| opendevreview | Verification of a change to openstack/bifrost master failed: Revert "Pin systemd to 252-15 for CentOS Stream 9" https://review.opendev.org/c/openstack/bifrost/+/891014 | 00:09 |
|---|---|---|
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM Enable OVN https://review.opendev.org/c/openstack/ironic/+/885087 | 00:18 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM Enable OVN https://review.opendev.org/c/openstack/ironic/+/885087 | 00:28 |
| iurygregory | mmalchuk, just fyi normally we try to avoid bare rechecks, since the job timed out in the patch you can just add recheck job timed_out =) | 01:38 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM Enable OVN https://review.opendev.org/c/openstack/ironic/+/885087 | 03:07 |
| opendevreview | Verification of a change to openstack/bifrost master failed: Revert "Pin systemd to 252-15 for CentOS Stream 9" https://review.opendev.org/c/openstack/bifrost/+/891014 | 03:14 |
| TheJulia | goodnight | 04:14 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Retry connecting vmedia through a DVD device if available https://review.opendev.org/c/openstack/ironic/+/888746 | 05:49 |
| arne_wiebalck | Good morning, Ironic! | 06:57 |
| rpittau | good morning ironic! o/ | 07:15 |
| arne_wiebalck | good morning rpittau o/ | 07:24 |
| arne_wiebalck | rpittau: quick question, for https://review.opendev.org/c/openstack/ironic-python-agent/+/891609 and the failure in metalsmith integration, I don't think this is related to the patch itself, but would appreciate a more experienced set of eyes :) | 07:26 |
| arne_wiebalck | rpittau: would you have a moment to take a brief look? | 07:27 |
| mmalchuk | morning Ironic o/ | 07:29 |
| rpittau | hey arne_wiebalck :) | 07:30 |
| rpittau | arne_wiebalck at a glance the issue seems related to the tinycore setup, so not related to the patch itself | 07:30 |
| rpittau | I would recheck in a couple of hours | 07:30 |
| rpittau | I feel like metalsmith has been a bit abandoned, we're quite overdue for its upgrade to CS9, had a patch open but no time to finish it https://review.opendev.org/c/openstack/metalsmith/+/869374 :/ | 07:32 |
| arne_wiebalck | rpittau: thanks! | 07:32 |
| rpittau | np! | 07:32 |
| opendevreview | Riccardo Pittau proposed openstack/metalsmith master: [WIP] Add centos9 based job https://review.opendev.org/c/openstack/metalsmith/+/869374 | 07:33 |
| rpittau | at least rebased it, let's see how it goes | 07:33 |
| dtantsur | rpittau: overdue is a bit an overstatement, CS8 only expires in mid-2024 | 08:02 |
| rpittau | true, I was comparing to the rest of our projects | 08:03 |
| mmalchuk | rpittau tinyipa keystone integration failed everywhere. looks like not systemd issue? | 09:08 |
| mmalchuk | on CI | 09:09 |
| mmalchuk | folks, please merge: https://review.opendev.org/c/openstack/bifrost/+/890789 | 09:11 |
| opendevreview | Dmitry Tantsur proposed openstack/bifrost master: testenv: use host-model/host-passthru as a CPU model for VMs https://review.opendev.org/c/openstack/bifrost/+/891693 | 11:34 |
| dtantsur | zigo: hey, do you have any threads to pull to get https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1002025 fixed? We may start to discourage people from using Debian-based ramdisks otherwise.. | 11:37 |
| iurygregory | good morning Ironic | 11:55 |
| rpittau | mmalchuk: everywhere no, in jammy, just rechecked the patch | 12:10 |
| mmalchuk | rpittau ok, waiting for updates | 12:16 |
| mmalchuk | thanks | 12:16 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Retry connecting vmedia through a DVD device if available https://review.opendev.org/c/openstack/ironic/+/888746 | 12:39 |
