra from ES. Robert provided the patch to fix iLO issues but 4 out of 5 times the server hits this error ironic fails | 19:53 |
| Chandra | ES=HP Enterprise Services | 19:53 |
| Chandra | iLO fix = ignore iLO and just log a message | 19:53 |
| Chandra | once this error is show the server is unusable | 19:54 |
| lifeless | Chandra: we've not seen this. Whats the content of that file ? | 19:56 |
| *** ellenh has quit IRC | 20:00 | |
| *** sysexit has joined #openstack-ironic | 20:01 | |
| *** dwalleck_ has joined #openstack-ironic | 20:02 | |
| Chandra | it is pushing linux operating system. it is just a text file InitiatorName=iqn.1993-08.org.debian:01:408c6a497fc4 | 20:02 |
| Chandra | this the content on under cloud | 20:02 |
| rloo | devananda: how backwards-compatible do we need to be? baremetal needs to be able to migrate to ironic, but do we need to support (or migrate) ironic-icehouse to ironic-juno? | 20:03 |
| lifeless | Chandra: so, I don't have a /etc/iscsi/initiatorname.isci on any of my systems | 20:04 |
| devananda | rloo: yep | 20:04 |
| rloo | devananda: eg, I thought I saw one patch where some config options got moved from default to <something>. And stuff in driver_info -> instance_info? | 20:04 |
| lifeless | Chandra: I only have an iscsid.conf | 20:04 |
| jroll | rloo: there's a deprecated_group parameter for config options, and we have a migration script for moving things to instance_info | 20:05 |
| lifeless | oh, hang on, let me look on a different machine | 20:05 |
| lifeless | ah there it is | 20:05 |
| *** dwalleck has quit IRC | 20:05 | |
| *** blamar has quit IRC | 20:05 | |
| *** dwalleck_ has quit IRC | 20:06 | |
| lifeless | Chandra: is it possible you have another UC node running somewhere? | 20:06 |
| lifeless | Chandra: from the look of it we should be making that file unique, which golden images won't automatically do. | 20:06 |
| rloo | jroll, devananda: ok, so for configs, we deprecate and remove at some point in the future. for others (in db) we migrate. | 20:06 |
| lifeless | Chandra: by somewhere, I mean on the same LAN | 20:06 |
| Chandra | nope | 20:06 |
| Chandra | it is just 5 machines in isolated VLAN. out of this 5. One is uCloud and other 4 are for over cloud | 20:07 |
| rloo | jroll, devananda: we need to document too, eg driver_info[pxe_root_gb] -> instance_info[root_gb] so people can update their code. | 20:07 |
| devananda | rloo: so there are two types of upgrades we need to support, in addition to a migration from baremetal | 20:07 |
| Chandra | and seed VM is in same network of coruse | 20:07 |
| devananda | rloo: stable release -> next stable release | 20:07 |
| jroll | rloo: the config option actually moves to a different group, but has a comment like # Deprecated group/name - [DEFAULT]/sql_max_overflow | 20:07 |
| devananda | rloo: point in the past (whether stable or arbitrary point in trunk) -> current trunk | 20:07 |
| jroll | rloo: if there's a *new* config, something like: | 20:08 |
| jroll | # This option is deprecated and may be removed in a future | 20:08 |
| jroll | rloo: (as far as I know, I don't run this project :P) | 20:08 |
| * rloo now wishes we didn't have a stable release of ironic yet ;) | 20:10 | |
| devananda | rloo: re: document, yes. we should begin producing more operator documentation, including upgrade docs, a bit later in this cycle | 20:10 |
| devananda | rloo: heh | 20:10 |
| devananda | rloo: so we dont need to do ^ | 20:10 |
| devananda | from teh TC perspective, a project doesn't need to provide an upgrade path until after graduation | 20:10 |
| devananda | that's based on the assumption that a project's "graduation" means the project is stable and ready to use | 20:11 |
| devananda | which is not what "graduation" means in OpenStack -- it means it is part of the integrated gate and integrated release cycle | 20:11 |
| devananda | we, as a project, can be stable before that | 20:11 |
| devananda | and I think we need to be | 20:11 |
| rloo | ok, so we/you, the first ironic 'stable' is icehouse, right? | 20:11 |
| devananda | cause folks who are goign to use Ironic will care. and really, taht's us (and other teams who work for the people who pay us) | 20:11 |
| *** radsy has joined #openstack-ironic | 20:12 | |
