| opendevreview | Merged openstack/sushy-tools master: Openstack - get/set_boot_image support volume boot https://review.opendev.org/c/openstack/sushy-tools/+/942212 | 00:03 |
|---|---|---|
| opendevreview | Satoshi Shirosaka proposed openstack/ironic master: Remove multinancy page https://review.opendev.org/c/openstack/ironic/+/942306 | 00:03 |
| opendevreview | Verification of a change to openstack/ironic-tempest-plugin master failed: Add retries while waiting for SSH on server https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/942009 | 00:09 |
| opendevreview | cid proposed openstack/ironic master: Follow-up: Apply Inspection Rules https://review.opendev.org/c/openstack/ironic/+/942112 | 00:18 |
| opendevreview | cid proposed openstack/ironic master: API/Testing: Inspection rules migration https://review.opendev.org/c/openstack/ironic/+/939217 | 00:18 |
| frickler | I've noticed some unstable looking failures in the devstack ironic job lately, maybe someone wants to take a look? https://zuul.opendev.org/t/openstack/builds?job_name=ironic-tempest-bios-ipmi-direct-tinyipa&project=openstack/devstack | 06:12 |
| opendevreview | Michal Nasiadka proposed openstack/networking-generic-switch master: doc: Rework support matrix for trunk driver https://review.opendev.org/c/openstack/networking-generic-switch/+/942338 | 07:35 |
| opendevreview | Vasyl Saienko proposed openstack/ironic-tempest-plugin master: Validate network data for portgroups https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/940678 | 07:36 |
| opendevreview | Michal Nasiadka proposed openstack/networking-generic-switch master: doc: Rework support matrix for trunk driver https://review.opendev.org/c/openstack/networking-generic-switch/+/942338 | 07:42 |
| opendevreview | Michal Nasiadka proposed openstack/networking-generic-switch master: doc: Rework support matrix for trunk driver https://review.opendev.org/c/openstack/networking-generic-switch/+/942338 | 07:48 |
| vsaienko | TheJulia: thanks for review, I've pushed a new revision https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/940678, I've added latest microversion in requests with portgroups, as if not specify it client uses smallest version available. | 07:48 |
| opendevreview | Michal Nasiadka proposed openstack/networking-generic-switch master: doc: Rework support matrix for trunk driver https://review.opendev.org/c/openstack/networking-generic-switch/+/942338 | 07:51 |
| opendevreview | Michal Nasiadka proposed openstack/networking-generic-switch master: doc: Rework support matrix for trunk driver https://review.opendev.org/c/openstack/networking-generic-switch/+/942338 | 07:53 |
| opendevreview | Vasyl Saienko proposed openstack/ironic-tempest-plugin master: Validate network data for portgroups https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/940678 | 07:53 |
| rpittau | good morning ironic! o/ | 08:07 |
| opendevreview | Vasyl Saienko proposed openstack/ironic master: Enable atop on jobs https://review.opendev.org/c/openstack/ironic/+/942340 | 08:14 |
| opendevreview | Vasyl Saienko proposed openstack/ironic-tempest-plugin master: Validate network data for portgroups https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/940678 | 08:34 |
| rpittau | before I forget again, don't you forget to register for the next OpenInfra vPTG! :) | 10:49 |
| rpittau | #link http://ptg.openinfra.dev/ | 10:49 |
| rpittau | the topics etherpad is ready to be populated! | 10:50 |
| rpittau | #link https://etherpad.opendev.org/p/ironic-ptg-april-2025 | 10:50 |
| opendevreview | Vasyl Saienko proposed openstack/ironic-tempest-plugin master: Validate network data for portgroups https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/940678 | 11:53 |
| opendevreview | Vasyl Saienko proposed openstack/networking-generic-switch master: Adds vlan aware VMs support for Cumulus NVUE and DellOS10 https://review.opendev.org/c/openstack/networking-generic-switch/+/940966 | 12:00 |
| opendevreview | Julia Kreger proposed openstack/ironic master: ci: set a longer nova tempest build interval https://review.opendev.org/c/openstack/ironic/+/942355 | 13:24 |
| opendevreview | Julia Kreger proposed openstack/ironic master: trivial: follow-up on ftp enable/disable flag https://review.opendev.org/c/openstack/ironic/+/942357 | 13:37 |
