| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Redfish UefiHttp boot support https://review.opendev.org/c/openstack/ironic/+/900964 | 00:16 |
|---|---|---|
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Add HTTP versions of network boot interfaces https://review.opendev.org/c/openstack/ironic/+/900965 | 00:16 |
| TheJulia | heh, yeah... nope :) | 00:17 |
| TheJulia | so the latter one goes a bit further than the spec, and the first one kind of went a little more specific off from the spec when I realized "this would just be better to let sushy handle the details | 00:17 |
| opendevreview | Boushra Sondos Bettir proposed openstack/ironic master: [WIP] Currently working on adding support for OVN type switches which Neutron already supports. https://review.opendev.org/c/openstack/ironic/+/900568 | 01:09 |
| opendevreview | Verification of a change to openstack/ironic master failed: [api-ref] Complete port name and shard documentation https://review.opendev.org/c/openstack/ironic/+/899097 | 01:12 |
| opendevreview | Michal Nasiadka proposed openstack/ironic-python-agent master: docs: improve rootpwd password generation command https://review.opendev.org/c/openstack/ironic-python-agent/+/899380 | 07:52 |
| rpittau | good morning ironic! o/ | 08:12 |
| opendevreview | Michal Nasiadka proposed openstack/bifrost master: inspector: add console=ttyS0 by default only when testing https://review.opendev.org/c/openstack/bifrost/+/901028 | 11:52 |
| dtantsur | JayF: "just run it locally" is what I used to do fwiw | 13:27 |
| opendevreview | Graeme Moss proposed openstack/networking-generic-switch stable/zed: Fix regression plugging 802.3ad port group https://review.opendev.org/c/openstack/networking-generic-switch/+/900989 | 13:27 |
| TheJulia | good morning | 14:11 |
| opendevreview | Graeme Moss proposed openstack/networking-generic-switch stable/zed: Fix regression plugging 802.3ad port group https://review.opendev.org/c/openstack/networking-generic-switch/+/900989 | 14:11 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Generic API for attaching/detaching virtual media https://review.opendev.org/c/openstack/ironic/+/894918 | 15:12 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Generic API for attaching/detaching virtual media https://review.opendev.org/c/openstack/ironic/+/894918 | 15:16 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Generic API for attaching/detaching virtual media https://review.opendev.org/c/openstack/ironic/+/894918 | 15:16 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Generic API for attaching/detaching virtual media https://review.opendev.org/c/openstack/ironic/+/894918 | 15:53 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Generic API for attaching/detaching virtual media https://review.opendev.org/c/openstack/ironic/+/894918 | 15:54 |
| TheJulia | Do we know what is going on with the metal3 integration job, looks like the playbook run is failing on a missing package | 15:57 |
| TheJulia | like... trying to pull "https://download.opensuse.org/repositories/devel:/kubic:/libcontainers:/stable/xUbuntu_22.04/amd64/libcap2-bin_2.48-1_amd64.deb" | 15:57 |
| rpittau | JayF, TheJulia, dtantsur, the patch for the bugfix eol is passing CI now, we need to skip some validations though as all the release tools are "stable-branches-centric" https://review.opendev.org/c/openstack/releases/+/900810 | 15:58 |
| rpittau | depending on the support from the release team we could think about adding the missing validations, but it doesn't look trivial | 15:58 |
| JayF | ack; thank you | 15:59 |
| dtantsur | TheJulia: interesting, not such failures are knows in metal3 | 16:17 |
| dtantsur | rpittau: I think the infra tooling will expect - instead of / | 16:18 |
| rpittau | dtantsur: uhmmm where? | 16:27 |
| dtantsur | rpittau: you're creating tag bugfix/X.Y-eol | 16:27 |
| dtantsur | I think it's meant to be bugfix-X.Y-eol | 16:27 |
| rpittau | oh, so bugfix-X.Y-eol ? | 16:27 |
| rpittau | ok, I'll double-check that | 16:28 |
| rpittau | thanks | 16:28 |
| JayF | dtantsur: are you coming to doc meeting? | 16:32 |
| dtantsur | sorry, yeah | 16:32 |
| opendevreview | Graeme Moss proposed openstack/networking-generic-switch stable/zed: Add support for Dell OS10 https://review.opendev.org/c/openstack/networking-generic-switch/+/901061 | 17:06 |
| JayF | Hey; we just had a small chat about docs, as a followup to PTG, notes here: https://etherpad.opendev.org/p/ironic-docs-2023-11-15 -- basically one of my downstream peers at GR-OSS is going to work to find resources to help improve our docs. Only thing we need to do now is if we know of examples of "good" documentation in OSS (closer to our problem set the better -- so maybe | 17:11 |
| JayF | other openstack or bm provisioning projects?), we should add them to the proper place on that etherpad (line 28-ish) | 17:11 |
| TheJulia | FWIW, it is a perception problem. We've had people also say in the past our docs are extremely verbose and detailed | 17:17 |
| rpittau | JayF: thanks for organizing that :) | 17:17 |
| TheJulia | ... possibly to the underlying problem, how to find what your looking for, maybe? | 17:18 |
| TheJulia | dunno | 17:18 |
| * TheJulia goes back to unit tests | 17:18 | |
| rpittau | and time to go for me, see ya tomorrow o/ | 17:19 |
| JayF | TheJulia: I think it's a levelling problem. We have a lot of docs at the super duper duper basic level, a lot of docs at the super duper detailed level. The middle is a pretty sizable gap. Plus a pretty bad discovery issue. | 17:21 |
| TheJulia | yup | 17:21 |