| *** radsy has joined #openstack-ironic | 20:12 | |
| devananda | rloo: yes | 20:12 |
| devananda | rloo: stable beta | 20:12 |
| rloo | and for ironic havana, we'll need : migration (for db), deprecated (for configs), and doc for other changes. | 20:12 |
| devananda | rloo: havana? no | 20:12 |
| lifeless | Chandra: can you check that the initiatorname in the seed and UC is different ? | 20:12 |
| rloo | sorry, juno! | 20:12 |
| devananda | rloo: ah. yes :) | 20:13 |
| devananda | rloo: the db migration code should be easy | 20:13 |
| rloo | for docn: am only thinking this in case I get ding'd to work on it. Wonder if we should start an etherpad. or do the docimpact thing. or ?? | 20:13 |
| devananda | we have migrations every time a db change is made | 20:13 |
| devananda | for the driver_info -> instance_info patch sets | 20:13 |
| devananda | we should require those db changes to include a data migration script | 20:13 |
| devananda | which extracts and transforms the relevant driver_info bits from a running deployment | 20:14 |
| *** blamar has joined #openstack-ironic | 20:14 | |
| devananda | and include doc'n of how to execute that migration at that point in the code | 20:14 |
| Chandra | this is from seed vm InitiatorName=iqn.1993-08.org.debian:01:7562e5fb4aa0 | 20:14 |
| devananda | there are folks for whom thsi will actually matter | 20:14 |
| Chandra | this is from under cloud InitiatorName=iqn.1993-08.org.debian:01:408c6a497fc4 | 20:15 |
| Chandra | both are different content | 20:15 |
| devananda | rloo: the project needs a doc writer, really ... | 20:15 |
| devananda | rloo: which is not you :) | 20:15 |
| rloo | and we'll need to mention what changed so people can update their code. eg 'pxe_root_gb' moved from driver_info to instance_info and renamed to 'root_gb'. And 'swap_mb' is now 'swap_gb'. | 20:16 |
| Chandra | btw one node in overcloud was built but the other 3 nodes failed | 20:16 |
| devananda | rloo: also, after feature freeze, we should have a jam session to write that | 20:16 |
| devananda | rloo: by reviewing all the completed specs and bugs from juno | 20:16 |
| devananda | rloo: *I think we should ... | 20:16 |
| rloo | devananda: ok. so maybe have people put DocImpact in their commits, and capture that in the specs. So I don't need to worry about trying to collect that info now ;) | 20:17 |
| adam_g | devananda, should https://review.openstack.org/#/c/88336/ merge frst? | 20:18 |
| devananda | adam_g: it technically can't | 20:18 |
| devananda | adam_g: even though, yes, it should | 20:18 |
| adam_g | okay, actually it shouldn't matter. | 20:19 |
| devananda | adam_g: because the tempest tests for 88336 will fail until the ksclient fixes merge | 20:19 |
| devananda | rloo: I like the ability to grep the git log for DocImpact. however, AIUI, that's supposed to be used to alert the openstack-docs team, which it's not doing right now (since they dont have any docs for ironic yet, anyway) | 20:21 |
| adam_g | devananda, okay ksc fix is +A'd | 20:21 |
| rloo | devananda: oh yeah, you're right, that's what DocImpact does. I suspect they'll ignore it so it should be fine? At some point, they won't ignore it, and we should just get in the habit of using it. | 20:22 |
| *** foexle has joined #openstack-ironic | 20:23 | |
| devananda | rloo: that's probably also a fair approach | 20:24 |
| adam_g | devananda, do you reckon the tempest API tests would pass against the pxe_ipmi driver if driver details were made available to the test (ipmi addr/user/passwd) | 20:24 |
| devananda | adam_g: using a real driver there would make it more than an API test, though | 20:26 |
| devananda | adam_g: passing valid IPMI credentials to the API in a smoke test is going to result in the conductor *actually* doing something t othat hardware | 20:26 |
| adam_g | devananda, right. gahhh | 20:26 |
| devananda | yep | 20:26 |
| devananda | adam_g: also, passing *inavlid* IPMI credentials will cause it to do things, too. like display errors in the API results | 20:27 |
| devananda | eg, it'll try to check the power state, and time out. or send UDP packets flying around the network... | 20:27 |
| Chandra | robert any ideas where I should look? | 20:27 |
| *** blamar has quit IRC | 20:28 | |
