| opendevreview | Takashi Kajinami proposed openstack/ironic-python-agent master: Remove unnecessary +x mode https://review.opendev.org/c/openstack/ironic-python-agent/+/947305 | 03:05 |
|---|---|---|
| opendevreview | Takashi Kajinami proposed openstack/ironic-ui master: Remove unnecessary +x bit https://review.opendev.org/c/openstack/ironic-ui/+/947306 | 03:06 |
| opendevreview | Takashi Kajinami proposed openstack/ironic-python-agent master: Remove unnecessary +x mode https://review.opendev.org/c/openstack/ironic-python-agent/+/947305 | 03:08 |
| opendevreview | Takashi Kajinami proposed openstack/ironic-ui master: Remove unnecessary +x mode https://review.opendev.org/c/openstack/ironic-ui/+/947306 | 03:09 |
| opendevreview | Takashi Kajinami proposed openstack/python-ironicclient master: Remove unnecessary +x mode https://review.opendev.org/c/openstack/python-ironicclient/+/947307 | 03:09 |
| tkajinam | these can be caught once we introduce pre-commit check to these repos but fixing these earlier may be better, I believe | 03:16 |
| opendevreview | Kaifeng Wang proposed openstack/ironic master: Add several NV GPU types https://review.opendev.org/c/openstack/ironic/+/947213 | 03:19 |
| freemanboss[m] | Good morning ironic \o/ | 07:46 |
| queensly[m] | Good morning | 07:47 |
| AmarachiOrdor[m] | Good morning Ironic! | 07:56 |
| rpittau | good morning ironic! o/ | 08:09 |
| arne_wiebalck | good morning rpittau and Ironic o/ | 08:13 |
| rpittau | hey arne_wiebalck :) | 08:14 |
| abongale | Good Morning Ironic! | 08:24 |
| abongale | !!Abhishek@97!! | 08:26 |
| opendevmeet | abongale: Error: "!Abhishek@97!!" is not a valid command. | 08:26 |
| rpittau | abongale: I suggest to change your password :) | 08:28 |
| opendevreview | Amarachi Ordor proposed openstack/bifrost master: docs: updated test guide, improve how-to clarity, add troubleshooting tips https://review.opendev.org/c/openstack/bifrost/+/946953 | 08:29 |
| freemanboss[m] | Hope we all submitted? | 08:30 |
| freemanboss[m] | Best of luck guys | 08:30 |
| abongale | @rpittau I already did :) | 08:33 |
| rpittau | ;) | 08:38 |
| AmarachiOrdor[m] | <freemanboss[m]> "Hope we all submitted?..." <- > <@freemanboss:matrix.org> Hope we all submitted?... (full message at <https://matrix.org/oftc/media/v1/media/download/AR-Ilh4bHnfWw4flkSueQcOeErsuowMbePcDh1EHnlbDmfbKQa0VrO7P_606ff30p40R4MrG085fwnhDg8m3-MpCeWhdmjDgAG1hdHJpeC5vcmcvRWlBbFdOV2dyYVdxa2R1aVprc2dNa0Nt>) | 08:39 |
| opendevreview | Merged openstack/ironic-ui master: Remove unnecessary +x mode https://review.opendev.org/c/openstack/ironic-ui/+/947306 | 09:28 |
| opendevreview | Merged openstack/python-ironicclient master: Remove unnecessary +x mode https://review.opendev.org/c/openstack/python-ironicclient/+/947307 | 09:30 |
| freemanboss[m] | rpittau: cid masghar please I am awaiting a reply on this. Also I am confused to while my test is still failing cause I have done what I understand which is to reduce the line length. https://review.opendev.org/c/openstack/bifrost/+/946724 | 09:57 |
| opendevreview | Merged openstack/ironic stable/2025.1: Improve is_image_available https://review.opendev.org/c/openstack/ironic/+/947276 | 10:06 |
| opendevreview | Merged openstack/sushy-tools master: nova-driver: speed up systems enumeration https://review.opendev.org/c/openstack/sushy-tools/+/946507 | 10:11 |
| opendevreview | Merged openstack/ironic-python-agent master: Remove unnecessary +x mode https://review.opendev.org/c/openstack/ironic-python-agent/+/947305 | 10:31 |
| opendevreview | Satoshi Shirosaka proposed openstack/ironic stable/2024.2: Improve is_image_available https://review.opendev.org/c/openstack/ironic/+/947432 | 10:36 |
| opendevreview | Verification of a change to openstack/ironic-python-agent master failed: Base standalone job on redfish standalone https://review.opendev.org/c/openstack/ironic-python-agent/+/946580 | 10:40 |
| opendevreview | Merged openstack/ironic master: Fix ISO+GPT image handling https://review.opendev.org/c/openstack/ironic/+/946235 | 10:48 |
| Ayo[m] | Freeman Boss: hi there, I replied a comment on opendev and it’s showing the draft pin? Suggestions please | 10:59 |
| queensly[m] | Ayo: you can click on small "a" on your keyboard. A notification will pop up. Click on send. That will send your response and the draft will change to resolved. | 11:01 |