| frickler | dtantsur: seems there is a newer version in bookworm and bullseye-backports, can you check if that has the issue, too? https://packages.debian.org/bullseye-backports/lshw | 12:51 |
| dtantsur | I guess we don't have backports enabled in IPA | 13:18 |
| TheJulia | good morning | 13:28 |
| TheJulia | dtantsur: perhaps we need nordix's job reporting to us in a 3rd party CI mode... | 13:32 |
| dtantsur | I doubt they have resources for that | 13:32 |
| dtantsur | morning! | 13:32 |
| mmalchuk | morning! | 13:42 |
| mmalchuk | folks, please review and merge: https://review.opendev.org/c/openstack/bifrost/+/890789 | 13:42 |
| TheJulia | arne_wiebalck: o/ how many conductors do you run these days? | 13:43 |
| TheJulia | dtantsur: posted comments on why update_node, lmk if that makes sense. I critically don't want internal tasks dying if we can help it at all | 13:44 |
| TheJulia | otherwise the risk of workflow ops derailing is quite high | 13:44 |
| dtantsur | State changes....... | 13:45 |
| TheJulia | yup | 13:45 |
| TheJulia | we have to record | 13:46 |
| dtantsur | TheJulia: I wonder if we can hack it further to only care for provisioning state changes. I'd prefer external operations (like ones from inspector) not to be retried forever. | 13:46 |
| TheJulia | I finally see what you were talking about w/r/t to logic though | 13:46 |
| dtantsur | but yeah, somehow protecting taskmanager stuff makes sense to me | 13:46 |
| TheJulia | maybe lets try incremental towards that, add a todo if see if we can sanely do that | 13:46 |
| dtantsur | I'm afraid of making things worse if e.g. a call from Inspector locks a node virtually forever.. but then again, it will probably fail to release the lock, so *shrug* | 13:47 |
| dtantsur | let's go for it, I guess | 13:47 |
| TheJulia | well, that sort of happens today | 13:48 |
| TheJulia | well, the lock was held and not released which blocked inspector | 13:48 |
| TheJulia | at least on the first logs, second one at a cursory look looks similar because it started at the same basic time after power state sync | 13:48 |
| arne_wiebalck | TheJulia: o/ | 13:53 |
| arne_wiebalck | TheJulia: let me check ... | 13:53 |
| arne_wiebalck | TheJulia: 21 CGs with 25 conductors | 13:54 |
| arne_wiebalck | TheJulia: for ~9000 nodes | 13:54 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Retool sqlite retries https://review.opendev.org/c/openstack/ironic/+/891333 | 14:25 |
| opendevreview | Elod Illes proposed openstack/ironic-lib master: DNM: dummy change to test gate health https://review.opendev.org/c/openstack/ironic-lib/+/891789 | 14:57 |
| opendevreview | Elod Illes proposed openstack/metalsmith master: DNM: dummy change to test gate health https://review.opendev.org/c/openstack/metalsmith/+/891790 | 14:57 |
| opendevreview | Elod Illes proposed openstack/networking-generic-switch master: DNM: dummy change to test gate health https://review.opendev.org/c/openstack/networking-generic-switch/+/891793 | 14:58 |
| opendevreview | Elod Illes proposed openstack/python-ironic-inspector-client master: DNM: dummy change to test gate health https://review.opendev.org/c/openstack/python-ironic-inspector-client/+/891823 | 15:07 |
| opendevreview | Verification of a change to openstack/bifrost master failed: Revert "Pin systemd to 252-15 for CentOS Stream 9" https://review.opendev.org/c/openstack/bifrost/+/891014 | 15:11 |
| opendevreview | Elod Illes proposed openstack/networking-baremetal master: DNM: dummy change to test gate health https://review.opendev.org/c/openstack/networking-baremetal/+/891842 | 15:23 |
| dtantsur | TheJulia: we have a few more instances reported at https://github.com/metal3-io/cluster-api-provider-metal3/issues/1082 | 15:38 |