| *** lazy_prince has quit IRC | 20:31 | |
| devananda | jroll: ya'll want to do autodiscovery with IPA, I believe, but that isn't specifically covered by any of your open specs? | 20:33 |
| devananda | jroll: you may want to look at https://review.openstack.org/#/c/100951/5/specs/juno/discover_hardware_properties.rst | 20:35 |
| jroll | devananda: yeah, right now for us it's hand-wavy "one day we'll do this somehow" | 20:37 |
| *** theDavidAiken has quit IRC | 20:38 | |
| devananda | jroll: ack. | 20:39 |
| * jroll looks | 20:39 | |
| devananda | jroll: i think we will need to see a spec descibing how IPA will do it and compare to this proposal for iLO discovery | 20:40 |
| devananda | jroll: and compare the APIs proposed by both | 20:40 |
| devananda | jroll: before approving either one | 20:40 |
| devananda | jroll: iow, not likely in juno -- but good to keep in the back of our collective minds | 20:40 |
| jroll | devananda: right, it's totally different methods, we want the agent to push to ironic, iLO is calling for the reverse | 20:41 |
| devananda | right | 20:42 |
| devananda | jroll: and I think that IPA should support a user-initiated "go interrogate this hardware" ability | 20:42 |
| jroll | I agree | 20:42 |
| devananda | jroll: however the user interaction is going t obe totally different here, based on driver | 20:42 |
| devananda | jroll: iLO looks like it will be able to do discovery w/o interupting workload (or even powering on) | 20:43 |
| devananda | well, i'd like the API to be the same. even if the time to get a result is different | 20:43 |
| jroll | devananda: indeed, that's a downside to our approach | 20:44 |
| jroll | I will definitely review that | 20:44 |
| devananda | jroll: expected limitation of inband approach to things | 20:44 |
| devananda | thanks | 20:44 |
| jroll | and maybe I can make a spec for our approach next week | 20:44 |
| devananda | jroll: also, some of that may be retrievable via standard IPMI | 20:44 |
| jroll | devananda: indeed, but I have no idea as to what is and isn't available there :) | 20:46 |
| devananda | that reminds me, linggao - it'd be great to get your input on https://review.openstack.org/#/c/100951/5/specs/juno/discover_hardware_properties.rst as well | 20:47 |
| devananda | linggao: iirc, you had proposed something similar using pyghmi, before we moved to the specs repo | 20:47 |
| linggao | devananda, I'll take a look. | 20:49 |
| linggao | devananda, there was blueprint proposed by Hao Meng, https://blueprints.launchpad.net/ironic/+spec/ipminode-auto-discovery | 20:56 |
| *** blamar has joined #openstack-ironic | 20:56 | |
| devananda | linggao: ah! that's what i was thinking of. thanks | 20:57 |
| linggao | devananda, we discussed it internally. For mac address, ipmi will return all the nics available on the node. The out-of-band discovery cannot decide which one is the primary install nic. | 20:58 |
| linggao | devananda, and I saw someone wrote on the white board for this bp: "The blueprint seems to be missing an important point: it seems not possible to retrieve disk layout/capacity only with IPMI (AFAIK, in the general case)" | 20:59 |
| adam_g | russell_h, RE: https://review.openstack.org/#/c/99770/ , what sort of callbacks does the agent driver need from ramdisks where this might be useful? im looking at the pxe ramdisk + vendor passthru now and wondering if it makes sense to try to replace that or not | 21:00 |
| linggao | devananda, currently, I think Ironic has in-band discovery built in in the deploy agent within ramdisk. (at least we discussed it before) | 21:01 |
| JayF | adam_g: (responding without context of the spec) the entire agent deploy works on callbacks | 21:01 |
| JayF | adam_g: you tell teh agent 'do a deploy' then when it's done, on the next heartbeat, it tells ironic it's done | 21:01 |
| linggao | devananda: and the new python deploy agent (teeth-agent) should implement this capability. | 21:01 |
| devananda | linggao: indeed | 21:01 |
| devananda | linggao: i believe jroll is going to start drafting a spec for how ironic-python-agent will perform node discovery, but he could use your help (or help from haomeng) with the ipmi portions of it | 21:02 |
| devananda | jroll: ^^ | 21:02 |
| linggao | devananda, sure. glad to contribute. | 21:03 |
