| opendevreview | Steve Baker proposed openstack/ironic master: Utility functions for graphical console drivers https://review.opendev.org/c/openstack/ironic/+/939505 | 00:56 |
|---|---|---|
| opendevreview | Steve Baker proposed openstack/ironic master: Add ironic-novncproxy service https://review.opendev.org/c/openstack/ironic/+/939191 | 00:56 |
| opendevreview | Steve Baker proposed openstack/ironic master: [WIP] Add systemd provider for console containers https://review.opendev.org/c/openstack/ironic/+/941614 | 00:56 |
| opendevreview | Steve Baker proposed openstack/ironic master: [WIP] implement drivers redfish-graphical, fake-graphical https://review.opendev.org/c/openstack/ironic/+/941615 | 00:56 |
| opendevreview | Andrew Bonney proposed openstack/ironic master: Filter physnets when ports are pre-allocated to a segment https://review.opendev.org/c/openstack/ironic/+/941343 | 08:40 |
| opendevreview | Andrew Bonney proposed openstack/ironic master: Filter physnets when ports are pre-allocated to a segment https://review.opendev.org/c/openstack/ironic/+/941343 | 08:44 |
| opendevreview | Merged openstack/metalsmith master: Add ability to specify columns used for json output https://review.opendev.org/c/openstack/metalsmith/+/928993 | 11:23 |
| opendevreview | Harald Jensås proposed openstack/ironic master: [WIP] OVN UEFI IPv6 https://review.opendev.org/c/openstack/ironic/+/941352 | 12:16 |
| *** tkajinam is now known as Guest9129 | 13:39 | |
| TheJulia | JayF: w/r/t https://review.opendev.org/c/openstack/ironic/+/941607, it LTGM | 14:20 |
| TheJulia | Err LGTM | 14:20 |
| opendevreview | Harald Jensås proposed openstack/ironic master: OVN UEFI IPv6 CI job https://review.opendev.org/c/openstack/ironic/+/941352 | 14:36 |
| cardoe | How far back does that need to be backported? | 15:07 |
| TheJulia | Likely to all current stable branches once merged | 15:08 |
| kubajj | Should it be possible to backport to unmaintained branches as well? (Just out of curiosity. I've got a weird error message when I tried to backport to antelope in IPA, maybe cause by the change https://review.opendev.org/c/openstack/ironic-python-agent/+/936679 failing? But no clue) | 15:23 |
| TheJulia | It can be possible | 15:24 |
| TheJulia | eh, that looks like CI is just unhappy | 15:24 |
| TheJulia | I guess the question is how much CI we keep on the oldest branches | 15:24 |
| opendevreview | Merged openstack/ironic-python-agent master: Trivial:Remove codespell job (moved to pre-commit) https://review.opendev.org/c/openstack/ironic-python-agent/+/941456 | 15:51 |
| JayF | shermanm: jrosser: We tracked it down to a very, very slow BMC (thank you power_wait+power_state_change_timeout) and a bad IPA build with some permissions issues | 16:06 |
| JayF | we have it deployed, thank you for the help | 16:06 |
| opendevreview | Merged openstack/ironic master: Pass agent token to get command results https://review.opendev.org/c/openstack/ironic/+/941607 | 16:12 |
| jrosser | JayF: good it’s working - would be interesting to hear if you find any surprises with the hp ampere | 16:36 |
| JayF | jrosser: we got some interesting errors out of it before we slowed Ironic down; I've encouraged downstream engineers to share those in an upstream bug but I generally let them do it so I don't spill any beans :D | 16:37 |
| jrosser | I’ve wondered how closely related it is to the supermicro as they’re both openbmc and almost certainly derived from the same reference design | 16:37 |
| JayF | You should not assume that the hardware we're using has an off-the-shelf BMC configuration. | 16:37 |
| TheJulia | would they be issues like... the bmc state takes like 5 minutes to update? | 16:38 |
| jrosser | obliquely relatedly, has anyone seen virtual media boot validate the certificate of the web server that the boot image is retrieved from? | 16:44 |
