| opendevreview | Verification of a change to openstack/ironic stable/2023.2 failed: [CI][stable only] fix zuul config https://review.opendev.org/c/openstack/ironic/+/926637 | 00:43 |
|---|---|---|
| opendevreview | Merged openstack/ironic-inspector master: Fix versions in release notes https://review.opendev.org/c/openstack/ironic-inspector/+/926994 | 00:49 |
| opendevreview | Merged openstack/ironic-lib stable/2024.1: Fix invalid UTF-8 characters in execute output https://review.opendev.org/c/openstack/ironic-lib/+/926716 | 00:54 |
| opendevreview | Merged openstack/ironic-lib stable/2023.2: Fix invalid UTF-8 characters in execute output https://review.opendev.org/c/openstack/ironic-lib/+/926718 | 00:54 |
| opendevreview | Verification of a change to openstack/ironic-lib stable/2023.1 failed: Fix invalid UTF-8 characters in execute output https://review.opendev.org/c/openstack/ironic-lib/+/926719 | 00:54 |
| *** dtantsur_ is now known as dtantsur | 01:48 | |
| opendevreview | OpenStack Proposal Bot proposed openstack/ironic master: Imported Translations from Zanata https://review.opendev.org/c/openstack/ironic/+/927699 | 03:54 |
| opendevreview | Derek Higgins proposed openstack/ironic master: Set node "alive" when inspection finished https://review.opendev.org/c/openstack/ironic/+/927828 | 08:25 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-tempest-plugin master: Check inspection data and abortion in the standalone tests https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/927928 | 12:15 |
| opendevreview | Merged openstack/ironic master: Imported Translations from Zanata https://review.opendev.org/c/openstack/ironic/+/927699 | 13:16 |
| TheJulia | dtantsur: if you could today take a quick glance and see my comments on https://review.opendev.org/c/openstack/ironic-python-agent/+/927823, I think as is I could likely just add a reno and move it along, but wanted to see what you think before doing so. | 13:18 |
| TheJulia | Thanks in advance | 13:18 |
| * dtantsur puts on his queue | 13:19 | |
| TheJulia | Much appreciated! | 13:19 |
| TheJulia | I also noted an idea we could use to detect a conflict, but it would be a PITA | 13:23 |
| TheJulia | and require the ramdisk to use a loopabck | 13:23 |
| * TheJulia walkies the doggo | 13:25 | |
| *** vicent_ is now known as vicent | 13:38 | |
| *** vicent is now known as vicentfg | 13:38 | |
| *** vicentfg is now known as vicent | 13:38 | |
| TheJulia | Iury sent me a photo and comment from the summit, sounds like it is going well | 13:53 |
| TheJulia | or did to be more precise | 13:53 |
| TheJulia | cid: quick question on https://review.opendev.org/c/openstack/ironic-python-agent/+/926973 | 14:00 |
| cid | TheJulia: Actually, I ripped those right out of the log in the bug description. | 14:06 |
| cid | Now that you've mentioned it, let me quickly check | 14:06 |
| TheJulia | o/ rosmaita | 14:07 |
| rosmaita | good morning! | 14:07 |
| TheJulia | cid: much appreciated | 14:07 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927965 | 14:08 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/26.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927966 | 14:08 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/25.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927967 | 14:08 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2024.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927968 | 14:09 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/24.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927969 | 14:09 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.2: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927970 | 14:09 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927972 | 14:09 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent master: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927974 | 14:10 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2024.1: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927976 | 14:10 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2023.2: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927978 | 14:10 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2023.1: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927979 | 14:10 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.13: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927981 | 14:11 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.12: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927983 | 14:11 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.9: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927984 | 14:11 |
| JayF | rosmaita: ^ I think thaqt is all that you need? | 14:11 |
| rosmaita | think so ... will let you know in a few minutes | 14:12 |
| TheJulia | It didn't list the unmaintained ones | 14:12 |
| JayF | yeah those go to um | 14:12 |
| TheJulia | https://review.opendev.org/q/I7fac5c64f89aec39e9755f0930ee47ff8f7aed47 | 14:12 |
| JayF | or, uh, nowhere? | 14:12 |
| JayF | Weird, must have gotten turned off from -unmaintained and -stable too | 14:12 |
| JayF | that's unfortunate, I was using those to trigger reviews | 14:12 |
| rosmaita | ok, i should be able to find them in gerrit | 14:13 |
| JayF | I'm going to push the IPA UM patches now too, but they don't need to go in the bug or any text since it's just a warning. | 14:14 |
| rosmaita | ok | 14:14 |
| JayF | Apparently xena UM still hasn't had the gitreview merged in, yikes | 14:19 |
| JayF | I stacked mine on top of t hat | 14:19 |
| JayF | same with v + w | 14:21 |
| TheJulia | looks like unmatained yoga and xena for ironic are not in happy states | 14:22 |
| JayF | TheJulia: looks like mine is failing codespell in IPA; I'm going to fix and repush :( just two words | 14:23 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent master: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927974 | 14:23 |
| dtantsur | We can review and merge all this stuff now, right? | 14:24 |
| JayF | yes, please | 14:24 |
| TheJulia | yes | 14:24 |
| dtantsur | okay, I'm listening to something on a meeting, should be free in a few minutes | 14:25 |
| JayF | the idea is we get as much of it merged as possible before we put the official notice out in a couple hours | 14:25 |
| JayF | yeah, CI gonna have to run so no real rush as long as it's in the next 30-45m | 14:25 |
| dtantsur | This meeting makes wanna flit the table, so I might need to re-assemble it first :) | 14:26 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.13: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927981 | 14:27 |