| adam_g | JayF, and that currently relies entirely on the vendor passthrough API? | 21:03 |
| JayF | adam_g: I believe so, yes | 21:03 |
| JayF | adam_g: because heartbeats work via vendor_passthru | 21:03 |
| *** jdob has quit IRC | 21:04 | |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Adding swift temp url support https://review.openstack.org/81391 | 21:04 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Factor out TFTPImageCache https://review.openstack.org/100734 | 21:04 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Factor out deploy info from PXE driver https://review.openstack.org/100735 | 21:04 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Add methods to ipmitool driver https://review.openstack.org/100364 | 21:04 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Add ironic-python-agent deploy driver (DO NOT MERGE) https://review.openstack.org/101020 | 21:04 |
| jroll | wheeeeeee | 21:04 |
| jroll | devananda, linggao, I am going to start that next week | 21:04 |
| jroll | (probably) | 21:05 |
| adam_g | JayF, ah, cool. found the relevant code in the agent. thanks | 21:05 |
| JayF | devananda: news you will care about -> had a long chat in #-infra today, and it looks like we're going to do the builds of IPA... in the check | 21:05 |
| linggao | jroll, do you have a spec for it? | 21:05 |
| JayF | devananda: like using -infra stuff, because they have images that run trusty as of about 5 minutes ago | 21:05 |
| jroll | linggao: I mean, I am probably going to start on a spec next week | 21:06 |
| linggao | jroll, cool. | 21:06 |
| jroll | linggao: but it's not a high priority for me right now, so please don't count on that :| | 21:06 |
| jroll | I will *try* | 21:07 |
| *** dtantsur is now known as dtantsur|afk | 21:07 | |
| linggao | jroll, :) | 21:08 |
| jroll | devananda: this is an attempt to make a *basic* agent driver that is similar to the PXE driver. no new functionality, etc. mind skimming and pointing out anything crazy I'm doing? https://review.openstack.org/101020 | 21:08 |
| devananda | JayF: so dib builds the deploy ramdisk within devstack today | 21:10 |
| devananda | JayF: which is less than fantastic, since that takes a while | 21:10 |
| devananda | JayF: i'd much rather have an infra job that produces the IPA disk image, and then the functional test of ironic+ipa just downloads the latest build which passed all tests | 21:11 |
| JayF | devananda: the idea is that we'd add a check + gate job to IPA that tests to ensure it can build the coreos image, and builds+uploads the built agents to tarballs.o.o | 21:11 |
| devananda | JayF: is that what ya'll discussed? | 21:11 |
| devananda | JayF: right. and that job tests "does this image work with current ironic + ipa-driver" before uploading the tarball | 21:12 |
| devananda | JayF: and we co-gate all three projects | 21:12 |
| JayF | devananda: As a step two, yes. | 21:12 |
| openstackgerrit | Alex Weeks proposed a change to openstack/ironic-specs: Add metric reporting spec. https://review.openstack.org/100729 | 21:12 |
| JayF | devananda: right now I want to solely scope myself to building an agent for others to test | 21:12 |
| linggao | jroll, what is the major different between the agent derive and pxe driver? | 21:12 |
| JayF | devananda: then when jroll and others finish their work to get IPA working in devstack, they can possibly integrate with those downloaded ones | 21:13 |
| jroll | linggao: the ramdisk | 21:13 |
| JayF | devananda: just trying to bite off one days' worth of work at a time :) | 21:13 |
| devananda | JayF: ack. so you are focusing on the disk image building parts specifically? or IPA? or both | 21:13 |
| jroll | linggao: the ramdisk exposes a rest API, in which we will eventually have "secure erase disks", "update firmware", etc | 21:13 |
| JayF | devananda: I'm going to work, on Friday, on the part to build an image based on git://ipa/imagebuild/coreos with every 'check' and 'gate' for the agent | 21:14 |
| linggao | jroll, for now, the *basic* part, are they identical to the pxe driver? | 21:14 |
| *** romcheg has quit IRC | 21:14 | |
| linggao | I mean the flow. | 21:14 |
| JayF | devananda: once that exists, I suspect jroll and dwalleck will leverage it to make testing via devstack easier | 21:14 |