| opendevreview | Verification of a change to openstack/ironic master failed: Fix devstack plugin for centos https://review.opendev.org/c/openstack/ironic/+/941720 | 13:58 |
| TheJulia | cardoe: Could you take a glance at https://review.opendev.org/c/openstack/ironic/+/941343 please | 13:59 |
| TheJulia | Thanks in advance! | 13:59 |
| cardoe | I +2’d it. I went through that flow yesterday when I started writing up some of my flow and how it interacts. | 14:03 |
| cardoe | His code is correct for L3 routed networks for sure. | 14:05 |
| cardoe | I just can’t phrase it correctly to neutron folks that L2 networks can’t do the right thing. | 14:06 |
| cardoe | I tried to explain this week that I have two VXLAN based networks that are physically separate and I need physical_network. And I keep getting told that it’s an overlay network not a physical one. | 14:06 |
| TheJulia | just in irc ? | 14:06 |
| cardoe | Yes at the weekly meeting. | 14:07 |
| TheJulia | So I guess they are unable to wrap their head around the physical modeling | 14:07 |
| TheJulia | ... hmmmmm | 14:07 |
| TheJulia | could it be time to request a higher bandwidth call? | 14:08 |
| cardoe | Maybe. I’m going to finish my write up first and share it with them. | 14:12 |
| opendevreview | Verification of a change to openstack/ironic master failed: ci: focus ironic-tempest-bios-ipmi-direct-tinyipa https://review.opendev.org/c/openstack/ironic/+/942204 | 14:33 |
| frickler | cardoe: so you want someone (neutron?) to set up bridging between those two vxlans? or between a neutron vxlan and a physical port (aka flat network)? | 14:33 |
| * TheJulia wonders if there is a slight difference in terminology | 14:39 | |
| opendevreview | cid proposed openstack/ironic master: More reliable TinyIPA build with network retries https://review.opendev.org/c/openstack/ironic/+/942368 | 14:51 |
| opendevreview | cid proposed openstack/ironic-python-agent-builder master: More reliable TinyIPA build with network retries https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/942369 | 14:51 |
| frickler | pretty sure there is, what would the world be like if terminology was unique and well defined like in maths? ;-) | 15:02 |
| frickler | who is the resident expert for redfish virtual-media things? is there a list of vendors or similar that support the USB-for-configdrive option? seems supermicro doesn't, not that that would surprise me | 15:04 |
| dtantsur | frickler: it's usually someone from our team :) But we don't have such a list. In fact, we have serious doubts that it works on real hardware. | 15:05 |
| dtantsur | I think we found Dell disliking more than one virtual media device at the same time. | 15:05 |
| frickler | dtantsur: I'm wondering whether one could merge the config-drive data into the boot iso. teaching cloud-init about that might be another step, but if I could access them via some custom tooling that would already be very helpful | 15:07 |
| dtantsur | frickler: I think we partly do that when Ironic is building the ISO. It's going to be a problem when the configdrive is provided as a blob. | 15:11 |
| frickler | ah, hmm, but that would still be an iso blob, which could be mounted and copied from? | 15:12 |
| dtantsur | frickler: it will be in realistic cases, but strictly speaking we put no restrictions on what a configdrive is. | 15:12 |
| dtantsur | (it's not an objections, just something we need to consider) | 15:13 |
| dtantsur | By the way, are we talking about the "ramdisk deploy" / live ISO use case? | 15:14 |
| dtantsur | Or are you looking for having a USB device connected to a normally provisioned instance? | 15:14 |
| dtantsur | (should have asked that as the first thing) | 15:14 |
| frickler | ok, I was under the assumption that it should always mimic what one would see from the metadata service | 15:14 |
| dtantsur | We don't prevent people from doing crazy things in this area :) | 15:14 |
| dtantsur | Again, not a blocking concern, just maybe we'll need a way to opt out | 15:15 |
| frickler | I'm actually talking about ramdisk deploy, for the live iso the image isn't (re)built by ironic afaict? | 15:15 |
| dtantsur | Riiiiight | 15:15 |
| dtantsur | Yeah, it sounds like an entirely new feature | 15:16 |
| dtantsur | Now, my concern would be: can we actually reliably rebuild an arbitrary CD image? | 15:17 |
| frickler | does encoding metadata in a specifically named ipmi/redfish user account count as crazy thing? ;-D that's my current workaround | 15:17 |