| JayF | If after we're done, someone doesn't do a summit talk with a clever title like "Unlocking SQLite in Ironic", we've failed ;) | 15:45 |
| TheJulia | https://review.opendev.org/c/openstack/ironic/+/891335 is growing on me, but it looks like in one of the logs it all started in the window between getting a lock to begin with , and then loosing it to do the actual node_update.... | 15:47 |
| opendevreview | Merged openstack/ironic master: Add a list of children nodes unit test https://review.opendev.org/c/openstack/ironic/+/889746 | 15:50 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Add some service steps https://review.opendev.org/c/openstack/ironic/+/891630 | 15:53 |
| TheJulia | dtantsur: is there a way to look at the logs and get the hash for the ironic code ? | 15:54 |
| TheJulia | so we can line that up to the state of the repo? | 15:54 |
| TheJulia | or does it just build with whatever is on master branch when it runs | 15:54 |
| TheJulia | i guess I want to understand if https://review.opendev.org/c/openstack/ironic/+/887835 is present or not | 15:57 |
| TheJulia | hmm, both instances are actually rooted in an update_node request from the api | 16:02 |
| rpittau | good night! o/ | 16:11 |
| * TheJulia has idea | 16:21 | |
| mmalchuk | TheJulia now you can push the button, all branches passed CI | 16:33 |
| mmalchuk | https://review.opendev.org/q/Iab88b3a28378cc22f767204bef84128bc9627210 | 16:33 |
| mmalchuk | thanks in advance | 16:33 |
| mmalchuk | and https://review.opendev.org/c/openstack/bifrost/+/890789 | 16:34 |
| mmalchuk | too | 16:34 |
| JayF | no need to ping individuals :) | 16:34 |
| JayF | we're all looking | 16:34 |
| dtantsur | TheJulia: ironic may log its version on start-up.. otherwise, only inspecting quay.io/metal3-io/ironic | 16:34 |
| mmalchuk | JayF understand but TheJulia did it when you stop she | 16:35 |
| JayF | mmalchuk: the stable/2023.1 backport of that is on the way to the gate | 16:35 |
| mmalchuk | JayF thanks | 16:35 |
| TheJulia | dtantsur: got a few minutes to chat? | 16:35 |
| dtantsur | TheJulia: sure | 16:35 |
| TheJulia | https://meet.google.com/ycx-qhav-kmd | 16:36 |
| mmalchuk | JayF is here a best/correct way to ping all cores? | 16:36 |
| JayF | mmalchuk: if you populate the hashtag field of your patch with ironic-week-prio | 16:38 |
| JayF | and link it in here asking for review without pinging anyone | 16:38 |
| JayF | it should be a timely review, especially if it's V+1 and is a trivial backport like that was | 16:38 |
| mmalchuk | any example? | 16:38 |
| JayF | I know, at least for me personally, I read almost 100% of the messages put in here, even when I've been away :) | 16:38 |
| JayF | mmalchuk: for the hashtag; just click "Show More" at the top of the change, and the field will show up: https://home.jvf.cc/~jay/hashtag.png | 16:40 |
| JayF | mmalchuk: just hit the edit-pencil beside hashtags and enter 'ironic-week-prio' | 16:40 |
| JayF | then your change shows up in our review dashboard: https://review.opendev.org/q/hashtag:%22ironic-week-prio%22+(status:open) | 16:40 |
| mmalchuk | oh cool, no one other project uses hashtag there) lol | 16:41 |
| JayF | It's very much a project-by-project thing, I think other projects use hashtags for other things | 16:41 |
| JayF | it's just a tag field that was created when everyone called tags hashtags ;) | 16:41 |
| mmalchuk | yep, but not for reviews) | 16:42 |
| JayF | yeah, and it's not like, you won't get reviewed without that tag | 16:42 |
| JayF | it's just a way of saying "this is ready to go" | 16:42 |
| JayF | My order of reviewing is generally: updating reviews on patches I already reviewed; unreviewed stuff in ironic dash; everything else | 16:43 |
| mmalchuk | I like it | 16:45 |