| devananda | JayF: great | 21:15 |
| jroll | linggao: doorbell, hold that thought | 21:16 |
| linggao | jroll, like for example, ramsidk expose the iscsi, then the agent driver dd the image on the iscsi. | 21:16 |
| devananda | linggao: that flow is very different | 21:16 |
| devananda | linggao: eg, IPA downloads the image from glance directly. no iscsi | 21:16 |
| linggao | devananda, that's great improvement, for scaling. | 21:17 |
| devananda | that's the hope :) | 21:18 |
| openstackgerrit | Alex Weeks proposed a change to openstack/ironic-specs: Add metric reporting spec. https://review.openstack.org/100729 | 21:20 |
| *** martyntaylor has quit IRC | 21:20 | |
| aweeks | ^ I haven't addressed all of the comments yet, but most of them with that patch | 21:21 |
| Shrews | adam_g: ping | 21:24 |
| adam_g | Shrews, hey | 21:24 |
| Shrews | adam_g: on 99519, should we not do the same thing for rebuild() ? | 21:24 |
| Shrews | adam_g: i.e., catch InstanceDeployFailure and cleanup | 21:25 |
| jroll | linggao: so yeah, the flow is similar but a bit different. we just pass an image URL to the agent and it downloads and writes to disk | 21:26 |
| adam_g | Shrews, hmm. im not as familiar with instance rebuilds.. is that the expected behavior on a failed rebuild? does the instance get kicked back to the scheduler for a retry on another node? | 21:26 |
| linggao | jroll, way to go! thanks. | 21:27 |
| devananda | adam_g: an interesting devstack change, when you have a minute to look - https://review.openstack.org/#/c/98317/ | 21:27 |
| Shrews | adam_g: i don't believe that it does | 21:27 |
| JayF | linggao: and as a note, this isn't theory. It works. Today. Pretty well, in fact, considering it's pre-pre-alpha software, haha | 21:28 |
| linggao | JayF, how do you build the images that has IPA in the ramdisk? | 21:29 |
| adam_g | devananda, yeah haven't had time to play with that. im also interested to know if / how https://review.openstack.org/#/c/92257/ affects us wrt stable testing | 21:29 |
| devananda | adam_g: oh, hum. dib has releases. I thought we were going to add it to irnoic/requirements.txt to resolve that? | 21:30 |
| linggao | jroll, what protocal does the IPA use to download the images? http? | 21:30 |
| adam_g | devananda, yeah. but if devstack is installing from a local checkout (master) and ironic installs a specific version from its requirements, which one is actually used? | 21:30 |
| jroll | linggao: today, http. in the future, who knows :) | 21:30 |
| devananda | adam_g: if dib doesn't have a stable *branch* then I think that patch is likely eventually to break stable branch testing of ironic | 21:31 |
| jroll | linggao: initial thoughts are bittorrent and multicast | 21:31 |
| adam_g | Shrews, so destroy() will make the node in question available to nova again for provisioning new instances | 21:31 |
| devananda | adam_g: switching to infra to ask that question of folks there | 21:31 |
| JayF | linggao: https://github.com/openstack/ironic-python-agent/tree/master/imagebuild | 21:31 |
| adam_g | devananda, actually this is a knowledge gap for me. the same thing happens with client libraries and requierments | 21:32 |
| JayF | linggao: it's our assumption that at some point we'll also get dib elements in there, although since we aren't going to be using them, we're a bit hesistant to spend time writing them | 21:32 |
| devananda | adam_g: hm, ok. not switching to infra then. | 21:32 |
| devananda | adam_g: so I *think* that requirements are synced in part to solve this problem | 21:33 |
| Shrews | adam_g: i think that should probably apply for rebuild(), then | 21:33 |
| devananda | mordred may be able to tell us more about that | 21:33 |
| openstackgerrit | A change was merged to openstack/ironic: Cleanup virt driver tests and verify final spawn https://review.openstack.org/99518 | 21:33 |
| devananda | adam_g: my assumption is that all the stable branches of all the projects now have the same versions of things in their requirements files (may not have always been true) | 21:33 |
| linggao | JayF, thanks for the link. Does it just build the ramdisk? | 21:34 |
| adam_g | devananda, right. | 21:34 |
