| opendevreview | Merged openstack/ironic-python-agent master: Fix errors in the function erase_devices_express https://review.opendev.org/c/openstack/ironic-python-agent/+/939803 | 01:21 |
|---|---|---|
| opendevreview | Steve Baker proposed openstack/ironic master: Mask all driver_internal_info in node output https://review.opendev.org/c/openstack/ironic/+/939504 | 03:10 |
| opendevreview | Steve Baker proposed openstack/ironic master: Utility functions for graphical console drivers https://review.opendev.org/c/openstack/ironic/+/939505 | 03:10 |
| opendevreview | Steve Baker proposed openstack/ironic master: Add novnc-proxy service https://review.opendev.org/c/openstack/ironic/+/939191 | 03:10 |
| opendevreview | Steve Baker proposed openstack/ironic master: WIP add VNC auth type to novnc-proxy https://review.opendev.org/c/openstack/ironic/+/939192 | 03:10 |
| opendevreview | Steve Baker proposed openstack/ironic master: WIP create idrac graphical console driver https://review.opendev.org/c/openstack/ironic/+/939193 | 03:10 |
| opendevreview | cid proposed openstack/ironic master: Apply Rules: inspection rules migration https://review.opendev.org/c/openstack/ironic/+/939218 | 07:42 |
| opendevreview | cid proposed openstack/ironic master: API/Testing: Inspection rules migration https://review.opendev.org/c/openstack/ironic/+/939217 | 07:42 |
| opendevreview | Jakub Jelinek proposed openstack/ironic-python-agent stable/2024.2: Fix errors in the function erase_devices_express https://review.opendev.org/c/openstack/ironic-python-agent/+/939938 | 07:56 |
| opendevreview | Verification of a change to openstack/ironic-python-agent stable/2024.2 failed: Fix RAID volume name https://review.opendev.org/c/openstack/ironic-python-agent/+/939700 | 08:15 |
| rpittau | good morning ironic! o/ | 08:36 |
| kubajj | good morning rpittau, and ironic! o/ | 09:00 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Run metal3 integration job using UEFI boot (default) https://review.opendev.org/c/openstack/ironic/+/939694 | 09:13 |
| opendevreview | Merged openstack/ironic-python-agent stable/2024.2: Fix RAID volume name https://review.opendev.org/c/openstack/ironic-python-agent/+/939700 | 12:29 |
| opendevreview | cid proposed openstack/python-ironicclient master: [WIP]Add CLI support for migrated inspection rules https://review.opendev.org/c/openstack/python-ironicclient/+/937628 | 12:30 |
| opendevreview | cid proposed openstack/python-ironicclient master: [WIP]Add CLI support for migrated inspection rules https://review.opendev.org/c/openstack/python-ironicclient/+/937628 | 12:34 |
| opendevreview | cid proposed openstack/ironic master: DB: inspection rules migration https://review.opendev.org/c/openstack/ironic/+/939318 | 12:38 |
| opendevreview | cid proposed openstack/ironic master: Apply Rules: inspection rules migration https://review.opendev.org/c/openstack/ironic/+/939218 | 12:38 |
| opendevreview | cid proposed openstack/ironic master: API/Testing: Inspection rules migration https://review.opendev.org/c/openstack/ironic/+/939217 | 12:38 |
| opendevreview | Jakub Jelinek proposed openstack/ironic-python-agent stable/2024.1: Fix RAID volume name https://review.opendev.org/c/openstack/ironic-python-agent/+/939959 | 12:38 |
| opendevreview | Jakub Jelinek proposed openstack/ironic-python-agent stable/2023.2: Fix RAID volume name https://review.opendev.org/c/openstack/ironic-python-agent/+/939960 | 12:42 |
| TheJulia | good morning | 14:13 |
| TheJulia | looks like the metal3 job might not be happy again? | 14:32 |
| rpittau | TheJulia: yep, looking into it | 14:38 |
| rpittau | some weird credentials issue | 14:38 |
| TheJulia | joy | 14:39 |
| TheJulia | rpittau: so I figured out the start of the v6 job vailure | 14:39 |
| TheJulia | err, failure | 14:39 |
| TheJulia | goes back to the change for Noble | 14:39 |
| rpittau | ah | 14:39 |
| TheJulia | Was there any specific reason you took the bios boot flag off the job? | 14:39 |
| rpittau | it's non-voting :/ | 14:39 |
| TheJulia | Yeah | 14:40 |
| TheJulia | I think I figured out one of the issues which was causing failures which kept it non-voting to begin with | 14:40 |
| rpittau | mmm let me think about it | 14:40 |
| TheJulia | no rush, I'm re-building a bios stack locally now to try and reproduce it because in bios mode it just doesn't work either now | 14:41 |
| TheJulia | in weird/annoying/frustrating ways :) | 14:41 |
| rpittau | so I removed the bios flag I guess because OVMF should work with both ipv6/ipv4 noew | 14:43 |
| rpittau | I guess my guess was wrong :D | 14:43 |
| TheJulia | it works sometimes | 14:43 |
| TheJulia | ... kind of | 14:43 |