| dtantsur | It's not just files, I believe, it also has some sort of a bootloader-ish section? | 15:17 |
| dtantsur | wut? | 15:17 |
| dtantsur | frickler: sounds scary to me, unless I'm missing something | 15:18 |
| frickler | well I essentially just need three bytes or digits. so I create an extra user META123 that isn't actually used and can check for that on the server with ipmitool => deployment proceeds happily | 15:19 |
| dtantsur | Oh, that's just as cursed as I thought :D | 15:20 |
| dtantsur | You can probably (ab)use UEFI variables for that too | 15:20 |
| dtantsur | CoreOS images have a cool feature: they leave a designated place on the image where you can put your first-boot stuff | 15:21 |
| dtantsur | I wish something like this was more common | 15:21 |
| frickler | for the bootloader stuff, there's the esp_image thing already, and that mostly works for me. just chasing some bug that seems related to https://bugs.launchpad.net/ironic/+bug/2081719 | 15:21 |
| frickler | oh, UEFI variables is also a good idea, that might even be more generic | 15:22 |
| frickler | dtantsur: like just "echo my-metadata | dd skip=well-known-offset of=image-file"? that sounds nice, too | 15:23 |
| dtantsur | yeah, along these lines | 15:23 |
| dtantsur | This allows us (in OpenShift) to have a service that serves node-specific ISO images on fly | 15:23 |
| frickler | might be possible to have some dib element that does something similar | 15:24 |
| * dtantsur nods | 15:25 | |
| frickler | dtantsur: that was pretty helpful, thx. /me goes back to isolating the esp_image/bootloader bug now | 15:28 |
| dtantsur | On the topic of bugs: https://bugs.launchpad.net/ironic/+bug/2098977 is very annoying, and I'll be grateful if someone with spare cycles could do the code reshuffling | 15:34 |
| cardoe | frickler: no. neutron controls bare metal networks when integrated with Ironic and ML2 mechanisms like networking-generic-switch. I'm trying to work with physical VXLANs. | 15:51 |
| cardoe | dtantsur: 100% agree with that bug. I believe what we need to do is have a "sushy.quirks.<vendor>" or some such. | 15:52 |
| dtantsur | I'm going to file another bug like that (HPE not supporting changing boot mode) | 16:02 |
| dtantsur | ... by "like that" I mean: could benefit from quirks | 16:02 |
| opendevreview | Merged openstack/sushy master: explicit oem extn names and fix parallel tests https://review.opendev.org/c/openstack/sushy/+/942166 | 16:02 |
| cardoe | dtantsur: so rather than us trying 20 different methods we have a quirks and either when we connect to the device we apply the quirks to the connection or maybe we magic load the quirks | 16:04 |
| dtantsur | It's a good idea (probably tie it together with the oem rework) | 16:05 |
| dtantsur | TheJulia: the HPE bug I promised you https://bugs.launchpad.net/ironic/+bug/2098983 | 16:07 |
| frickler | cardoe: in my world there is no such thing as "physical vxlan". from what you've said in the neutron channel last week, for you this means a vxlan with the source address for the encapsulation tied to a specific physical network? else you need to share your definition for it | 16:18 |
| cardoe | You're right. VXLAN isn't really physical. | 16:20 |
| cardoe | There's a few nuances here that all tie together. | 16:23 |
| cardoe | So I have a box. It's physically plugged into 4 different switches. | 16:23 |
| cardoe | Two switches make up a leaf pair on fabric X. Two switches make up a leaf pair on fabric Y. | 16:24 |
| cardoe | Today Ironic marries up neutron ports to bare metal ports using the physical_network field and then pxe_enabled field. | 16:26 |
| cardoe | So what I've started doing is creating VLAN segments for my VXLAN network | 16:30 |
| cardoe | With a physical_network field. | 16:30 |
| cardoe | That at least let's me land on one of two correct ports. But still doesn't let me fully select (there's a different bug for that) | 16:31 |
| cardoe | But now I've got a network segment range for VXLAN for my "primary" VXLAN network. | 16:32 |
| cardoe | tenants go and create a network and they get a VNI assigned from that pool. | 16:33 |
| cardoe | But for my OTHER network, I want to create a VNI as a provider. Since neutron doesn't let me have a physical_network for VXLAN, I cannot have two pools. | 16:33 |