| JayF | linggao: yeah, there's some info about how in the readme under coreos/ | 21:34 |
| devananda | adam_g: so installing dib master while testing stable/ironic is awkward | 21:35 |
| JayF | linggao: I intend on working Friday to get a builder working for those in infra, now that there's support for Trusty + docker | 21:35 |
| devananda | adam_g: i think if dib wants to be a service that devstack installs, it should have branches | 21:35 |
| devananda | since that's how devstack controls what verions of its services are started | 21:35 |
| adam_g | devananda, one sec, let me check somethign in my devstack env | 21:35 |
| linggao | JayF, so the end goal is to use this image builder to build the deplpy ramdisk and use diskimage-builder to build the rest (user ramdisk, kernel and whole image)? | 21:37 |
| JayF | linggao: well, as far as images, that's kinda an exercise for the deployer? You point nova at an image in glance, and it's passed all the way through | 21:37 |
| JayF | linggao: for building the ramdisk that ipa runs in, we at Rackspace intend on using CoreOS like under imagebuild/coreos for any agents we run on physical hardware | 21:38 |
| jroll | linggao: just a note, today IPA only supports deploying whole disk images | 21:38 |
| *** mrda-away is now known as mrda | 21:38 | |
| mrda | Morning ironic! | 21:39 |
| mordred | what did I do? | 21:39 |
| *** romcheg has joined #openstack-ironic | 21:39 | |
| devananda | mordred: question on testing of stable/XXX branches of things in devstack | 21:40 |
| mordred | kk | 21:40 |
| devananda | mordred: what if a "service" doesn't have a stable/XX branch? | 21:40 |
| devananda | but we want to test stable/icehouse | 21:40 |
| devananda | of ironic, nova, etc | 21:40 |
| mordred | it falls back to master | 21:40 |
| devananda | ok | 21:40 |
| devananda | so diskimage-builder doesn't have a stable branch | 21:40 |
| devananda | afaik they dont plan to have one | 21:40 |
| mordred | yah. so you'll get master of dib | 21:41 |
| mordred | same with python-novaclient, for instance | 21:41 |
| devananda | but they're releasing pips | 21:41 |
| devananda | and are in requirements.txt | 21:41 |
| devananda | so we should use the versioned pip from requirements, not install the service from git | 21:41 |
| linggao | JayF, jrol, have you seen this sepc? https://review.openstack.org/#/c/97150/. If IPA only supports whole disk image, then we do not need this feature. | 21:41 |
| devananda | right? | 21:41 |
| linggao | JayF, jroll: or I miss something. | 21:42 |
| JayF | linggao: I don't really know much at all about the pxe driver or use cases. I've not run the pxe driver so don't have a lot of familiarity. | 21:42 |
| jroll | linggao: still, someone may wish to use whole disk images with the PXE driver. also, we plan to support not-whole-disk-images. maybe. | 21:42 |
| jroll | s/maybe/probably/ | 21:42 |
| linggao | JayF, jroll. thanks for all the info. Nice talking to you. | 21:45 |
| jroll | linggao: ofc :) | 21:46 |
| jroll | linggao: I also plan to write up a thing about our environment and how we run ironic in the near future | 21:46 |
| jroll | what patches are working, etc | 21:46 |
| linggao | jroll, that will be very helpful. | 21:46 |
| jroll | :) | 21:47 |
| linggao | got to go. bye folks. | 21:47 |
| devananda | jroll: i'd love to see that | 21:47 |
| devananda | linggao: ciao! | 21:47 |
| *** linggao has quit IRC | 21:47 | |
| jroll | devananda: I'll ping you when it's up | 21:48 |
| jroll | devananda: do you think that be better as a wiki page, or blog post, or? | 21:48 |
| *** romcheg has quit IRC | 21:48 | |
| devananda | jroll: blog | 21:48 |
| devananda | jroll: it represents a point-in-time | 21:48 |
| devananda | wikis are graveyards for info, not snapshots | 21:48 |
| jroll | devananda: sure, but I'd also love to keep it up to date | 21:49 |
| devananda | jroll: then wiki or etherpad | 21:49 |
| jroll | devananda: sort of, how we run ironic and IPA and you can too! | 21:49 |
| jroll | ok, great. thanks :) | 21:49 |
| JayF | jroll: thanks for doing that | 21:49 |
| mrda | jroll: +1 | 21:49 |
| *** sysexit has quit IRC | 21:49 | |
| * JayF ^5 jroll | 21:49 | |
| jroll | damn, I didn't actually mean to commit that hard :P | 21:49 |