| jrosser | rather than just accept whatever random cert might be presented | 16:45 |
| dtantsur | some vendors start doing it by default. I think we've seen Dells? | 16:46 |
| dtantsur | Can be turned on/off in GUI | 16:46 |
| JayF | dtantsur: an error around being unable to adjust power state in post, and something generically invalid | 16:49 |
| JayF | dtantsur: all resolved by upping ironic timeouts and power_wait | 16:50 |
| JayF | dtantsur: but it still looked like something we'd wanna retry in sushy | 16:50 |
| TheJulia | cid: if you still around, https://review.opendev.org/c/openstack/ironic-python-agent/+/941489 needs a minor release not fix so we can backport it | 16:51 |
| cid | Alright, taking a look now. | 16:54 |
| TheJulia | cardoe: I'm good with https://review.opendev.org/c/openstack/networking-generic-switch/+/939211 | 17:08 |
| opendevreview | cid proposed openstack/ironic-python-agent master: Ensure IPA is locked down in rescue mode https://review.opendev.org/c/openstack/ironic-python-agent/+/941489 | 17:19 |
| shermanm | I was happy to come across this in the queue, https://review.opendev.org/c/openstack/ironic/+/699953, it would definitely improve things for us | 17:31 |
| TheJulia | Yeah, I wnat to get that merged in | 17:32 |
| cardoe | TheJulia: I'm good with Andrew's change as well. Given how neutron works. | 17:39 |
| cardoe | It's still wrong wrt to the spec. One of the random things I need to track on my behavioral delta to NGS. | 17:39 |
| cardoe | But its neutron internals that are wrong. | 17:40 |
| cardoe | That's all I was trying to say before. | 17:41 |
| cardoe | +1 to that patch as well shermanm. | 17:42 |
| opendevreview | Julia Kreger proposed openstack/ironic master: trivial: de-distro dnsmasq version check https://review.opendev.org/c/openstack/ironic/+/941709 | 18:49 |
| opendevreview | Julia Kreger proposed openstack/ironic master: trivial: lock dnsmasq check to ubuntu https://review.opendev.org/c/openstack/ironic/+/941710 | 18:49 |
| TheJulia | I happen to be firing up devstack in fips mode on centos, so... those bit me | 18:52 |
| *** awb_ is now known as awb | 19:04 | |
| opendevreview | Julia Kreger proposed openstack/ironic master: Filter physnets when ports are pre-allocated to a segment https://review.opendev.org/c/openstack/ironic/+/941343 | 19:22 |
| TheJulia | cardoe: ^ the ironic patch Andrew mentions, I've added a release note | 19:22 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CI: Set multinode jobs to fail if binding fails https://review.opendev.org/c/openstack/ironic/+/941712 | 19:44 |
| opendevreview | Satoshi Shirosaka proposed openstack/ironic-python-agent master: WIP Add ContainerHardwareManager https://review.opendev.org/c/openstack/ironic-python-agent/+/941714 | 19:52 |
| shermanm | (very much not an ironic question): some time back I recall mention of who to talk to about bugs / feature requests for gophercloud? I was looking into adding support for blazar, but wasn't sure where to start | 20:19 |
| JayF | they have a github, you could open an issue | 20:21 |
| JayF | adamcarthur5 might have some specific advice for getting started there | 20:21 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Fix devstack plugin for centos https://review.opendev.org/c/openstack/ironic/+/941720 | 20:49 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Fix devstack plugin for centos https://review.opendev.org/c/openstack/ironic/+/941720 | 20:53 |
| opendevreview | Satoshi Shirosaka proposed openstack/ironic-python-agent master: WIP Add ContainerHardwareManager https://review.opendev.org/c/openstack/ironic-python-agent/+/941714 | 21:02 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Fix devstack plugin for centos https://review.opendev.org/c/openstack/ironic/+/941720 | 21:05 |