| JayF | TheJulia: Given docs is almost a form of marketing; perception issues with docs /are/ docs issues :) | 17:21 |
| TheJulia | agree completely | 17:22 |
| JayF | This is the stuff Caterina is good at, I'm hoping she'll help in the right kinda way | 17:22 |
| JayF | getting someone outside Ironic to look is probably the thing we've needed for a while | 17:22 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: DNM Testing nova change https://review.opendev.org/c/openstack/ironic/+/900958 | 17:28 |
| *** JasonF is now known as JayF | 18:47 | |
| opendevreview | Julia Kreger proposed openstack/ironic master: Redfish UefiHttp boot support https://review.opendev.org/c/openstack/ironic/+/900964 | 19:24 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: DNM Testing nova change https://review.opendev.org/c/openstack/ironic/+/900958 | 19:27 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Redfish UefiHttp boot support https://review.opendev.org/c/openstack/ironic/+/900964 | 19:29 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM: Add redfish https CI job https://review.opendev.org/c/openstack/ironic/+/901090 | 19:29 |
| TheJulia | crazy question, what if we had a tempest test which tried to deploy on *every* boot_interface configured? | 19:31 |
| JayF | we'd have 'test coverage', I believe, as the kids say :D | 19:31 |
| * JayF not sure what you're asking/maybe missing context | 19:31 | |
| TheJulia | heh | 19:32 |
| TheJulia | so what I'm sort of thinking with https booting... is today say we have "pxe (grub... please use grub), ipxe, redfish-virtual-media, and now adding redfish-https, and https-ipxe, and https-pxe to that list, *feels* like we should just rotate through them instead of having separate jobs | 19:34 |
| JayF | rotate meaning | 19:34 |
| JayF | order by rand() | 19:34 |
| JayF | or meaning, we do N boots for N interfaces in N jobs | 19:34 |
| TheJulia | we would likely need to do some test magic maybe | 19:34 |
| TheJulia | but kind of | 19:34 |
| JayF | or we do N boots in N interfaces in *1* job | 19:34 |
| TheJulia | n test classes or n tests inside of a test class | 19:34 |
| JayF | I wonder if like | 19:34 |
| JayF | we can segment some of this | 19:34 |
| JayF | e.g. | 19:34 |
| JayF | boot interface needs to just boot a thing | 19:35 |
| JayF | that screams "use ramdisk deploy", right? | 19:35 |
| JayF | if we use ramdisk deploy + standalone, how long can a boot interface test take? | 19:35 |
| TheJulia | potentially | 19:35 |
| JayF | then we could have a "test all the boot interfaces job" that tries hard to NOT touch anything else | 19:35 |
| TheJulia | I'm not sure the PXE interface inherently does ramdisk deploy out of the box, bu it would likely be a quick fix if it doesn't | 19:35 |
| TheJulia | so the one challenge with only verifying it becomes pingable, is the machine is pingable via EFI firmware | 19:36 |
| JayF | which while maybe not working from a "stuff as much coverage into as few jobs as possible" standpoint, would help from a "actually able to tell what breaks when a job fails" standpoint | 19:36 |
| TheJulia | or iPXE | 19:36 |
| TheJulia | so we don't *really* know without getting to a steady state | 19:36 |
| JayF | ramdisk doesn't preclude ssh, does it? | 19:36 |
| TheJulia | today, we don't try to wait that long because we have no key injection means | 19:37 |
| TheJulia | keep in mind, it has to be completely ready to rock and roll and we're basically hands-off | 19:37 |
| JayF | In our current devstack model; yes. We have the technology and knowledge to e.g. toss an image/kernel/initrd into an ISO and use it for a ramdisk-based deploy, (with something injected) yes? | 19:38 |
| TheJulia | vmedia we can sort of ping and be sure, we know the bootloader-> OS is causing networking to fire up | 19:38 |
| TheJulia | we would have to inject some means to know what it is | 19:38 |
| JayF | well I'm thinking, if we could even get something like a port open inside the ramdisk | 19:38 |
| JayF | just literally nc -l 9999 or something nuts like that | 19:39 |
| TheJulia | or verify it, or ssh in, or just fire up cirros and ensure cubsw1n;) | 19:39 |
| JayF | heck, even *look for the ssh port being open* | 19:39 |
| JayF | there are lots of canaries that we've booted a thing, when we take agent as a possibility out of it | 19:39 |
| JayF | assuming tempest has more than just a menu of "ping or ssh" | 19:39 |
| JayF | because you just need to answer "is this running a real OS?" | 19:39 |
| JayF | especially if it's an isolated job where it's impossible that there would've been another OS on that disk, ever | 19:40 |
| JayF | heck, even provision diskless VMs even | 19:40 |
| JayF | (if we wanna be certain certain) | 19:40 |
| TheJulia | likely best not to over-think it too far beyond the base idea | 19:53 |
| JayF | we don't have to do any of that stuff, there are just lots of intersting things we could do to go about it | 19:56 |
| TheJulia | indeed | 20:02 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: DNM Testing nova change https://review.opendev.org/c/openstack/ironic/+/900958 | 21:27 |
| opendevreview | Verification of a change to openstack/ironic-inspector stable/xena failed: Handle bracketed IPv6 redfish_address https://review.opendev.org/c/openstack/ironic-inspector/+/896326 | 22:24 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Redfish UefiHttp boot support https://review.opendev.org/c/openstack/ironic/+/900964 | 23:15 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM: Add redfish https CI job https://review.opendev.org/c/openstack/ironic/+/901090 | 23:15 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!