| JayF | lol mmalchuk I think I accidentally landed zed before :/ 2023.1 going to gate now | 16:51 |
| JayF | don't worry, I've already reported my rule violation to myself, and I've been punished severely 😂 | 16:52 |
| JayF | ah no, I landed a *different* 2023.1 change | 16:54 |
| JayF | your hashtag paying off already :P | 16:54 |
| dtantsur | TheJulia: to illustrate my thoughts, this part caches stuff, and I don't understand what and why: https://opendev.org/openstack/oslo.db/src/branch/master/oslo_db/sqlalchemy/enginefacade.py#L1012-L1055 | 17:03 |
| TheJulia | likewise :( | 17:05 |
| dtantsur | :( | 17:05 |
| dtantsur | anyway, long weekend here, talk to y'all on Monday | 17:05 |
| TheJulia | o/ | 17:05 |
| mmalchuk | JayF lol | 17:07 |
| TheJulia | ... wow, I think i just speeded up our unit tests locally | 17:07 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Don't hold on to the API request dbapi https://review.opendev.org/c/openstack/ironic/+/891872 | 17:16 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM: Test - Revert "Don't actually heartbeat with sqlite" https://review.opendev.org/c/openstack/ironic/+/891874 | 17:22 |
| TheJulia | lets get metal3 to lock! | 17:22 |
| JayF | TheJulia: how does that recover if e.g. a connection is lost to the DB? | 17:25 |
| JayF | TheJulia: is the client smart enough to make the existing instance work or would we fail a request | 17:26 |
| TheJulia | it is supposed to reconnect | 17:26 |
| JayF | nice | 17:27 |
| TheJulia | in oslo.db's realm | 17:27 |
| TheJulia | but we shall see | 17:27 |
| TheJulia | I'm just surprised how much faster unit test runs are locally | 17:27 |
| TheJulia | which means "there was way more going on there" | 17:27 |
| JayF | yeah, not caching connections is a very strange thing to have lived this long | 17:27 |
| TheJulia | well, it also gets lazy loaded basically | 17:31 |
| opendevreview | Merged openstack/bifrost stable/2023.1: Create the log file for the disk-image-create command https://review.opendev.org/c/openstack/bifrost/+/891124 | 19:20 |
| mmalchuk | merged... the next one can be +W | 19:39 |
| JayF | refresh it my friend | 19:39 |
| JayF | you can't beat me to the punch during working hours ;) | 19:39 |
| * JayF is very very empathetic to being frustrated at waiting for a thing to backport | 19:40 | |
| * JayF looks at https://review.opendev.org/c/openstack/oslo.messaging/+/891742 | 19:41 | |
| mmalchuk | yep, I can't do that, I'm not core( | 19:41 |
| JayF | oh, I know you can't, but I'm just saying I try to get the stuff landed when someone cares about it | 19:41 |
| JayF | b/c it's no fun to wait | 19:41 |
| mmalchuk | I'm care for all my work) | 19:41 |
| mmalchuk | and for your too | 19:42 |
| JayF | What is your professional interest in OpenStack/Ironic? | 19:42 |
| JayF | aka where do you work lol | 19:42 |
| mmalchuk | I'm Cloud Architect in the Bank | 19:42 |
| JayF | and have you gotten any Ironic socks+stickers yet | 19:42 |
| JayF | I work finance-adjacent as well | 19:43 |
| * JayF works for https://opensource.gresearch.com | 19:43 | |
| opendevreview | Verification of a change to openstack/bifrost stable/2023.1 failed: Remove extra symbols accidentally added https://review.opendev.org/c/openstack/bifrost/+/890789 | 20:21 |
| JayF | hmm, I don't see bifrost CI enough to know if that's likely a real err or not | 20:22 |
| JayF | given the change hard to see it being related | 20:22 |
| JayF | but I'll leave it to someone with more confidence to recheck | 20:23 |
| mmalchuk | its random | 20:25 |
| mmalchuk | same job passed on test but failed on gate | 20:26 |
| JayF | So, in this case you're probably right; but we try to avoid corrolating it exactly like that | 20:27 |