| opendevreview | Jay Faulkner proposed openstack/ironic stable/2024.2: Pass agent token to get command results https://review.opendev.org/c/openstack/ironic/+/941722 | 21:06 |
| JayF | stevebaker[m]: | 21:11 |
| JayF | whoops | 21:11 |
| JayF | stevebaker[m]: I approved https://review.opendev.org/c/openstack/ironic/+/939505/8#message-8a9c62a4093c2d91b8c00757be34dedb9b15da1f but had a security-related question, if you don't mind making sure you make sure we're safe there it'd be great | 21:12 |
| TheJulia | JayF: anything with secret or password in the name gets scrubbed by the api surface code | 21:13 |
| JayF | ack, I knew "password" was, I wasn't sure about "secret" | 21:13 |
| TheJulia | yeah, it is | 21:13 |
| JayF | tyvm for confirming | 21:13 |
| JayF | I was like, 99% sure but it's good to be 100 | 21:14 |
| TheJulia | I double checked for the oci authentication stuffs | 21:14 |
| TheJulia | (and yeah, it scrubbed it on the api side as expected (annoyingly so!)) | 21:14 |
| TheJulia | that being so, it is *always* good to double check | 21:14 |
| stevebaker[m] | JayF: that key is why I proposed this change in the first place, which hasn't actually landed yet https://review.opendev.org/c/openstack/ironic/+/939504 | 21:19 |
| JayF | what are you talking about? that's approved, and it totally didn't get approved 13 seconds after you pasted it in IRC and I saw it :P | 21:20 |
| JayF | (I had read that two days ago and wanted to see if Kaifeng had anything else to chime in) | 21:20 |
| JayF | I'm sure I'll approve it again when I get to it in my massive tabs of review queue :D | 21:21 |
| stevebaker[m] | :D | 21:21 |
| JayF | TheJulia: is your comment on https://review.opendev.org/c/openstack/ironic-python-agent/+/941300/3/ironic_python_agent/extensions/standby.py#75 still valid or is that what the `return r` fixed? | 21:26 |
| stevebaker[m] | Hey the graphical console work includes this whole other repo for the actual browser container (README will be written next week). We need to decide where the scripts should live, and also how others might build and consume a container. One option is that the contents of bin live in the ironic repo, and other existing repos have a way of building a container, like kolla | 21:32 |
| stevebaker[m] | This whole other repo https://github.com/steveb/ironic-vnc-container | 21:32 |
| JayF | I think it should go in ironic-the-program | 21:33 |
| JayF | Given we have history of embedding useful/needed scripts, even if containerfiles, e.g. ironic-python-agent/imagebuild (pre ipa-builder), we can probably slot it into ironic | 21:34 |
| JayF | just make sure that setuptools is setup in a way where they don't end up in pypi bundles | 21:34 |
| JayF | (or maybe we do install them as resources? hmmmmm) | 21:34 |
| stevebaker[m] | so pull that whole repo into the ironic repo? | 21:34 |
| JayF | That's what I'd think to do. I'm not sure everyone would agree :) | 21:35 |
| JayF | we have examples in ipa repo, which is basically just bonus code | 21:35 |
| JayF | maybe even just tools/vnc-container/[repo contents] | 21:35 |
| JayF | or something similarly better named | 21:35 |
| stevebaker[m] | yeah that would certainly be easiest. The container is currently opinionated centos | 21:36 |
| JayF | maybe specifically namespace the container to centos then? | 21:37 |
| JayF | so if others want to contribute more, more power to them | 21:37 |
| JayF | I suspect my downstream will run what they are given :) | 21:37 |
| stevebaker[m] | getting these scripts into pypi or whatever would actually be useful, because there will be other container build pipelines that need to consume them https://github.com/steveb/ironic-vnc-container/tree/main/bin | 21:37 |