| cardoe | Which the answer might be oh just run two neutrons. But then how do I have Ironic know which Neutron to talk to for port binding? | 16:36 |
| shermanm | tbh "physical network" sounds like the right abstraction to me, all it really defines is: "this set of segmentation ids is logically independent from that set" | 16:43 |
| shermanm | I don't see it as different from the vlan case? You have layer 2 traffic, with protocol specific l2 segment IDs, and you want to specify where those are valid | 16:43 |
| vsaienko | ironic community, I'm checking this failed deployment https://zuul.opendev.org/t/openstack/build/5f838eed9dec42ba83f55713d376c234/logs and it looks like request didn't reach to nova-compute ironic. It stuck somewhere between conductors and superconductor. Are you aware of any opeend bugs? and do we really need to use this deployment topology with cells? When its not a feature of ironic itself | 17:00 |
| rpittau | bye everyone see you on Monday o/ | 17:01 |
| opendevreview | Merged openstack/ironic master: Fix devstack plugin for centos https://review.opendev.org/c/openstack/ironic/+/941720 | 17:02 |
| TheJulia | ... I think it is required | 17:22 |
| masghar | Hello ironic o/ | 17:42 |
| TheJulia | hello! | 17:42 |
| masghar | Wondering if this is still a valid bug: https://bugs.launchpad.net/ironic/+bug/1630319 | 17:42 |
| TheJulia | So, it is valid, but odds are we'll never "fix" it | 17:43 |
| masghar | Because no matter how much we try API errors can not be perfect? | 17:45 |
| masghar | Or for lack of time? | 17:45 |
| JayF | I imagine opportunity cost is a big reason | 17:46 |
| TheJulia | lack of time + desire to avoid breaking changes + it really is trying to drive us towards perfect when we've been around so long that making substantial changes like that also is not ideal for users | 17:46 |
| JayF | even if we had the time to do it; we could use that time to do much more valuable stuff instead | 17:47 |
| TheJulia | yeah | 17:47 |
| masghar | Makes sense! Shall I close it, then? | 17:47 |
| TheJulia | I'd mark it wontfix at this point | 17:47 |
| masghar | Alright, thanks! | 17:47 |
| TheJulia | Thank you! | 17:48 |
| masghar | :) | 17:48 |
| masghar | I think this one, too is not too important to do? https://bugs.launchpad.net/ironic/+bug/1596726 | 18:23 |
| opendevreview | Verification of a change to openstack/ironic master failed: Trivial: Enable disabling tftp setup https://review.opendev.org/c/openstack/ironic/+/941742 | 18:28 |
| JayF | looking | 19:13 |
| JayF | masghar: that's one of those things that likely would be JFDI if it needed to be done; no need for a bug ticket calling for it | 19:14 |
| JayF | (JFDI = just ... do it) | 19:14 |
| JayF | I think we might need to update https://docs.openstack.org/ironic/latest/contributor/dev-quickstart.html#running-unit-tests-locally now that more stuff is integrated into the pep8 job (cc: satoshi) -- I don't think codespell is a target in most places anymore | 19:18 |
| cardoe | So some stupid questions wrt to me fighting ESXi with skrobul and jamesdenton_ | 19:25 |
| cardoe | VMware has their own boot loader which I need to run instead of IPA. | 19:26 |
| cardoe | Just trying to look for the best way to do this in a way that could be upstreamable. | 19:26 |
| JayF | anaconda driver is what you wanna use as the example | 19:26 |
| JayF | it boots an enterprise linux installer | 19:26 |
| JayF | only real anaconda-level integration is a callback in our default ks template | 19:27 |
| cardoe | okay. do we run any tempest tests or anything with it today? | 19:27 |
| cardoe | There's quite a bit of if boot_mode != "kickstart" (do this stuff) else special case. | 19:27 |
| cardoe | I cannot get the anaconda one to work. It always throws an error that stage2_id is missing. | 19:28 |
| JayF | yep there is :( | 19:28 |
| JayF | I knoow the anaconda one worked at some point, I thought we had CI for this but I think it bitrotted | 19:28 |
| cardoe | I've added debug logging and the image properties come back and it sees stage2_id | 19:29 |
| JayF | cern uses it for sure cc: kubajj | 19:29 |
| JayF | maybe kuba has some insight? | 19:29 |
| JayF | > ironic-standalone-anaconda https://zuul.opendev.org/t/openstack/build/27bf264b40764bfabc2e8968f08369a6 : SUCCESS in 52m 19s | 19:30 |
| JayF | runs in ironic-tempest-plugin | 19:30 |
| JayF | but note it's standalone, so if the error is in the nova/ironic handoff, that could 100% be a bug | 19:31 |