| JayF | bceause imagine we introduce a failure that happens 25% of the time, if we just recheck because it's the same job that already passed, we eventually get in recheck hell | 20:28 |
| JayF | every openstack project has been their due course of time in recheck hell (or at least recheck purgatory) where you can't merge anything until you pay off all the "random test failure" tech debt | 20:28 |
| mmalchuk | we need to fix CI | 20:29 |
| mmalchuk | make it more stable | 20:29 |
| mmalchuk | rpittau know how | 20:29 |
| JayF | For master, we're doing about as well as we have in a long time | 20:29 |
| JayF | bifrost was impacted by a breakage but we knew what it was and it was being fixed; that's pretty good lol | 20:29 |
| JayF | for the stable/ branches, they often get less CI love keeping them stable which isn't ideal but there's only so many of us | 20:29 |
| JayF | I'll note that operators who run openstack in production have all the skills needed to troubleshoot failing CI jobs :P | 20:30 |
| TheJulia | hmmm | 20:31 |
| mmalchuk | can we get the ssh access to the CI job hosts to debug? | 20:31 |
| TheJulia | so seems we can't use tftp at all with ovn | 20:31 |
| JayF | TheJulia: why not? | 20:34 |
| JayF | mmalchuk: generally speaking, all the logs you need should be right there. Click on the job results, then go to Artifacts or Logs tab, and all the service logs are there | 20:34 |
| JayF | mmalchuk: when something is extra-breaky, we often ask the zuul admins to hold a node for us when it fails so we can ssh in; but that's a last resort sorta thing. | 20:35 |
| TheJulia | It seems like OVN's NAT is side stepping the main kernel stack | 20:35 |
| TheJulia | so the connection tracker helper does nothing | 20:35 |
| JayF | Can we disable firewalling for CI purposes? | 20:36 |
| TheJulia | nope, because it is natting the network | 20:38 |
| JayF | oh, yeah, of course | 20:38 |
| JayF | hmmm | 20:38 |
| JayF | multinode, so we can't just yolo-yolo forward all possible ports | 20:38 |
| JayF | you sure OVS isn't just operating in a different network namespace? it really completely avoids the nf stack with raw sockets? damn | 20:38 |
| TheJulia | stack@np0034976153:~/devstack$ sudo ip netns list | 20:39 |
| TheJulia | ovnmeta-825eb402-b4f1-4a60-a435-5ebb9e260f99 (id: 0) | 20:39 |
| TheJulia | stack@np0034976153:~/devstack$ | 20:39 |
| TheJulia | totally different concept modeling :( | 20:39 |
| JayF | this is borderline infuriating :( | 20:41 |
| JayF | and I've not been banging head against it like you | 20:41 |
| JayF | print out a logo and burn it in effagy for stress relief, it'll help :| | 20:41 |
| TheJulia | heh | 20:41 |
| JayF | It drives me bonkers when I pretty much fully understand a thing and someone just reimplements it and goes around the thing | 20:42 |
| JayF | TheJulia: is there any value in running OVN on our jobs that can use vmedia? | 20:42 |
| TheJulia | eh, might work | 20:43 |
| JayF | if we don't currently, we probably could support vmedia in sushy-tools | 20:43 |
| TheJulia | trying to see if ther eis a "I don't want nat" option | 20:43 |
| opendevreview | Merged openstack/bifrost stable/zed: Create the log file for the disk-image-create command https://review.opendev.org/c/openstack/bifrost/+/891125 | 21:02 |
| mmalchuk | next could be yoga change | 21:06 |
| JayF | go check the timestamp, W+2'd two minutes before you got to it | 21:08 |
| JayF | you're not going to win the race with me mmalchuk ;) | 21:08 |
| * JayF just found https://docs.openstack.org/ironic/latest/contributor/index.html | 21:09 | |
| mmalchuk | lol | 21:09 |
| JayF | with even more links to different devstack configs | 21:09 |
| JayF | /o\ | 21:10 |