| JayF | we have codespell voting on bugfix/9.13 and master; I fixed the patch both places and am leaving it elsewhere | 14:27 |
| dtantsur | Unrelated side note: I have problems with IRC recently, so it's perfectly possible for me to miss pings | 14:27 |
| JayF | I can follow up on stable branches to get codespell re-passing, but it looks like stable/2024.1 needs some config as well to get it passing, so I'm punting that until post-CVE | 14:28 |
| JayF | fungi: can you make the queued jobs on 927974,2 start running? https://zuul.opendev.org/t/openstack/status#927974 | 14:31 |
| TheJulia | dtantsur: joy | 14:31 |
| JayF | TheJulia: https://zuul.opendev.org/t/openstack/status#927965 master ironic patch failing tempest-funcional-python3 and tox-cover (which is post_Failure so probably not you?) | 14:31 |
| JayF | looking at tempest logs | 14:32 |
| TheJulia | yup, downside of dropping a ton of patches on CI at once | 14:32 |
| JayF | > 2024-09-04 14:27:14.723231 | controller | The specified regex doesn't match with anythingERROR: InvocationError for command /opt/stack/tempest/.tox/tempest/bin/tempest run --regex ironic_tempest_plugin.tests.api --concurrency=1 (exited with code 1) | 14:32 |
| TheJulia | we just DoS-ed ourselves | 14:32 |
| JayF | that doesn't look like DoS, that looks like real-failure | 14:33 |
| TheJulia | well, in general, we just slowed everything down | 14:33 |
| dtantsur | JayF: oh, so it's not just on my patch?? | 14:35 |
| dtantsur | see #openstack-qa | 14:35 |
| JayF | oh no | 14:35 |
| TheJulia | Did we suddenly just stop merging plugins?! | 14:35 |
| TheJulia | err | 14:35 |
| TheJulia | merging is the wrong word | 14:35 |
| TheJulia | pulling in | 14:35 |
| dtantsur | sorry, I should have brought this up, but I was sure I was doing something wrong with my patch.. | 14:35 |
| TheJulia | no worries | 14:36 |
| TheJulia | https://zuul.opendev.org/t/openstack/stream/8e4588458fa7459b9ddfb98d73508a98?logfile=console.log | 14:43 |
| TheJulia | err | 14:43 |
| TheJulia | wrong channel | 14:43 |
| fungi | JayF: taking a look | 14:50 |
| JayF | fungi: dont | 14:50 |
| JayF | fungi: see #openstack-qa | 14:50 |
| JayF | tl;dr tempest is broken | 14:50 |
| fungi | oh fun. well, that's always how it goes, you know? | 14:51 |
| dtantsur | JayF: could you run a quick sanity check with me? | 14:52 |
| dtantsur | https://review.opendev.org/c/openstack/ironic-python-agent/+/927974/2/ironic_python_agent/extensions/standby.py#368 | 14:52 |
| dtantsur | Is it me or do we call qemu-img convert even after dd when is_raw? | 14:52 |
| dtantsur | there is no "else" or return? | 14:52 |
| JayF | #%#@$#@!@#% | 14:53 |
| dtantsur | :( | 14:53 |
| JayF | I'm going to fix it in each patch separately | 14:53 |
| JayF | if I have to rebase these again my brain will leak out my ears | 14:54 |
| dtantsur | understandable | 14:54 |
| dtantsur | you may want to fixup the "cherry-picked from" lines after that | 14:54 |
| JayF | I'm going to be honest, dunno what you mean | 14:55 |
| JayF | can you be 100% explicit with what Ineed to do while I do this change? | 14:55 |
| dtantsur | ah, you don't have them. that's one way :) | 14:55 |
| dtantsur | I mean, normally, you cherry-pick with -x flag to generate "Cherry picked from: <commit ID>" in your commit message | 14:56 |
| TheJulia | dtantsur: cherry-pick lines only apply to clean cherry-picks, almost none of this is clean | 14:56 |
| TheJulia | fwiw | 14:56 |
| dtantsur | mmmm? I always leave them | 14:56 |
| JayF | and also it was done using text patches and am and all | 14:56 |
| dtantsur | anyway, ignore this one. let's focus on important things. | 14:56 |
| JayF | yep | 14:56 |
| TheJulia | AIUI, your supposed to drop them when it is not clean | 14:56 |
| TheJulia | anyway, just a minor detail for tooling which I don't think even gets used in our ecosystem any longer | 14:57 |
| dtantsur | I was under the opposite impression. Interesting. | 14:57 |
| JayF | my understanding == dtantsur's fwiw | 14:57 |
| TheJulia | (HP magical rebase walking tool which used the tagging heavily) | 14:57 |
| JayF | and that you usually put the Conflicts:\n\tfilename stuff | 14:57 |
| dtantsur | TheJulia: `git cherry-pick -x` is neither HPE nor openstack specific though | 14:57 |
| dtantsur | anyway. finishing the sanity check of IPA patches, Ironic up next | 14:57 |
| TheJulia | in our ecosystem, it is not used anymore afaik, but whatever | 14:58 |
| TheJulia | Anyway, I have a meeting/call to jump on in 2 minutes | 14:58 |
| TheJulia | will be slightly distracted, I guess we're pending QA to force push whatever is going on | 14:58 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent master: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927974 | 14:58 |
| JayF | dtantsur: ^ can you have a look before I push that the rest of the way thru | 14:58 |
| dtantsur | looking | 14:59 |
| dtantsur | seems good to me | 14:59 |
| JayF | ty | 14:59 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2024.1: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927976 | 15:02 |
| fungi | JayF: i can enqueue security fixes into the gate pipeline if you want, now that the tempest fix has merged | 15:03 |
| JayF | fungi: start with master? 927974 IPA, 927965 Ironic | 15:04 |
| fungi | are they all patchset #1? i need to know the patchsets, but can look them up if you're not sure whether they got revisions pushed | 15:04 |
| JayF | not for mine, at least | 15:04 |
| JayF | 927974,3 | 15:05 |
| JayF | this does not bypass CI, correct? | 15:05 |
| JayF | just moves it to where it only has to pass once? | 15:05 |
| JayF | Ironic is 927965,1 | 15:05 |
| dtantsur | we do lose non-voting jobs like bifrost in this case | 15:06 |
| fungi | correct | 15:06 |
| fungi | you do lose feedback from jobs that are configured to only run in check | 15:06 |
| JayF | I don't have a strong opinion on if that matters here tbh | 15:06 |
| fungi | or rather get a limited amount of time to react to them really | 15:06 |
| JayF | I'd rather get them in and fix fallout | 15:06 |
| JayF | would rather have a broken, secure bifrost in master than a working, insecure one IMO | 15:06 |
| JayF | but it's not just my call | 15:06 |
| fungi | also make sure the changes have all the necessary votes to go into the gate (code review +2, workflow +1) | 15:07 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2023.2: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927978 | 15:07 |
| JayF | fungi: good call, we're missing those | 15:07 |
| JayF | I think dtantsur is still looking at the ironci change, I just +2d it | 15:07 |