| JayF | that could sneak past ci | 19:31 |
| opendevreview | Verification of a change to openstack/ironic-tempest-plugin master failed: Add retries while waiting for SSH on server https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/942009 | 19:39 |
| cardoe | https://opendev.org/openstack/ironic/src/commit/d3c74c6a4bc6917cc97e834d4d8db26ad9d6e905/ironic/drivers/modules/deploy_utils.py#L623 basically that line barfs that stage2_id is missing | 19:44 |
| cardoe | But https://opendev.org/openstack/ironic/src/commit/d3c74c6a4bc6917cc97e834d4d8db26ad9d6e905/ironic/drivers/modules/deploy_utils.py#L543 has it | 19:45 |
| kubajj | JayF: sorry, we just managed to deploy it and first users are using it in production, no insight in particular, just that it works (when I tested it) | 19:46 |
| cardoe | JayF: the other funky bit is that VMware needs a lot more files than just stage2, kernel and ramdisk | 19:51 |
| cardoe | We've got a handful of files with hardcoded things on a glance image like "ks_cfg": <glance uuid> which will write out "ks.cfg" for the node. But then we've got "stage2_id": <glance uuid> which will write out a stage2.$disk_file_extension (where disk_file_extension is another image property) | 19:52 |
| cardoe | I'm wondering if we should have a more generic mechanism? something configured in ironic.conf as to what image properties can be used? | 19:55 |
| cardoe | I was also thinking about virtual media boot. It wouldn't be a hard lift to genisoimage this for Anaconda and VMware. | 19:56 |
| cardoe | In both of those cases, you'd really want a "base ISO" and then you're dropping some customization in. | 19:57 |
| cardoe | I'll annoyingly ping TheJulia as well to weigh in. | 19:57 |
| cardoe | The real answer is probably toss this on the PTG agenda. | 19:57 |
| JayF | 100% a PTG topic | 20:02 |
| JayF | and when I was porting in some of the anaconda driver | 20:02 |
| JayF | it was very clear that there may be space for a generic "installer" driver | 20:02 |
| JayF | but we had zero call for it, and other automatic distro install processes (e.g. ubuntu/debian + presseed) are not as mature | 20:03 |
| JayF | so it wasn't worth it at the time | 20:03 |
| JayF | but I am onboard in theory if we can find a good shape for it | 20:03 |
| TheJulia | wait, what am I chiming in on? | 20:04 |
| TheJulia | Banana bread is tasty, does that work?! | 20:04 |
| cardoe | One day some of these gremlins will be solved and they won't be throwing yaks in front of me to shave. | 20:04 |
| * TheJulia reads | 20:05 | |
| cardoe | TheJulia: TLDR maybe virtual media boot will be better if you bring us all banana bread | 20:05 |
| TheJulia | cardoe: so, regarding more generic mechanisms, I'm all for it. I think i wired it so stage2 could be omitted when I revamped the docs, I think I put both examples in | 20:05 |
| JayF | it's cardoe's giant plan to cause Ironic to implode with irony by adding an esxi driver /s | 20:06 |
| TheJulia | I feel bad, but at the same time anaconda is a hyper-super-hyper specific opinionated path and I'm the wrong person to be driving that ship because I'm just a servant leader | 20:06 |
| TheJulia | so #PatchesWelcome | 20:06 |
| cardoe | I wish I could poke a group at large to drive it as well. | 20:06 |
| TheJulia | eh, I'm more a fan of small groups of strategic things | 20:07 |
| TheJulia | personally :) | 20:07 |
| TheJulia | but seriously, that driver (and also the way RH internally uses anaconda is much more of a multi-stage setup with some lead in to do a whole install. That is obviously different from vmware so things do get a bit more weird | 20:08 |
| TheJulia | but! virtualmedia->banana_bread == happiness() | 20:08 |
| TheJulia | oh, yeah, the skipping existing stage2 is currently contingent on the image_source ending with a path | 20:09 |
| TheJulia | sounds like in your case it might not, or might be entirely omit-able? | 20:09 |
| TheJulia | (which, I guess in theory is toats valid with linux, but pointless maybe with vmware anaconda?!) | 20:10 |
| TheJulia | cardoe: officialy, I wouldn't really have any objection to a "modify an iso" interface to do injection | 20:51 |
| opendevreview | Satoshi Shirosaka proposed openstack/ironic-python-agent master: WIP Add ContainerHardwareManager https://review.opendev.org/c/openstack/ironic-python-agent/+/941714 | 20:51 |