| devananda | jroll: i still think a blog would be great, even with a footer saying "and we'll keep this>> page updated as things evolve" | 21:49 |
| devananda | or something | 21:49 |
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic: Destroy instance to clear node state on failure https://review.openstack.org/99519 | 21:49 |
| jroll | devananda: sure. | 21:50 |
| jroll | I've been meaning to start blogging nyway | 21:50 |
| jroll | anyway* | 21:50 |
| adam_g | devananda, 99519 needed a rebase | 21:50 |
| jroll | oh, devananda, did you see my ping to take a quick look at https://review.openstack.org/#/c/101020/ and make sure it's not insane? | 21:57 |
| devananda | jroll: ooh, i missed that | 21:59 |
| * devananda adds that to review queue | 21:59 | |
| jroll | devananda: goal is to be as close as possible to pxe driver there | 21:59 |
| devananda | jroll: awesome | 21:59 |
| devananda | jroll: also, please link the bp in the commit msg | 21:59 |
| jroll | devananda: I will | 21:59 |
| jroll | devananda: unit tests before good commit messages :) | 21:59 |
| jroll | (tests exist but were not committed as I broke them all) | 22:00 |
| *** blamar has quit IRC | 22:05 | |
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic: Add diskimage-buidler to requirements.txt https://review.openstack.org/101038 | 22:06 |
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic-specs: External event callback API https://review.openstack.org/99770 | 22:12 |
| devananda | adam_g: so actually i think we should move the deploy-ironic element out of dib | 22:12 |
| devananda | adam_g: that's the only part of dib which we really depend on -- that's what goes *in* the image. we dont care how the image is actually built | 22:12 |
| devananda | adam_g: also, global-requirements needs a version pin for dib | 22:12 |
| devananda | orw hat ever we move deploy-ironic element to | 22:13 |
| adam_g | devananda, https://review.openstack.org/#/c/100385/ | 22:13 |
| devananda | \o/ | 22:13 |
| adam_g | devananda, but the deploy-ironic element depends on other stuff in dib | 22:13 |
| devananda | adam_g: we'll need to do the same things for IPA eventually (queue the conversation JayF and I had earlier) | 22:14 |
| JayF | IPA already vendors the image builders in our codebase. | 22:15 |
| devananda | vendors? | 22:15 |
| adam_g | devananda, are the IPA elements going to stay in the IPA source or go into dib/tie? | 22:15 |
| devananda | my brain fails to verb that word | 22:15 |
| devananda | JayF: ^ ? | 22:15 |
| JayF | vendoring the dependency... aka the imagebuilder is in the codebase | 22:15 |
| JayF | imagebuild/coreos right now, I'm assuming imagebuilder/dib-elements will exist in the future as well | 22:15 |
| devananda | JayF: cool | 22:15 |
| JayF | I never understood why I'd want a dependency of product a (ironic/pxe driver) in the repo of product be (dib/pxe deploy ramdisk) | 22:16 |
| devananda | so eg, you should be able to take any dib version (assuming they dont ever break backwards compat, whic hthey swear they wont) | 22:16 |
| devananda | and build the stuff in IPA/imagebuilder/dib-elements with it | 22:16 |
| devananda | JayF: we have that all the time in openstack. | 22:16 |
| JayF | I'm sure they promise that. If it were up to me I'd probably lock in a version, but I'm paranoid and have been burned before by 'we won't break backwards compat' promises :) | 22:16 |
| adam_g | the DIB interface/framework may stay stable but the elements themselves will evolve over time | 22:18 |
| devananda | adam_g: right | 22:18 |
| devananda | ironic is dependent upon the elements, no tthe DIB framework | 22:18 |
| JayF | In the case of IPA, I don't think we even want to proscribe what image to use. DIB is an option. The CoreOS image we use is an option. Some other idea that someone is willing to upstream and maintain is an option. Building your own using an external repo is okay. | 22:19 |
| devananda | adam_g: all the deps have meregd. want to officialy open stable/icehouse? https://review.openstack.org/#/c/88336/ | 22:20 |
| JayF | Images/ramdisks are just environments that software run in, and I'm not really keen on the idea that we should all use the same environments to run that software. | 22:20 |
| adam_g | devananda, nice | 22:21 |