| dtantsur | yeah, gimme a couple of minutes | 15:08 |
| JayF | it's OK, I need to buy you a schnitzel or something for catching that in review for ipa | 15:09 |
| * dtantsur does not object to a Schnitzel :D | 15:09 | |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.13: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927981 | 15:09 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.12: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927983 | 15:09 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.9: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927984 | 15:10 |
| JayF | all ipa patches updated for code review | 15:10 |
| JayF | and I learned the syntax for "patch exactly one file" :) | 15:10 |
| JayF | (diff file1 file2; patch file-to-patch mypatch.patch) | 15:11 |
| JayF | I've only ever done the full directory syntax | 15:11 |
| fungi | so enqueue --project=openstack/ironic-python-agent --change=927974,3 and --project=openstack/ironic --change=927965,1 at the moment? | 15:11 |
| JayF | looking to see if they have the votes | 15:11 |
| JayF | TheJulia: https://review.opendev.org/c/openstack/ironic-python-agent/+/927974/ needs your +2A | 15:11 |
| JayF | and we're waiting for Dmitry to finish on the Ironic patch | 15:12 |
| JayF | and he caught something in the IPA one, so good to get a review in | 15:12 |
| * JayF going through stable/um/bugfix ironic patches and reviewing | 15:12 | |
| fungi | oh cool | 15:12 |
| JayF | TheJulia: https://review.opendev.org/c/openstack/ironic/+/927968/1/doc/source/admin/security.rst you have erroneous stuff in your stable patch[es?] | 15:14 |
| JayF | only in the bottom of a doc, I +2'd anyway in case we don't want to lose CI progress | 15:17 |
| dtantsur | TheJulia: 2 quick questions for you as well: https://review.opendev.org/c/openstack/ironic/+/927965/1/ironic/drivers/modules/deploy_utils.py#1455 and https://review.opendev.org/c/openstack/ironic/+/927965/1/ironic/drivers/modules/image_cache.py#369 | 15:17 |
| JayF | I think you found something real in deploy_utils | 15:18 |
| dtantsur | Could be | 15:19 |
| dtantsur | Sigh | 15:19 |
| JayF | and btw; I think the bug is not in requests | 15:19 |
| dtantsur | I guess this proves that I cannot really review code from someone's screen.. | 15:19 |
| JayF | but in like, some keystoneauth middleware | 15:19 |
| JayF | dtantsur: +++++++++++== | 15:20 |
| JayF | dtantsur: secret gerrit when :) | 15:20 |
| dtantsur | Yeah, true, I like that you can work on advisories in github | 15:20 |
| JayF | I know even for me personally, with code I'm writing myself, often I'll push it, self-review in gerrit/github, and find stuff I missed otherwise | 15:21 |
| dtantsur | Happens all the time | 15:21 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2024.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927968 | 15:25 |
| JayF | btw, I think after this experience, we could probably use a refactor around the ipa image handling code | 15:34 |
| JayF | and maybe a rewriting of the ironic image cache too | 15:34 |
| fungi | okay, so just to confirm, there's no need to direct enqueue anything into the gate pipeline just yet | 15:34 |
| JayF | fungi: yeah, I don't think so, Ironic patch has valid review comments and https://review.opendev.org/c/openstack/ironic-python-agent/+/927974/ not yet approved | 15:35 |
| fungi | i'll probably be disappearing in another half hour, but maybe frickler can help if needed | 15:35 |
| TheJulia | dtantsur: replied on https://review.opendev.org/c/openstack/ironic/+/927965/1/ironic/drivers/modules/deploy_utils.py#1455 and porposed an idea, please take a look | 15:35 |
| dtantsur | TheJulia: yeah, "call it again" is probably the simplest thing we can do | 15:36 |
| JayF | TheJulia: will there be something else to catch a redirect loop? | 15:36 |
| JayF | or will that blow up the conductor with a recursive call stack err | 15:36 |
| dtantsur | it's not a loop though: we only handle one redirect, it seems | 15:36 |
| TheJulia | JayF: the underlying code only allows a single redirect | 15:36 |
| TheJulia | and ultimately IPA will catch, we just need to double check how far ipa patches go back | 15:36 |
| dtantsur | (which is weird, but it's the existing state of things) | 15:37 |
| JayF | oh yeah, you're right | 15:37 |
| * JayF was thinking something else that was wrongish | 15:37 | |
| TheJulia | so it can easily be a fix after the needful thing as well | 15:37 |
| JayF | TheJulia: IPA fixes stop at the UM edge, so Zed is the first unpatched IPA | 15:37 |
| TheJulia | I am aware of that | 15:37 |
| JayF | Then what did > we just need to double check how far ipa patches go back < refer to? | 15:38 |
| JayF | the only reason I'm like, -0.5 to a follow-up, is the OSSA references the patch URLs directly | 15:39 |
| JayF | so I think my preference is fixed in place, but I can be convinced otherwise | 15:39 |
| TheJulia | I get what your doing, I get what your saying. I'm a single person righ tnow | 15:39 |
| TheJulia | I already said we need to check where redirect logic went in and then figure out the rest from there which *includes fixing* or *following-up* | 15:40 |
| TheJulia | So, lets just take one step at a time, I'm trying to write a response to dmitry now on his other question | 15:40 |
| TheJulia | one item at a time. | 15:40 |
| TheJulia | dtantsur: replied on https://review.opendev.org/c/openstack/ironic/+/927965/1/ironic/drivers/modules/image_cache.py#369 | 15:45 |
| TheJulia | okay, redirect logic merged in Zed. So need to get fixed and rebased | 15:46 |
| TheJulia | joy | 15:47 |
| TheJulia | that is a painful amount of work | 15:47 |
| TheJulia | dtantsur: would an inline comment on https://review.opendev.org/c/openstack/ironic/+/927965/1/ironic/drivers/modules/image_cache.py bring more clarity to it? | 15:52 |
| JayF | <not-urgent> https://review.opendev.org/c/openstack/ossa/+/928005 was posted by me just now, adding Ironic repos to official VMT oversight. I've marked it W-1 pending review from Ironic cores. | 15:55 |
| fungi | yeah, notably, the repositories are expected to adhere to the requirements enumerated in that file | 15:58 |
| JayF | Yeah, I think the only pieces we'd need is consensus from cores, and the buttons flipped in lp so VMT sees security issues first | 16:01 |
| JayF | I'm putting it on next meeting agenda | 16:02 |
| dtantsur | TheJulia: probably worth it.. What I don't understand: it looks like force_raw=False no longer has effect unless the image inspection is disabled. | 16:03 |
| TheJulia | okay, you re-phrasing it has me feeling the need to look again | 16:04 |