| rpittau | lol | 14:43 |
| TheJulia | and can like take close to 10 minutes to boot | 14:43 |
| TheJulia | and has a giant pile of issues | 14:44 |
| rpittau | \o/ | 14:44 |
| rpittau | TheJulia: metal3 job should actually work now in legacy mode, what's the latest failure you saw ? | 14:45 |
| TheJulia | https://zuul.opendev.org/t/openstack/build/614ef512ea504506bc4437caef98c496 | 14:46 |
| rpittau | ok, looking there then | 14:48 |
| rpittau | lol again | 14:48 |
| rpittau | Failed to deploy instance: Disk volume where '/shared/tftpboot/master_images' is located doesn't have enough disk space. Required 512 MiB, only 409 MiB available space present | 14:48 |
| iurygregory | redfish standards strikes again, yay \o/ | 14:50 |
| TheJulia | huh? | 14:51 |
| iurygregory | we have a bug from Intel Hardware... | 14:52 |
| iurygregory | :D | 14:52 |
| iurygregory | which doesn't like when we insert the virtual media using https... | 14:52 |
| * TheJulia blinks | 14:53 | |
| TheJulia | iurygregory: jssfr has some intel gear which requires nfs... so... yeah :( | 14:53 |
| iurygregory | SMB for this one it seems =( | 14:54 |
| TheJulia | that seems... awful. | 14:54 |
| iurygregory | TheJulia, yeah! | 14:55 |
| iurygregory | and we don't have support for that XD | 14:55 |
| jssfr | iurygregory, oh! | 14:56 |
| jssfr | ours can do either SMB or NFS | 14:56 |
| jssfr | (but not https....) | 14:56 |
| iurygregory | jssfr, maybe after they upgrade the FW it can be NFS =D | 14:56 |
| iurygregory | jssfr, do you use vmedia on it and via ironic or nope? | 14:57 |
| jssfr | iurygregory, do you have logs from this room from 2025-01-16T10:56:22Z | 14:57 |
| jssfr | +? | 14:57 |
| jssfr | we discussed it back then | 14:58 |
| jssfr | iurygregory, we can't use it with ironic (yet!) because ironic can't offer NFS :) | 14:58 |
| iurygregory | not sure, but I can check the irc logs | 14:58 |
| jssfr | but I intend to write patches in the next months or so. | 14:58 |
| jssfr | iurygregory, I'll paste it, sec | 14:58 |
| iurygregory | jssfr, got it =) I had this feeling | 14:58 |
| TheJulia | so, somehow my local test machine is not running noble even though I'm 95% sure I used a noble ISO | 14:59 |
| iurygregory | not even sure what I should tell the customer now :D | 15:00 |
| iurygregory | TheJulia, huh?! O.o maybe it boot from the disk? | 15:00 |
| TheJulia | no idea, on both counts | 15:00 |
| jssfr | iurygregory, https://paste.debian.net/hidden/c8f722ee/ | 15:00 |
| jssfr | iurygregory, so the basic plan is to detect the error condition, and if it happens, substitute the http_url with some nfs_url template | 15:01 |
| TheJulia | iurygregory: well, regarding the customer, likely going to have to tell them "unfortunately, it seems intel's virtual media requires use of CIFS or NFS, which is not presently supported" | 15:01 |
| iurygregory | TheJulia, yeah... | 15:02 |
| iurygregory | jssfr, nice! | 15:02 |
| jssfr | iurygregory, but as I wrote yesterday or so, we're currently working around it a little and we will only start actually implementing it around end of february I guess. | 15:03 |
| iurygregory | I like that they have "CIFS" "HTTPS" "NFS" in their docs https://www.intel.com/content/dam/support/us/en/documents/server-products/intel-server-obmc-redfish-interface.pdf but they also have " This action is used to attach remote media to virtual media. Media insertion currently supports SMB protocol only. Username and password are required fields in the URL" | 15:03 |
| iurygregory | jssfr, ack o/ tks for the info | 15:03 |
| jssfr | iurygregory, oof, that's more recent than what we have | 15:04 |
| jssfr | so I guess we should go for CIFS straight away :( | 15:04 |
| jssfr | *sigh* I don't want to have to debug CIFS authentication (and that seems to inevitably always happen when I touch CIFS) | 15:05 |
| rpittau | jssfr: Intel docs are also contradictory | 15:05 |
| iurygregory | funny that they don't say SMB in the table only in the text :D | 15:05 |
| rpittau | check page 72 here https://www.intel.com/content/dam/support/us/en/documents/server-products/intel-server-obmc-redfish-interface.pdf | 15:05 |
| jssfr | I mean ours definitely supports NFS---I tried that and it works | 15:05 |
| TheJulia | cifs can also be done anonymously, fwiw | 15:05 |
| rpittau | TheJulia: not in this case :) | 15:05 |
| TheJulia | https://media3.giphy.com/media/v1.Y2lkPTc5MGI3NjExY2JraGZjZHVzY3J4bnVpZ29sdHAwMWdjYWtxejBudmliZXVwZmE1YSZlcD12MV9pbnRlcm5hbF9naWZfYnlfaWQmY3Q9Zw/3oEdTYo6XY8CZ4LFQs/giphy.gif | 15:05 |
| rpittau | anyway, another nice war story | 15:06 |
| * TheJulia had the gif on standby | 15:06 | |