| TheJulia | cardoe: the whole base requirement would really just be how to make it repeatable and abstractable too | 20:51 |
| JayF | So satoshi has been running into an intermittant bug around (seemingly) reboot_requested with his example IPA stuff | 21:09 |
| JayF | Image not found [uuid of correct ipa image] errors on the reboot | 21:09 |
| JayF | *however* I've not been able to reproduce cleanly | 21:09 |
| cardoe | okay I think I feel like I'm taking crazy pills... https://etherpad.opendev.org/p/cardoe-ironic-anaconda-img-props the image properties are nested in another sub key called properties? | 21:12 |
| cardoe | Yeah so that's the question... if you set image properties should be in nested key called "properties" with the parent key of "properties" being a bunch of other stuff. | 21:17 |
| cardoe | vsaienko: since I know you're here... https://etherpad.opendev.org/p/nova-trunk-details-941227 | 21:23 |
| * JayF wonders if it's possible we're accepting aki/ari and then bail on it during a reboot_requested | 21:25 | |
| JayF | seems impossible though | 21:25 |
| cardoe | What I don't know is if Ironic is wrong or if Glance is wrong. | 21:28 |
| JayF | I have a hypothesis, perhaps? | 21:29 |
| JayF | more of either a shot in the dark, or a tendency to wanna blame myself for breakages | 21:29 |
| JayF | https://github.com/openstack/nova/commit/93b90d2b6a58df26f6fbf0ff0fe20cf37fcbf8af | 21:30 |
| JayF | I wouldn't expect that to be breaky, but it absolutely moved some things around | 21:30 |
| cardoe | Well no this is on the "get_image_properties()" call inside of Ironic | 21:31 |
| JayF | okay good | 21:31 |
| cardoe | Which is talking to Glance. | 21:31 |
| JayF | cardoe: oh we are chasing cousin bugs | 21:31 |
| * JayF is trying to find out why satoshi's devstack's glance occassionally goes "WHAT IMAGE? I DON'T KNOW THAT UUID" | 21:32 | |
| JayF | then literally the next second it works | 21:32 |
| JayF | at least the failure is reproducable now | 21:32 |
| cardoe | It's an Ironic thing. | 21:34 |
| cardoe | https://github.com/openstack/ironic/blob/d3c74c6a4bc6917cc97e834d4d8db26ad9d6e905/ironic/common/glance_service/service_utils.py#L26 | 21:35 |
| cardoe | If the "properties" aren't one of those things. It gets moved to a nested "properties" key. | 21:35 |
| cardoe | All of the tests use a flattened properties key like you get from glance. | 21:35 |
| cardoe | The "translate_from_glance()" is skipped in a bunch of tests because they mock the caller of that with the raw data that glance returns | 21:38 |
| JayF | https://etherpad.opendev.org/p/wtf-ipa-reboot-requested is the gist of where I'm at | 21:40 |
| JayF | we're poking in the same area but unrelated breakage | 21:40 |
| cardoe | Yep. It's the same issue. | 21:41 |
| cardoe | You wanna join me and jamesdenton_ on Zoom? we've got like 15 before we have to run. | 21:41 |
| JayF | I can but tbh will be digging my own issue | 21:42 |
| JayF | maybe we luck out and collide | 21:42 |
| JayF | if I accidentally do work for Rackspace past-me is going to whack me in the head though /s :P | 21:42 |
| jamesdenton_ | :D | 21:43 |
| cardoe | nah your issue is different. | 21:43 |
| cardoe | conductor cannot see your image | 21:43 |
| JayF | incoming test with this code change https://usercontent.irccloud-cdn.com/file/g8gQVgGL/image.png | 21:49 |
| * TheJulia returns from lunch | 22:02 | |
| opendevreview | cid proposed openstack/ironic-python-agent-builder master: More reliable TinyIPA builds with network retries https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/942369 | 22:05 |
| opendevreview | Julia Kreger proposed openstack/ironic master: ci: focus ironic-tempest-bios-ipmi-direct-tinyipa https://review.opendev.org/c/openstack/ironic/+/942204 | 22:16 |
| cardoe | TheJulia: you don't wanna do that. | 23:08 |
| cardoe | JayF: managed to confirm that flattening the dual properties into one made the anaconda deploy interface work | 23:08 |
| TheJulia | cardoe: do which item?!? | 23:08 |
| TheJulia | oh... wow | 23:09 |
| JayF | cardoe: https://bugs.launchpad.net/ironic/+bug/2099276 | 23:12 |
| TheJulia | grrrr... now my corgi wants to go for a walk | 23:15 |
| TheJulia | Thanks Jay! ;) | 23:15 |
| JayF | woof :D | 23:19 |
| cardoe | Well maybe. It seems like it doesn’t go onto the deployment network? I dunno getting info relayed. | 23:43 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