| JayF | IPA tests are failing for *examples* ?!, looking | 16:04 |
| JayF | I'm pretty sure those tests are invalid... | 16:05 |
| JayF | dtantsur: you got a sec to look at this with me? | 16:06 |
| dtantsur | I have 5-10 mins before I need to go afk for some time | 16:06 |
| JayF | dtantsur: I think https://zuul.opendev.org/t/openstack/build/b31966d670cd4d428693c52424c096b4 is failing because it's trying to import the hwms directly | 16:07 |
| TheJulia | dtantsur: okay, yeah, It does kind of look like that but I think it is context. I know with the tests that was the happy path to move stuff foward as well, so I think I feel pretty good there. Anyway, I'm runing the tests with the other changes to see how much testing I need to change/fix/add for the redirect stuffs | 16:07 |
| JayF | dtantsur: but because we need to have the cmd run first to setup the conf, that's not happening, so it kabooms | 16:07 |
| JayF | tl;dr: we would never in the real world import a hwm into a process that didn't /already/ have IPA and it's config setup | 16:07 |
| dtantsur | JayF: it's quite unfortunate that we cannot simply sanity-check the examples.. | 16:08 |
| TheJulia | okay, one test it looks like. Not horrible | 16:08 |
| JayF | dtantsur: I would argue we're not sanity checking them even right now. | 16:08 |
| dtantsur | Possibly. Feel free to disable. | 16:08 |
| dtantsur | TheJulia: force_raw=False not working any more has implications for networking traffic though. Do we at least mention that it does nothing? Potentially even deprecate? | 16:09 |
| dtantsur | * mention in release notes | 16:09 |
| TheJulia | dtantsur: so you realize your looking at the conversion code right? We would have already downloaded it | 16:10 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent master: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927974 | 16:11 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent master: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928011 | 16:11 |
| dtantsur | not sure if my message came through with my bloody internet: | 16:11 |
| dtantsur | I thought the idea behind force_raw=False is to deliver the original format to the node? E.g. if the throughput to the node is very limited and the raw image is very large. | 16:11 |
| dtantsur | it's separate from the question of whether or not we download image to the *conductor* | 16:12 |
| TheJulia | well, we would have already done so, the intent behind force_raw was largely around streaming and keeping it as raw on disk to begin with, so everything still hinges on force_raw, but has two separate conditional paths around image handling for compatability further on | 16:13 |
| dtantsur | force_raw pre-dates streaming by years | 16:13 |
| TheJulia | oh | 16:13 |
| TheJulia | OH | 16:14 |
| TheJulia | I see what yours eeing | 16:14 |
| TheJulia | the intenting is wrong | 16:14 |
| dtantsur | I suspected something like this | 16:14 |
| dtantsur | Thanks for looking into it, now I can do some exercising. Will check with y'all later. | 16:15 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2024.1: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927976 | 16:15 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2024.1: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928012 | 16:15 |
| TheJulia | k, thanks! | 16:16 |
| JayF | dtantsur: if I caught you, we need +2 on new IPA patches I just pushed | 16:17 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2023.2: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927978 | 16:17 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2023.2: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928013 | 16:17 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2023.1: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927979 | 16:19 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent stable/2023.1: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928014 | 16:19 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.13: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927981 | 16:20 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.13: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928015 | 16:20 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.12: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927983 | 16:22 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.12: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928017 | 16:22 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.9: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927984 | 16:23 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.9: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928019 | 16:23 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927965 | 17:06 |
| TheJulia | dtantsur: JayF ^ | 17:06 |
| JayF | ty, lgtm | 17:13 |
| TheJulia | cool cool, I'll roll-out the local diff against the rest of the branches shortly | 17:16 |
| JayF | TheJulia: https://review.opendev.org/c/openstack/ironic-python-agent/+/927974/4 and friends (including the "remove examples job" patches attached to each) need review/approval whenever you get a second. CI still running so not urgent r/n | 17:19 |
| TheJulia | hoping k | 17:20 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/26.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927966 | 17:20 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/25.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927967 | 17:21 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2024.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927968 | 17:22 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/24.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927969 | 17:25 |
| JayF | I put a topic on all the ipa changes needed for this: https://review.opendev.org/q/topic:%22ossa-2024-003%22 | 17:26 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.2: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927970 | 17:28 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927972 | 17:29 |
| opendevreview | Doug Goldstein proposed openstack/sushy master: fix spelling and make codespell pass https://review.opendev.org/c/openstack/sushy/+/927444 | 17:34 |
| TheJulia | toxtopic reset on the ironic patches | 17:37 |
| TheJulia | err | 17:37 |
| TheJulia | topic | 17:38 |
| TheJulia | looks like metal3-integration is failing on stable branches (why is it even there, since that job doesn't grok stable branches?!?) on artifacts in the nordix docs | 17:43 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2024.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927968 | 17:43 |