| mmalchuk | devstack? meh | 21:10 |
| mmalchuk | only pure openstack on baremetal! | 21:10 |
| * JayF looks around | 21:10 | |
| JayF | I don't have any BMCs whatsoever in my domicile :) | 21:10 |
| mmalchuk | you are working for finance... you always can bought some) | 21:13 |
| JayF | Don't tempt me with a good time | 21:14 |
| mmalchuk | I will) | 21:14 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM Enable OVN https://review.opendev.org/c/openstack/ironic/+/885087 | 21:16 |
| TheJulia | That might work more... although it seems tftp transfers like to stall on the CI host I'm holding | 21:17 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Don't hold on to the API request dbapi https://review.opendev.org/c/openstack/ironic/+/891872 | 21:29 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM: Test - Revert "Don't actually heartbeat with sqlite" https://review.opendev.org/c/openstack/ironic/+/891874 | 21:29 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: WIP: Revamp Ironic contributor docs, esp. around testing env https://review.opendev.org/c/openstack/ironic/+/891912 | 21:42 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM Enable OVN https://review.opendev.org/c/openstack/ironic/+/885087 | 21:54 |
| iurygregory | JayF, not sure if you already saw https://review.opendev.org/c/openstack/python-ironicclient/+/891560 =) | 22:04 |
| JayF | good stuff, ty | 22:04 |
| JayF | assume you've tested it well? | 22:04 |
| JayF | if you tell me you actually ran all the commands you can get a speedyjay review lol | 22:04 |
| iurygregory | I tested the logic only lol | 22:06 |
| iurygregory | didn't have a setup to test it (sorry) | 22:06 |
| JayF | whomp whomp | 22:06 |
| iurygregory | going to try next week | 22:06 |
| JayF | cool; yeah I looked at it and it looks OK | 22:07 |
| JayF | but for me "does it work" is the ultimate question for CLI lol | 22:07 |
| iurygregory | fighting dhcp to deploy a real bm with bifrost in the internal network :D (when I get this working I will be able to test) | 22:07 |
| iurygregory | or maybe I will just add some support in sushy-tools and test via VM =) | 22:08 |
| iurygregory | at least this would enable CI testing... if I update sushy-tools | 22:08 |
| JayF | I guess it's not a valid test to just run it, ensure the payload is right, and get the err | 22:09 |
| JayF | b/c you gotta ensure the resp payload is filled | 22:09 |
| JayF | iurygregory: instead of sushy-tools; maybe wire enough of it up in fake driver for testing? | 22:10 |
| JayF | we should use fake driver as a fixture more often for things like this | 22:10 |
| opendevreview | Merged openstack/bifrost stable/yoga: Create the log file for the disk-image-create command https://review.opendev.org/c/openstack/bifrost/+/891127 | 22:22 |
| JayF | mmalchuk: I think that's the end of the line; https://review.opendev.org/c/openstack/bifrost/+/891134 is not in a mergable state right now. If you need it, get it fixed up and I'll look tomorrow. | 22:26 |
| iurygregory | JayF, maybe but probably metal3 will need something that can be emulated to test it | 22:55 |
| iurygregory | but I think I added enough tests in the fake-firmware interface if I recall correctly | 22:55 |
| zigo | dtantsur: I happen to co-maintain the lshw package in Debian, this annoyed me a lot in the past as well, but https://bugs.debian.org/1002025 was fixed a long time ago, please upgrade to Bookworm ! :) | 23:30 |
| zigo | Alternatively, you can use the backports in one of my OpenStack unofficial debian.net repos: http://osbpo.debian.net/debian/dists/ | 23:31 |
| zigo | I believe I have a backport for at least bullseye-victoria and up, but probably even in buster. | 23:31 |