| queensly[m] | Has it been resolved Ayo ? | 11:09 |
| freemanboss[m] | queensly: great reply | 11:11 |
| opendevreview | Satoshi Shirosaka proposed openstack/ironic stable/2024.1: Improve is_image_available https://review.opendev.org/c/openstack/ironic/+/947435 | 11:19 |
| opendevreview | Abhishek Bongale proposed openstack/ironic-tempest-plugin master: fix: fail fast on deploy failure in Anaconda jobs https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/946886 | 12:04 |
| opendevreview | Merged openstack/python-ironicclient master: A new 'description' field to the port object https://review.opendev.org/c/openstack/python-ironicclient/+/944770 | 12:18 |
| opendevreview | Merged openstack/ironic master: Fix mismatch in expected loop functionality https://review.opendev.org/c/openstack/ironic/+/945475 | 12:59 |
| opendevreview | Takashi Kajinami proposed openstack/python-ironic-inspector-client master: Replace UPPER_CONSTRAINTS_FILE https://review.opendev.org/c/openstack/python-ironic-inspector-client/+/947481 | 13:30 |
| TheJulia | Question: Do we have a long term plan regarding python-ironic-inspector-client in the back of our brains? | 13:32 |
| rpittau | TheJulia: AFAIR that's the same as inspector :) | 13:34 |
| TheJulia | well, then how do folks engage with *ironic* for the same features? such as rules? | 13:35 |
| cardoe | So that's kinda what I was asking about as well schema wise for the inspection data. | 13:36 |
| cardoe | Because old inspector had things separated and you could see different data while new is just a TEXT in the DB with json and the fields are YOLO | 13:36 |
| rpittau | TheJulia: I meant moving what we need to ironic-client, as we did from inspector to ironic | 13:56 |
| JayF | That hasn't been done yet?! | 14:12 |
| JayF | I thought we were done with the inspector migration 😔 | 14:12 |
| TheJulia | Differing perceptions on the meaning of "done" | 14:23 |
| TheJulia | I suspect the client can talk to ironic directly, but yeah... | 14:23 |
| JayF | Well, IMO we should prioritize getting inspector "done" with a definition of "any repo matching *inspector* can be retired" :) | 14:39 |
| JayF | separate client to the same API is almost more decoder ring than separate client to separate service lol | 14:39 |
| dtantsur | I think the inspection rules CLI is on review already? | 14:40 |
| JayF | https://review.opendev.org/c/openstack/python-ironicclient/+/937628 | 14:42 |
| JayF | if that's all that's left then we're effectively done pending review | 14:42 |
| TheJulia | FWIW, I wasn't consciously aware of the exact state which is why I asked. | 14:44 |
| JayF | Yeah, I'm more kicking myself for not asking that during PTG :) | 14:44 |
| JayF | plus I'm a little amped up as I've been visited by an ant colony this morning lol | 14:45 |
| opendevreview | Satoshi Shirosaka proposed openstack/ironic stable/2024.1: Improve is_image_available https://review.opendev.org/c/openstack/ironic/+/947435 | 14:57 |
| dtantsur | I hope your new ant overlords were merciful, JayF :D | 15:31 |
| JayF | heh, I have a bunch of ant bait traps waiting for pickup at the store and my pest control dude is coming | 15:32 |
| JayF | I don't play games with insects, they are the real overlords of the earth | 15:32 |
| TheJulia | heh | 15:38 |
| * dtantsur is wondering who will win: AI or insects | 15:39 | |
| TheJulia | insects, of course. | 15:39 |
| dtantsur | in any case, the humanity will lose :( | 15:39 |
| TheJulia | Just think of those electricity loving ants ;) | 15:39 |
| dtantsur | :D | 15:39 |
| TheJulia | ... If I was an evil overlord, I'd setup ant colonies near the massive data centers | 15:41 |
| JayF | I know at one point, we were looking at being able to change deploy interface dynamically (e.g. so I can use direct deploy for images and ramdisk deploy for isos); Did we ever actually get things hooked up for that? (I'm thinking no?) | 16:35 |
| dtantsur | JayF: you can set deploy_interface in instance_info | 16:36 |
| JayF | So it's possible in standalone but not in integrated cases then :\ | 16:37 |
| dtantsur | I guess so, yeah | 16:37 |
| JayF | At least this is about the easiest downstream nova patch ever if we have to do it, but I guess I have another feature design to ponder in the medium/longer term | 16:42 |