| dtantsur | TheJulia: it should not run on stable branches, at least for now | 17:43 |
| TheJulia | vi know | 17:43 |
| TheJulia | err | 17:43 |
| TheJulia | i know | 17:43 |
| TheJulia | people don't remove it | 17:44 |
| TheJulia | dtantsur: what about bugfix branches? | 17:45 |
| dtantsur | TheJulia: I don't think the job is smart enough to figure out the required ironic-image branch.. | 17:46 |
| dtantsur | and BMO branch.. | 17:46 |
| TheJulia | k | 17:46 |
| dtantsur | so, not today | 17:46 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/26.0: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928047 | 17:47 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928048 | 17:48 |
| TheJulia | rutro | 17:48 |
| TheJulia | that was not what I intended | 17:48 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/25.0: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928049 | 17:49 |
| TheJulia | https://review.opendev.org/q/If04a5b97722cc1a8e125c3348e09339c3a7ce0eb to cleanup Ironic CI so we can merge things | 17:52 |
| dtantsur | TheJulia: master change approved with nits to follow-up | 17:52 |
| dtantsur | TheJulia: you need to remove the metal3 job from the gate as well | 17:53 |
| TheJulia | right | 17:53 |
| TheJulia | fuck | 17:53 |
| * TheJulia hates CI right now | 17:53 | |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/26.0: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928047 | 17:54 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/25.0: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928049 | 17:55 |
| dtantsur | Unfortunately, I'll need to drop now. I'll continue approving whatever is not approved by my morning | 17:55 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/24.0: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928054 | 17:56 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2024.1: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928057 | 18:01 |
| JayF | https://review.opendev.org/c/openstack/ironic-python-agent/+/927974 master patch is V+1 and needs +2A | 18:08 |
| TheJulia | speaking of things which don't show up in terminal windows, tailing spaces on the yaml | 18:17 |
| TheJulia | JayF: I can approve, or you can revise on master branch, up to you | 18:17 |
| JayF | please just land them as-is, I will fix if it is bothersome | 18:18 |
| JayF | my IDE has started doing that, I haven't found the setting yet to make it stop putting trailing spaces in | 18:18 |
| TheJulia | ok | 18:18 |
| opendevreview | Merged openstack/ironic bugfix/26.0: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928047 | 18:23 |
| opendevreview | Merged openstack/ironic-python-agent bugfix/9.13: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928015 | 18:26 |
| TheJulia | sigh, this is going to be a very very long day | 18:28 |
| opendevreview | Merged openstack/ironic-python-agent bugfix/9.12: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928017 | 18:31 |
| JayF | we need to approach it shift-based: we can handoff to the next TZ. This has been broken for over a decade, it's a marathon not a sprint (as someone once told mE) | 18:31 |
| opendevreview | Merged openstack/ironic bugfix/24.0: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928054 | 18:33 |
| opendevreview | Merged openstack/ironic bugfix/25.0: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928049 | 18:33 |
| TheJulia | we still have centos7 node-set usage on unmaintained/yoga and likely older branches | 18:36 |
| TheJulia | err, centos8 | 18:36 |
| cardoe | ugh sorry you guys are getting beat on with this. | 18:39 |
| cardoe | JayF: does your IDE respect editorconfig? We can add a pre-commit check for trailing spaces and put an editorconfig | 18:39 |
| JayF | it's pycharm | 18:40 |
| JayF | one of the most annoying things about it, it doesn't natively support our import ordering | 18:40 |
| JayF | because it won't split [openstack packages] and [other packages] as we tend to do | 18:40 |
| TheJulia | looks like I accidently pulled in an extra test on backporting 2023.2, fixing | 18:42 |
| cardoe | JayF: so that's where we setup pre-commit with tools that auto fix this stuff (instead of pep8/flake8 griping at you) and let your editor run on save. | 18:44 |
| JayF | cardoe: we have pre-commit configuration in ironic | 18:44 |
| JayF | cardoe: as long as you can also ensure anything configured there is in our tox-based linters, I'm game | 18:44 |
| JayF | .o(and wouldn't hate pre-commit being spread to other repos, too) | 18:44 |
| JayF | tbh I should add something to PTG notes; I think we should move to running our CI through pre-commit similar to how nova does | 18:44 |
| JayF | it's a really nice setup | 18:44 |
| cardoe | tox should just run pre-commit | 18:45 |
| JayF | that is how nova's setup works | 18:45 |
| cardoe | well I'm gonna be touching sushy first. | 18:45 |
| JayF | I'm onboard as long as the consensus of the core team is too :D | 18:46 |
| cardoe | I don't care what the choice is. I just wanna see you guys (and all OpenStack projects) worry about actual technical issues and solve those. | 18:48 |
| cardoe | And stop burning cycles on bad CI and code formatting. | 18:49 |
| JayF | You overestimate how fast I'd go without barriers in my way :D | 18:49 |
| JayF | my brain is usually the first thing I trip over when trying to run | 18:49 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.2: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927970 | 18:49 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.2: trivial: fix odd pep8 failures https://review.opendev.org/c/openstack/ironic/+/928071 | 18:49 |
| cardoe | Well on a positive note I do think you're good on https://review.opendev.org/c/openstack/ironic/+/924887 | 18:50 |
| cardoe | https://review.opendev.org/c/openstack/ironic/+/919779 as well | 18:52 |
| TheJulia | JayF: thoughts on disabling the snmp job on master branch (https://zuul.opendev.org/t/openstack/build/b380b3bd38054234a73db1c50ff4bba4/logs) | 18:52 |
| TheJulia | ? | 18:52 |
| JayF | I'm OK with it in general | 18:52 |
| JayF | but would like, for this change, for us to be 100% certain it's noise | 18:52 |
| TheJulia | well, it is definitely not toggling power on the job | 18:53 |
| JayF | ack; just -nv it then | 18:53 |
| JayF | if you fully remove it lmk so I can help cid re-enable it in his patches (he's trying to tackle that problem now) | 18:53 |
| TheJulia | I'll dig a little further to see if I can figure out what is going on at a high level, and then nv it | 18:57 |