| JayF | should it be it's own repo/project then? | 21:38 |
| stevebaker[m] | they could be packaged as resources I think | 21:39 |
| stevebaker[m] | like setup.cfg data_files to /usr/share/ironic/vnc-container or something | 21:40 |
| JayF | yeah | 21:41 |
| JayF | you are thinking along the lines as me now | 21:41 |
| JayF | and distro people can tar it up if they want :D | 21:41 |
| stevebaker[m] | yep, like RDO could create a sub-package with just those scripts, and a container build can depend on that | 21:43 |
| TheJulia | JayF: oh yeah, return r was all that was needed | 21:43 |
| JayF | I already +2d with that assumption as a condition | 21:43 |
| JayF | I have reviewed most everything outstanding AFAIK, if I missed something anyone needs please link it to me. | 21:44 |
| TheJulia | honestly, I'd pull either into ironic's repo or another repo, but preferably ironic's main repo if possible since this is on a path to become core functionality | 21:44 |
| TheJulia | agent just makes sense to be its own because it has a different use which is adjacent to address the overall task | 21:45 |
| stevebaker[m] | a sub directory of tools seems reasonable. Having it in the ironic repo will give it more visibility for adding support for other hardware | 21:47 |
| JayF | TheJulia: the idea of agent being in the same repo had NEVER occurred to me, because it's always been so separate | 21:47 |
| TheJulia | Yeah, even today I would say separate | 21:48 |
| TheJulia | but for vnc stuffs and service, just all together | 21:48 |
| JayF | I think if we were making the agent today, a LOT of things about it would be different | 21:51 |
| JayF | it's a testament to how flexible the design was that it's still a good tool 10 years later, but I want to go back in time and make it even better :D | 21:52 |
| TheJulia | hmmm | 21:58 |
| TheJulia | our bindep stuff dislikes us | 21:58 |
| TheJulia | so fun problem... our devstack plugin expects xinetd | 22:02 |
| TheJulia | or inetd at all | 22:03 |
| TheJulia | but... "nope" | 22:03 |
| TheJulia | Regarding bidnep, I mean centos | 22:04 |
| stevebaker[m] | xinetd isn't even a thing in centos now. its systemd socket activation | 22:05 |
| TheJulia | yeah | 22:05 |
| JayF | lolwhat | 22:05 |
| TheJulia | yeah | 22:05 |
| JayF | 2005 called it wants its service runner back | 22:06 |
| TheJulia | Modern Linux doesn't understand the non-multi-byte encoding | 22:06 |
| JayF | IDK if you need to fix it *right now* but if not, make sure it gets in a bug and I'm sure it'll get a look | 22:06 |
| TheJulia | I'm trying to do a fips enforce devstack on centos | 22:07 |
| TheJulia | because I'm not going to pay ubuntu for ubuntu pro | 22:07 |
| TheJulia | so I can turn on fips mode | 22:07 |
| JayF | \o/ | 22:07 |
| TheJulia | ... trying to decide if we just make tftp support optional... | 22:07 |
| TheJulia | at least in the plugin | 22:07 |
| JayF | If you don't need it for the testing you're doing, JFDI | 22:08 |
| JayF | but please do bug it | 22:08 |
| JayF | because that will break somewhere eventually and it's pretty straightforward to fix | 22:08 |
| TheJulia | yeah... trying to decide now | 22:08 |
| TheJulia | I could vmedia it | 22:08 |
| cardoe | uuhh stevebaker[m] had me at container with a +12 | 22:12 |
| cardoe | I know that's not gonna fly but I'm gonna end up building it as a container so the work is done for me | 22:12 |
| JayF | I mean, that's how the vnc console stuff will work | 22:13 |
| JayF | with a container-based proxy thinger | 22:13 |
| JayF | that's the official technical term /s | 22:13 |
| cardoe | So doesn't conflict with cid's changes but if ya want... https://review.opendev.org/c/openstack/ironic/+/940332 is a refactor that fixes an actual bug (pretty small corner case) | 22:14 |