| devananda | JayF: and yet we will need testing to run on something, and, by virtue of that being the thing we test the most, it will be seen as being proscribed | 22:21 |
| devananda | JayF: even if it's not | 22:21 |
| JayF | That's why we'll probably end up supporting DIB in IPA, and will also provide CI for the CoreOS image builder :) | 22:22 |
| devananda | JayF: ++ | 22:22 |
| JayF | I just want to replace DIB with containers and a docker-like build system (we actually use docker for this in IPA right now) | 22:22 |
| JayF | but my job is training secret agents to provision hardware, so I'll keep doing that for now ;) | 22:23 |
| JayF | Downside of open source w/code review: no sneaking in easter-egg API endpoints that return an image of a secret agent man | 22:23 |
| Shrews | rloo: ping! | 22:30 |
| Shrews | rloo: left a comment on 100571. past dinner time for me now :) | 22:36 |
| *** theDavidAiken has joined #openstack-ironic | 22:41 | |
| *** ellenh has joined #openstack-ironic | 22:47 | |
| openstackgerrit | Jay Faulkner proposed a change to openstack/ironic-python-agent: Fix invalid command in Dockerfile https://review.openstack.org/101046 | 22:48 |
| JayF | ^ easiest review you'll ever have jroll JoshNang russell_h others | 22:48 |
| jroll | holy shit, how did we land that | 22:50 |
| JayF | It's almost certainly my fault from a code-writers standpoint | 22:51 |
| JayF | but it would cause no failures because -f | 22:52 |
| JoshNang | oh wow | 22:53 |
| JoshNang | k, merging it | 22:53 |
| openstackgerrit | A change was merged to openstack/ironic-python-agent: Fix invalid command in Dockerfile https://review.openstack.org/101046 | 22:55 |
| JayF | TTM: 8 minutes | 22:57 |
| JayF | that might be an openstack record ;) | 22:57 |
| jroll | JayF: should look at TTM on pyghmi, there's approximately one main developer and one core reviewer, best I can tell (and those are the same person) | 23:00 |
| *** igordcard has quit IRC | 23:03 | |
| rloo | Shrews: thx! I'll reply on the patch. | 23:05 |
| *** Haomeng has joined #openstack-ironic | 23:07 | |
| *** radsy has quit IRC | 23:14 | |
| *** theDavidAiken has quit IRC | 23:15 | |
| *** Isotopp has quit IRC | 23:17 | |
| *** mdorman has quit IRC | 23:18 | |
| *** mordred has quit IRC | 23:24 | |
| *** radsy has joined #openstack-ironic | 23:25 | |
| *** radsy has joined #openstack-ironic | 23:25 | |
| *** Isotopp has joined #openstack-ironic | 23:25 | |
| *** Isotopp has quit IRC | 23:25 | |
| *** Isotopp has joined #openstack-ironic | 23:25 | |
| *** mordred has joined #openstack-ironic | 23:27 | |
| *** mordred has quit IRC | 23:30 | |
| *** mordred has joined #openstack-ironic | 23:30 | |
| devananda | SpamapS: were you seeing iscsi device is busy errors on hardware at some point? | 23:32 |
| devananda | JayF: clearly you need more tests then :) | 23:32 |
| JayF | devananda: it's more about rm -rf not complaining about pretty much anything. That's what -f is for :) | 23:34 |
| SpamapS | devananda: I don't remember ever seeing that on hardware no | 23:41 |
| lifeless | the iscsi intiator thing above turned out to be time skew | 23:42 |
| lifeless | I wonder if the deploy ramdisk should try to do NTP | 23:43 |
| lifeless | filing a bug on ironic about it | 23:43 |
| lifeless | since it needs to be documented at minimum | 23:43 |
| jroll | lifeless: it should :) | 23:43 |
| jroll | lifeless: we've ran into time skew issues ourself | 23:43 |
| JayF | FWIW the CoreOS IPA image does time syncing by default :) | 23:45 |
| devananda | lifeless: so there's https://review.openstack.org/#/c/90126/12 | 23:46 |
| lifeless | jroll: https://bugs.launchpad.net/tripleo/+bug/1331862 | 23:47 |
| devananda | ok - just got a call, need to run out and look at another apartment | 23:47 |
| devananda | bbi2h | 23:47 |
| lifeless | devananda: thats different but also interesting | 23:47 |
| jroll | lifeless: cool | 23:47 |
| JayF | jroll: fwiw our clock skew problems were conductor vs swift with temp_urls | 23:48 |
| jroll | ah, yeah | 23:49 |
| *** yfujioka has joined #openstack-ironic | 23:49 | |
| *** lucas-dinner has quit IRC | 23:53 | |
| Shrews | rloo: gah, tried the quickstart guide stuff for the 3rd time and I still get NodeLocked | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!