| JayF | My main concern is the security patch not regressing it | 18:58 |
| JayF | I think that's exstremely unlikely so you should set the bar pretty low for "what is going on ata high level" imo | 18:58 |
| TheJulia | Sep 04 17:40:53.590540 np0038391085 ironic-conductor[108063]: ERROR ironic.common.images [None req-4736d2d9-36f6-427b-9ee1-4f3549630743 None None] Security: The requested deploy image for node image cache is of format image iso and is not in the [conductor]permitted_image_formats list. | 19:01 |
| JayF | ramdisk support regression? | 19:02 |
| TheJulia | looks like it | 19:04 |
| TheJulia | looks like it explicitly asks for an iso to boot and the way the logic is now formed it doesn't accept it | 19:04 |
| TheJulia | I had *no* idea it did this | 19:04 |
| JayF | I'm happy to jump on a call and help you out with it if you want, or anything you need | 19:05 |
| JayF | cardoe: you really want a thing to track down for CI? | 19:07 |
| JayF | cardoe: curl: (28) Failed to connect to download.cirros-cloud.net port 80 after 130732 ms: Connection timed out ## happens all the darn time for ironic | 19:08 |
| JayF | and I think it's because we don't use the same cached cirros version as other projects | 19:08 |
| JayF | if we could get that cached, or somehow make it so the ones that *are* cached worked, that'd be a big removal of intermittant failures | 19:08 |
| TheJulia | so, We *do* have a ramdisk iso test | 19:09 |
| TheJulia | BaremetalRamdiskBootIsoSNMPIPXE | 19:09 |
| TheJulia | so... Two options, we could permit iso in the list, or I'm wondering if we're going to need to extend the list | 19:10 |
| JayF | https://review.opendev.org/c/openstack/ironic-python-agent/+/928019 bugfix/9.9 IPA CI is broken at HEAD | 19:10 |
| JayF | TheJulia: I would just say add iso | 19:11 |
| JayF | TheJulia: or allow ISO separately from that list | 19:11 |
| JayF | only big downside to adding it to existing list: you enable ISOs in IPA even though that's not required | 19:11 |
| TheJulia | I suspect since we're catching stuff in the image cache, and it gets cached | 19:12 |
| TheJulia | we need to explicitly permit it | 19:12 |
| JayF | ++ do it. I'll do a followup to IPA in master to make the default match | 19:12 |
| TheJulia | running unit tests because I'm assuming it is going to break something else | 19:14 |
| JayF | good call, I'm fixing the IPA bugfix branch | 19:14 |
| JayF | it was never pinned to stable/2024.1 | 19:14 |
| TheJulia | yeah, I guess I'll need to edit the release note and maybe even docs :( | 19:24 |
| TheJulia | it actually makes a ton of sense where it was wired in at | 19:24 |
| TheJulia | and I have 51 failing tests | 19:24 |
| * TheJulia cries | 19:24 | |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.9: Pin jobs to stable/2024.1 deps https://review.opendev.org/c/openstack/ironic-python-agent/+/928079 | 19:27 |
| TheJulia | JayF: thoughts on adding in at https://review.opendev.org/c/openstack/ironic/+/927967/2/ironic/common/images.py#866 ? | 19:27 |
| JayF | I mainly am :( at the fact you'll have to rebase again | 19:27 |
| JayF | you mean, instead of updating the list? | 19:27 |
| JayF | do something like expected format in [iso]: allow it for $reasons | 19:28 |
| JayF | maybe even check the node to see if it's using a ramdisk interface? | 19:28 |
| TheJulia | well, in those cases, we will never have an expected format | 19:28 |
| TheJulia | it is all special casing image retrieval | 19:28 |
| TheJulia | like "oh, please boot this iso directly, kthxbai" | 19:28 |
| JayF | I am not concerned about ramdisk path | 19:28 |
| TheJulia | and we're just downloading it at that point, and it gets caught because of the caching ation | 19:28 |
| JayF | we never touch those isos, shortcut away | 19:28 |
| JayF | I'm more thinking about what happens if I send a rotten iso as the image_source | 19:29 |
| TheJulia | okay, now to figure out why my local copy of the branch is failing 51 tests with no changes | 19:29 |
| JayF | on the OTHER drivers, where we might convert it | 19:29 |
| TheJulia | yeah | 19:29 |
| TheJulia | where we'll have some input signaling otherwise | 19:29 |
| JayF | I am +2 to it, +999999 to it if we load it into a devstack and make sure a bad iso is rejected on deploy | 19:30 |
| JayF | but we can do that test while the CI is running | 19:30 |
| TheJulia | ugh, what in the world is going on | 19:33 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.9: Pin jobs to stable/2024.1 deps https://review.opendev.org/c/openstack/ironic-python-agent/+/928079 | 19:33 |
| TheJulia | it has to go into the list | 19:37 |
| TheJulia | because it is going to detonate before it reaches the ramdisk exclusion check | 19:37 |
| TheJulia | because it gets cached out | 19:38 |
| TheJulia | ... which of course, doesn't explains the python issues I'm having | 19:38 |
| JayF | is there something I can do to help? | 19:40 |
| JayF | if not might step away for lunch + dog walk | 19:40 |
| JayF | but happy to help if I can | 19:40 |
| TheJulia | do me a favor and just pull down master branch and make sure tox -epy3 works cleanly for you | 19:41 |
| TheJulia | I just deleted my tox env | 19:41 |
| TheJulia | and are-running | 19:41 |
| JayF | I'm using -epy311 | 19:43 |
| JayF | so it doesn't run under 3.12 (my default python) | 19:43 |
| JayF | TheJulia: passed | 19:44 |
| JayF | 9538 tests in 52.1454 sec | 19:44 |
| TheJulia | still running here, but so far I haven't seen any obvious test failures | 19:45 |
| JayF | rerunning with -repy311 just to ensure I didn't get any cache help | 19:45 |
| TheJulia | I'm guessing maybe I just had a very unahppy tox | 19:45 |
| TheJulia | okay, yay | 19:45 |
| TheJulia | clean here | 19:45 |
| JayF | aight, I'll be back, going to walk the dog then go grab lunch | 19:47 |
| JayF | if you get a patchset posted, drop me a text and I'll get to a pc faster | 19:47 |
| JayF | (also; don't neglect your own sustenance o/) | 19:49 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927965 | 20:00 |
| opendevreview | Merged openstack/ironic-python-agent master: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928011 | 20:04 |
| opendevreview | Merged openstack/ironic-python-agent master: Inspect non-raw images for safety https://review.opendev.org/c/openstack/ironic-python-agent/+/927974 | 20:04 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/26.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927966 | 20:08 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/25.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927967 | 20:09 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2024.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927968 | 20:11 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/24.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927969 | 20:13 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.2: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927970 | 20:15 |