| rpittau | :D | 15:06 |
| iurygregory | yeah | 15:06 |
| iurygregory | one more chapter to our book of hardware horror | 15:06 |
| rpittau | we should sell the rights to make a series, it would run for at least 10 seasons | 15:07 |
| TheJulia | oh, It would be perpetual | 15:07 |
| TheJulia | it would end up like the Simpsons | 15:07 |
| iurygregory | ++ | 15:08 |
| rpittau | :D | 15:10 |
| rpittau | "we're gonna need a bigger boat" /dev/xvda1 37G 35G 404M 99% / | 15:29 |
| rpittau | TheJulia: I think the issue with metal3 job is not the job itself, we're not getting consistent hardware resources | 15:30 |
| TheJulia | is that *all* of the storage on that machine? | 15:30 |
| rpittau | this is form a successful job /dev/vda1 75G 35G 37G 50% / | 15:30 |
| JayF | The rackspace machines have an ephemeral partition | 15:31 |
| JayF | that can be used during ci runs iirc | 15:31 |
| rpittau | and the failing one is indeed rackspace | 15:32 |
| rpittau | althoough I don't see any other partition available | 15:33 |
| JayF | check lsblk | 15:33 |
| JayF | it may not be formatted | 15:33 |
| rpittau | I do not have lsblk in metal3 job \o/ | 15:35 |
| JayF | sfdisk --dump? | 15:35 |
| rpittau | nope | 15:35 |
| rpittau | well I will add the output | 15:35 |
| rpittau | and now I remember that we always run tinyipa on rax because of the fewer resources | 15:36 |
| JayF | we can ask in -infra what disk space is provided | 15:36 |
| JayF | and how to get more | 15:36 |
| JayF | they'd at least actually know whereas I'm guessing :P | 15:36 |
| rpittau | yep, we don't need too much more, 10 GB more is enough | 15:36 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Add lsblk output to metal3 logs https://review.opendev.org/c/openstack/ironic/+/939976 | 15:40 |
| TheJulia | so, it looks like noble's ipxe roms for bios mode don't support ipv6.... | 15:42 |
| * TheJulia goes and walks the corgi overlord | 15:42 | |
| rpittau | good night! o/ | 17:06 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: [doc] Add node history to admin guide https://review.opendev.org/c/openstack/ironic/+/940061 | 17:20 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: [doc] Add project-awareness to api-ref concepts https://review.opendev.org/c/openstack/ironic/+/940063 | 17:45 |
| JayF | Going to be honest, not 100% sure ^^^ is a good idea, or written well, but it always feels like a glaring omission so I took a swing at it | 17:46 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: [doc] Add project-awareness to api-ref concepts https://review.opendev.org/c/openstack/ironic/+/940063 | 17:48 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CI: Remove IPv6 testing https://review.opendev.org/c/openstack/ironic/+/940072 | 19:59 |
| TheJulia | ^^^ feels like a gut punch, but need to bre realistic | 20:00 |
| JayF | TheJulia: Are those edk2 bugs the primary blocker? | 20:04 |
| JayF | TheJulia: if so, going to see if I can get someone GR-OSS pointed at it so maybe it won't be gone forever | 20:04 |
| TheJulia | JayF: pretty much covers it | 20:05 |
| TheJulia | hjensas: see ^^^, I linked to two of your bugs | 20:05 |
| TheJulia | you opened way back in 2019 | 20:05 |
| JayF | sgtm, I'm asking. No promises but sometimes this works :) | 20:05 |
| TheJulia | Yeah, I think the base problem is likely there is minimal edk2 level of caring regarding ipv6 | 20:06 |
| rm_work[m] | <JayF> "rm_work: worth noting the flip..." <- ah ok, so priority 0 is how you make something manual, and any priority above that WILL be automatic? | 20:48 |
| JayF | yers | 20:56 |
| JayF | also for automatic, it can't take arguments | 20:56 |
| JayF | because there's nobody to pass the arguments | 20:56 |
| JayF | one pattern I've seen: step_that_takes_args priority:zero; step_with_builtin_default_args priority:nonzero; and step_with_builtin_default_args just calls the step_that_takes_args with some default args | 20:57 |
| rm_work[m] | right | 20:59 |
| rm_work[m] | cool, ok | 20:59 |
| JayF | also note that get_deploy_steps / get_clean_steps / get_service_steps are separate methods | 21:00 |
| rm_work[m] | yeah I see that if I don't define the method it just defaults to an empty list in the base manager class | 21:04 |
| JayF | (we don't have anyone who would be a good fit to look at those edk2/ipv6 issues it appears) | 21:38 |
| *** tosky_ is now known as tosky | 21:41 | |
| -opendevstatus- NOTICE: The Gerrit service on review.opendev.org will be offline momentarily while we reboot for a patch version upgrade of the software, but should return again within a few minutes | 22:45 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