| JayF | zigo: PTI for 2024.1 will include Debian oldstable still, in order to facilitate upgrades (we don't really have to respect that in IPA though) | 23:34 |
| JayF | zigo: sorry for using that name instead of number, I'm terrible at remembering number | 23:34 |
| zigo | JayF: Include in what? | 23:36 |
| JayF | here, I'll link it | 23:36 |
| JayF | https://review.opendev.org/c/openstack/governance/+/891225 | 23:36 |
| JayF | we still have to test next release on debian 11 | 23:36 |
| JayF | to ensure people can upgrade since we have 11 supported in this release | 23:36 |
| zigo | JayF: So next release, you'll test on both Deb 11 and 12 ? | 23:39 |
| zigo | From my view point, you're only 1 year late to match when I release Ironic in Debian ! :) | 23:40 |
| zigo | Zed was the release I supported both 11 and 12, starting with Antelope, I care only about Bookworm. | 23:40 |
| JayF | That is exciting and also always makes me a little nervous because our minions don't test those exact combos :D | 23:41 |
| JayF | Did you ever figure out all the bits to get your full-debian-style ironic stuff working, including IPA w/a debian-style ramdisk instead of a DIB one? | 23:42 |
| zigo | I got something funky here, trying to fix https://bugs.debian.org/1046841 : it seems some build process of Ironic are touching my gettext stuff in the Debian folder (ie: the debconf templates and translations), generating a not-wanted diff ... | 23:43 |
| zigo | I can't find what's doing that ... :/ | 23:43 |
| zigo | Sorry, I really meant: | 23:43 |
| zigo | https://bugs.debian.org/1046505 | 23:43 |
| zigo | JayF: Nearly, I still have the issue that Ironic doesn't like my "https://example.com/ironic-is-here-not-at-the-root-with-a-port-number/" type of endpoint ... | 23:45 |
| zigo | Appart from that, everyting seems working ... :/ | 23:45 |
| JayF | Can you make sure there's a bugs.launchpad.net ironic bug for that? With a really good reproduction case? | 23:46 |
| JayF | and for bonus points, email me the bug at jay at gr-oss dot io or CC me on it and I'll put it on my list | 23:46 |
| zigo | JayF: I traced it to a generator iterating the wrong way, and then got lost in the ironic client object model ... :( | 23:47 |
| zigo | But I'll try to find the time to try again. | 23:47 |
| JayF | yeah, just brain dump into a bug and someone may be able to follow up; especially if you are willing to draft a blogpost about it for ironicbaremetal.org :) | 23:47 |
| JayF | you don't have to solve it just show the reproduction and if you have leads share them :) | 23:48 |
| zigo | JayF: You can have fun with this though... https://salsa.debian.org/openstack-team/debian/supply/ | 23:48 |
| zigo | (that's my PoC ...) | 23:48 |
| zigo | That's how to solve it. | 23:48 |
| JayF | I'm past the end of my usual workday so I don't really have the capacity to look at that closely right now | 23:48 |
| JayF | I'll see if I can have a look tomorrow | 23:48 |
| zigo | s/solve/reproduce/ | 23:48 |
| zigo | It need instructions on how to deploy, I'm aware of that ... :/ | 23:49 |
| zigo | I will fill the bug. | 23:49 |
| JayF | thanks! | 23:50 |
| zigo | JayF: FYI, my "supply" thingy can setup a (nearly) working Ironic setup as a VM in any OpenStack cloud with one single command ... :P | 23:50 |
| JayF | yeah, that's basically what bifrost is set to do -- it's fun to see the different ways you can arrange the legos though | 23:51 |
| zigo | JayF: No, Bitfrost does only Ironic, I'm doing Keystone, Nova, Barbican, Neutron... | 23:51 |
| zigo | All of them setup at once in a single command, deployed with puppet. | 23:51 |
| zigo | :) | 23:51 |
| zigo | (forgot glance...) | 23:52 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