| TheJulia | can we make that official name? | 22:15 |
| TheJulia | JayF: bugs filed | 22:17 |
| JayF | cool | 22:18 |
| JayF | I don't have time today, but that'd be a nice relaxing one to take on if nobody gets to it before me :) | 22:18 |
| JayF | systems stuff always is like a warm blanket compared to the cold dead blackness of python :P | 22:18 |
| TheJulia | so. I think one thing to do is just to enable turning off tftp, and see if I can just low memory refish it | 22:18 |
| JayF | this is an awful suggestion dpeending on if you're trying to do CI or local testing | 22:19 |
| JayF | but just make it disablable | 22:19 |
| JayF | and manually setup tftp | 22:19 |
| TheJulia | yeah, that is the other major possibility | 22:19 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Trivial: Enable disabling tftp setup https://review.opendev.org/c/openstack/ironic/+/941742 | 22:30 |
| cardoe | You just nailed my favor commit message of the day... "enable disabling" | 22:31 |
| TheJulia | I <3 success | 22:32 |
| TheJulia | I've swapped my config over to virutal media, so time will tell | 22:33 |
| cardoe | You know... enabling ruff format and just setting the quote style to single, since we already set the line length... not that bad of a diff. | 22:38 |
| cardoe | The biggest change is some_func_call(arg1, arg2, {new line and indented over to arg1} arg3, arg4) is instead some_func_call({new line and 4 spaces in} arg1, arg2, arg3, arg4 {new line}) | 22:40 |
| cardoe | Found it easier to just enable it and only commit the section I was really changing than to try and edit my code to match the style after the fact. | 22:41 |
| opendevreview | Merged openstack/ironic-python-agent master: Ensure IPA is locked down in rescue mode https://review.opendev.org/c/openstack/ironic-python-agent/+/941489 | 22:46 |
| opendevreview | Merged openstack/ironic-python-agent master: Add token validation to GET command endpoints https://review.opendev.org/c/openstack/ironic-python-agent/+/941539 | 22:46 |
| opendevreview | Merged openstack/ironic-python-agent master: follow-up: update release note for bootable container work https://review.opendev.org/c/openstack/ironic-python-agent/+/941139 | 22:46 |
| TheJulia | cardoe: w/r/t your ML post, that is likelly the cleanest path, to at least try and whip up some templaes or even a superset driver to do some config embedding, But last time I looked... it didn't look like there was much available in the tooling because some of the "fork" looks like in name only | 22:47 |
| opendevreview | Merged openstack/ironic master: Utility functions for graphical console drivers https://review.opendev.org/c/openstack/ironic/+/939505 | 22:47 |
| opendevreview | Merged openstack/ironic master: Mask all driver_internal_info in node output https://review.opendev.org/c/openstack/ironic/+/939504 | 22:47 |
| opendevreview | Merged openstack/ironic master: oci: fix auth config loading https://review.opendev.org/c/openstack/ironic/+/941479 | 22:47 |
| cardoe | I figured let me ask before diving in head first. | 22:47 |
| cardoe | Gerrit and Zuul just got back from their coffee break. | 22:48 |
| opendevreview | Merged openstack/ironic-tempest-plugin master: Testing bad microversions on v1/allocations https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/937213 | 23:11 |
| TheJulia | ... coffee sounds good | 23:26 |
| cardoe | https://etherpad.opendev.org/p/cardoe-l2vni-segment-bug here's my jump to conclusions on the issue I saw in Andrew's code. | 23:28 |
| cardoe | jamesdenton: ^ | 23:28 |
| opendevreview | Merged openstack/ironic-tempest-plugin master: Testing bad microversions on v1/nodes/{uuid}/firmware https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/937214 | 23:38 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!