| JayF | fungi: rosmaita: JFYI; we will be merging a different version of fix to Ironic than we sent out for embargos. I know that's nondesirable but we caught a regression in CI. | 20:15 |
| rosmaita | JayF: it happens, and it will be documented on the gerrit reviews | 20:16 |
| JayF | +2 on master version, the changes lgtm, will see how CI reacts | 20:16 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927972 | 20:19 |
| opendevreview | Adam Rozman proposed openstack/ironic-python-agent master: WIP - root device encryption https://review.opendev.org/c/openstack/ironic-python-agent/+/926425 | 20:22 |
| opendevreview | Merged openstack/ironic-python-agent stable/2024.1: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928012 | 20:36 |
| TheJulia | JayF: stevebaker[m]: fyi, our unmaintained CI is broked on older bifrost jobs with a deleted nodeset | 20:51 |
| TheJulia | https://review.opendev.org/q/Ib82833168efc2a6e2b4f1178258840d08deb78ef+project:openstack/bifrost need reviews as well | 20:51 |
| stevebaker[m] | ಠಿ_ಠ | 20:52 |
| opendevreview | Merged openstack/ironic-python-agent stable/2023.2: Remove and disable examples job https://review.opendev.org/c/openstack/ironic-python-agent/+/928013 | 20:54 |
| JayF | I greatly approve of the look of disapprov..ing confusion :) | 20:55 |
| stevebaker[m] | lol | 20:57 |
| opendevreview | Merged openstack/ironic stable/2024.1: CI: Disable metal3-integration test job https://review.opendev.org/c/openstack/ironic/+/928057 | 21:02 |
| TheJulia | stevebaker[m]: I guess you now have a nice idea of why I've been so heads down the past month and a half :) | 21:03 |
| JayF | TheJulia: my openstack team meeting today w/Adam and C.I.D. was basically "here's where I've been" :D | 21:04 |
| TheJulia | heh | 21:05 |
| JayF | I made a slide deck about this, going to record a video to post once everything is merged | 21:05 |
| stevebaker[m] | TheJulia: yeah yikes. wow test_format_inspector.py creates all the image types! | 21:07 |
| TheJulia | stevebaker[m]: well, that was lifted from nova, but yeah | 21:07 |
| TheJulia | ugh | 21:09 |
| TheJulia | the iso logic on the snmp job is still problematic | 21:09 |
| TheJulia | https://8033075ee4587cf67d00-eeaf584229491d7ea79dcd495ddd4db1.ssl.cf1.rackcdn.com/927965/3/check/ironic-tempest-ramdisk-bios-snmp-pxe/2745a8f/controller/logs/screen-ir-cond.txt | 21:09 |
| TheJulia | I'm going to step away for a few minutes and clear my mind since I guess we force raw as well | 21:10 |
| TheJulia | so we almost have a default job state which is quite broken | 21:10 |
| opendevreview | cid proposed openstack/ironic-python-agent master: Check for the existence of an IPMI device https://review.opendev.org/c/openstack/ironic-python-agent/+/926973 | 21:17 |
| TheJulia | i guess the cleanest thing to then do is if source format is iso, skip | 21:19 |
| JayF | is there ever a case where I could ask Ironic to write an iso to disk? | 21:20 |
| JayF | I guess if it's safety checked, that's irrelevent | 21:20 |
| JayF | as long as it's safety checked I do not care if we respect the allowlist in the ramdisk/iso case | 21:21 |
| TheJulia | to disk, no, boot on the other hand, sure | 21:21 |
| TheJulia | yeah, we're past the safety check there | 21:21 |
| JayF | when I say "to disk" I really mean "touches qemu-img" | 21:21 |
| TheJulia | I tyeah | 21:21 |
| TheJulia | err, yeah | 21:21 |
| * TheJulia steps outside for a couple minutes and will update the master branch patch | 21:21 | |
| JayF | I'll take a gander and see if there's anything I can sus out | 21:21 |
| JayF | please do get some lunch if you haven't, too | 21:21 |
| JayF | it's 2:20pm; the day is flying | 21:22 |
| TheJulia | I actually ate already :) | 21:25 |
| cid | cid 's EOD o/ | 21:26 |
| * cid forgot this :( | 21:26 | |
| cid | o/ | 21:26 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.9: Pin jobs to stable/2024.1 deps https://review.opendev.org/c/openstack/ironic-python-agent/+/928079 | 21:26 |
| JayF | you don't have to wave every day but it is nice :D | 21:26 |
| JayF | \o | 21:26 |
| * cid noted | 21:27 | |
| TheJulia | :) | 21:27 |
| JayF | It's nice to know people are doing good ironic things :D | 21:27 |
| cid | 😄 | 21:28 |
| opendevreview | cid proposed openstack/ironic-python-agent master: Check for the existence of an IPMI device https://review.opendev.org/c/openstack/ironic-python-agent/+/926973 | 21:30 |
| JayF | TIL about https://docs.python.org/3/library/glob.html | 21:30 |
| TheJulia | https://www.irccloud.com/pastebin/q4tRf8ka/ | 21:32 |
| TheJulia | thoughts ? | 21:32 |
| opendevreview | cid proposed openstack/ironic-python-agent master: Check for the existence of an IPMI device https://review.opendev.org/c/openstack/ironic-python-agent/+/926973 | 21:32 |
| JayF | I think I'm good with that | 21:33 |
| * TheJulia runs tests | 21:33 | |
| TheJulia | in a sense, it is *already* raw for any and all ironic purposes | 21:35 |
| TheJulia | ?porposes? | 21:35 |
| * TheJulia expects a picture of pixie boots riding a dolphin | 21:36 | |
| JayF | hilariously enough, I'm about to upload a short about bears to the yt chan | 21:37 |
| JayF | (pandas / polars the ML libraries) | 21:37 |
| opendevreview | cid proposed openstack/ironic-python-agent master: Check for the existence of an IPMI device https://review.opendev.org/c/openstack/ironic-python-agent/+/926973 | 21:39 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927965 | 21:41 |
| TheJulia | okay, lets see if that works | 21:41 |
| JayF | https://www.youtube.com/shorts/4tybImC6KJc /me cheating on pixie boots with polars and pandas :P | 21:43 |
| JayF | TheJulia: that lgtm, +2 awaiting CI :) | 21:43 |
| * TheJulia closes some browser tabs | 21:45 | |
| JayF | Only IPA patch that seems in peril is bugfix/9.9 | 21:46 |
| JayF | and that entire branch CI is completely hosed | 21:46 |
| JayF | hilariously enough, my CVE fix fixes the tests because it uses less code from ironic-lib (and the wrong ironic-lib version is what is blowing up ci) | 21:47 |
| JayF | but I don't care, I need CI to run properly and pass, so https://review.opendev.org/c/openstack/ironic-python-agent/+/928079 has to get in | 21:47 |
| * TheJulia puts loud music on | 21:49 | |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/26.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927966 | 21:52 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/25.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927967 | 21:53 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2024.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927968 | 21:53 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/24.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927969 | 21:55 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.2: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927970 | 21:56 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927972 | 21:57 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent bugfix/9.9: Pin jobs to stable/2024.1 deps https://review.opendev.org/c/openstack/ironic-python-agent/+/928079 | 21:59 |
| TheJulia | now, ci | 22:00 |
| TheJulia | and annoying the corgi through RATM | 22:00 |
| JayF | some of those that make images, wanna hack your 'puter ... SOMMMA THOSE THAT MAKE IMAGES GONNA TAKE YOUR FILES | 22:03 |
| JayF | [headbanging continues] | 22:03 |
| TheJulia | lol | 22:04 |
| TheJulia | ugh | 22:05 |
| TheJulia | pep8 | 22:05 |
| TheJulia | I ran you, why you dislike me | 22:06 |
| * TheJulia sighs | 22:06 | |
| TheJulia | i guess I didn't | 22:08 |
| TheJulia | easy enough fix | 22:08 |
| JayF | at least it's not something more substantial to track down 🥲 | 22:11 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927965 | 22:19 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/26.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927966 | 22:19 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/25.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927967 | 22:19 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2024.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927968 | 22:19 |
| opendevreview | Julia Kreger proposed openstack/ironic bugfix/24.0: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927969 | 22:19 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.2: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927970 | 22:20 |
| opendevreview | Julia Kreger proposed openstack/ironic stable/2023.1: CVE-2024-44982: Harden all image handling and conversion code https://review.opendev.org/c/openstack/ironic/+/927972 | 22:20 |
| TheJulia | I am Julia, destroyer of CI | 22:22 |
| JayF | that's an oof on that change | 22:28 |
| JayF | I did the same thing before gerrit was open, locally | 22:29 |
| JayF | literally almost identical, not having it tabbed in enough | 22:29 |
| TheJulia | it was a fast fix once I just typed out a loop to fix it all | 22:30 |
| TheJulia | but yeah | 22:30 |
| TheJulia | looks like the snmp job failed and auto-retried on master | 22:34 |
| TheJulia | ughhhhhh | 22:34 |
| JayF | it failed on setup | 22:35 |
| JayF | network issue, not actual failure | 22:35 |
| TheJulia | ugh | 22:36 |
| TheJulia | Looks like we could use a second review on https://review.opendev.org/q/I5254b80717cb5a7f9084e3eff32a00b968f987b7 | 22:44 |
| TheJulia | at least, some of them. Stepping away for like 10 minutes and then I'll resume on the rest of the happy ones | 22:44 |
| JayF | bugfix/9.9 CI is not going to be fixed by me today, I've worked on it a while and am stuck, if someone can look overnight that'd be awesome or else I'll resume tomorrow | 22:46 |
| JayF | oh no https://usercontent.irccloud-cdn.com/file/L1KHirVe/image.png | 22:47 |
| JayF | it was the arm64 job | 22:47 |
| JayF | but that's a frightening old "friend" | 22:47 |
| * JayF is just going to assume some solar flares hit the CI machine | 22:48 | |
| TheJulia | We can disable the test_iso_9660 job | 22:48 |
| TheJulia | realistically I bet they are just the slower tests in the grand scheme of things | 22:49 |
| JayF | I am OK with that, but I also am suspicious it's actually broken vs just arm silly | 22:49 |
| JayF | yep | 22:49 |
| JayF | exactly | 22:49 |
| JayF | esp. since arm clouds are in flux (opendev lost a donor) | 22:49 |
| TheJulia | JayF: bugfix/9.9 for ipa is broken? | 22:51 |
| JayF | yes, on master | 22:51 |
| JayF | er HEAD | 22:51 |
| TheJulia | oh | 22:52 |
| TheJulia | I see | 22:52 |
| JayF | https://review.opendev.org/c/openstack/ironic-python-agent/+/928079 | 22:52 |
| TheJulia | funky, one back it is fine | 22:52 |
| JayF | is what I've been poking it with, it's still running I haven't checked status in zuul because tbh my brain is fried | 22:52 |
| JayF | the bugfix branches that are for the next release work, b/c they still use master constraints | 22:52 |
| JayF | 9.9 is 2024.1 based, so it needs the older constraints | 22:52 |
| TheJulia | okay, well, we just need to start merting the ipa side of the fixes | 22:54 |
| JayF | yeah, they are good afaict | 22:55 |
| JayF | even 2023.1 the failure was network | 22:55 |
| JayF | stevebaker[m]: can you please approve https://review.opendev.org/q/I5254b80717cb5a7f9084e3eff32a00b968f987b7 | 22:55 |
| JayF | I'm going to ensure all openstack/ironic patches open for the CVE get my +2, obviously pending CI being happy. | 22:58 |
| TheJulia | thanks | 23:00 |
| TheJulia | looks like we might be in for some rechecking on master branch | 23:06 |
| JayF | Can we 'follow the sun' on the rechecks? | 23:07 |
| JayF | to use the ops term | 23:07 |
| JayF | meaning like, go have an evening so we'll be fresh tomorrow in case real issue spawn again? | 23:07 |
| TheJulia | ugh, might need to nuke some further tests, py36 failing on https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_376/927975/4/check/openstack-tox-py36/37664f0/testr_results.html but it might be entirely load related, dunno | 23:09 |
| TheJulia | I'm going to keep an eye on on master branch and hopefully things will begin merging soon | 23:09 |
| TheJulia | it is clear we're not helping ourselves CI load wise | 23:09 |
| JayF | yeah, we should probably just focus on each branch as we go | 23:14 |
| JayF | btw, spent a little time while waiting poking at the new oslo.utils format_inspector by doing this https://github.com/jayofdoom/disk-image-checker | 23:15 |
| JayF | seems like something you might find useful for downstream use cases, fwiw | 23:15 |
| JayF | I'm stepping away from IRC for the workday; if you need something from me ping me directly and I'll be checking periodically. o/ | 23:20 |
| JayF | also if it's not after 9pm, most of you have my cell, sms me if you need a re-approval or are blocked on reviews | 23:21 |
| JayF | (PDT) | 23:21 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