| TheJulia | So we need some logic to map flavors to preferences then, at least it sounds like :) | 16:49 |
| TheJulia | err, TRAITS | 16:49 |
| opendevreview | Julia Kreger proposed openstack/metalsmith master: Fix metadata generation https://review.opendev.org/c/openstack/metalsmith/+/947523 | 16:49 |
| TheJulia | Now... to see if that works. | 16:49 |
| JayF | TheJulia: yeah, that's what I figured. A trait or some image metadata we could key on | 16:58 |
| JayF | TheJulia: going to let it background process to ponder since it's not happening (upstream) this cycle anyway | 16:58 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: Make port binding failure (configurably) fatal https://review.opendev.org/c/openstack/ironic/+/699953 | 17:56 |
| JayF | https://review.opendev.org/c/openstack/ironic/+/699953/17#message-b810d0a14fd31988ea2cb597960ea186449c17eb seems like it might be a cousin to what you're fixing TheJulia | 17:58 |
| JayF | cid and I isolated the failure to a log in neutron: Port c7fad300-44fd-4a3c-a7f9-233eca7b36b9 cannot update to ACTIVE because it is not bound | 17:58 |
| JayF | which makes me wonder: 1) does it matter if a port is bound for cleaning/servicing/provisioning and/or 2) what's special about the jobs that are failing to cause them to behave this way | 17:59 |
| JayF | and Refusing to bind due to unsupported vnic_type: baremetal is why the binding fails | 18:06 |
| * JayF wonders if the failing CI job represents a misconfiguration | 18:06 | |
| JayF | metalsmith job failing due to same case in Ironic/Neutron logs, but there's no log around refusing to bind from OVS | 18:11 |
| JayF | nevermind , just found the same log, so we have two jobs failing this way except only one is voting | 18:12 |
| * TheJulia attempts to context load | 18:14 | |
| JayF | Aha, jobs that fail have automated cleaning enabled, jobs that succeed have automated cleaning disabled | 18:14 |
| JayF | tl;dr https://review.opendev.org/c/openstack/ironic/+/699953 reliably fails CI in cleaning situations | 18:15 |
| * JayF is thinking more and more his #1 at 10:59:20 is the right path | 18:15 | |
| TheJulia | sorry, I got completely derailed on a bug with agent going utterly sideways if the *very* first load of steps ends up failing due to a ReadTimeout exception | 18:15 |
| JayF | egad | 18:15 |
| TheJulia | so, let me try to grok this from the beginning | 18:15 |
| JayF | you wanna jump into a sync? | 18:15 |
| JayF | cid and I are actively pairing on it | 18:15 |
| TheJulia | let me try to read what you've got thus far | 18:16 |
| JayF | I'll note for the IPA bit, if you wanna file a bug I can have a look if needed, that's a spot I know relatively well | 18:16 |
| TheJulia | Okay, so.. hmm | 18:16 |
| TheJulia | so, I need to dig at what john's change is doing again, but I *suspect* it needs some more conditional logic, but continuing to grok | 18:17 |
| TheJulia | yeah, you can send garbage and get bad starting state depending on jobs | 18:17 |
| TheJulia | Metalsmith has been doing something naughty all along | 18:17 |
| TheJulia | which, sort of breaks the expectations of flow because of our horrible nested UX nature | 18:17 |
| JayF | is there a reason we should be checking if the port is active/bound in the cleaning/servicing/deplooyment case | 18:18 |
| JayF | that to me is the big question; if no; we know the fix | 18:18 |
| JayF | if yes, there's something broken around that expectation we need to dig | 18:18 |
| JayF | (even if the broken is just "this job is setup wrongly") | 18:18 |
| TheJulia | yeah, I think that is the point I was trying to get at right before release, john's change doesn't account for the other issues and bad state, and we need to stort of get things in a consistent state in advance | 18:18 |
| TheJulia | okay, I think I'm good to acutally talk to humans, let me move back into the motorhome and hook up the mic | 18:19 |
| JayF | https://us06web.zoom.us/j/89137181068?pwd=NlEKVKCdKFZHuuicQZSphyB85Va7eb.1 | 18:20 |
| opendevreview | Julia Kreger proposed openstack/metalsmith master: WIP Fix metadata generation https://review.opendev.org/c/openstack/metalsmith/+/947523 | 18:50 |
| JayF | summary from the call: wow, we have no coverage for neutron networkif+automated cleaning enabled. We're going to get coverage for that case and /hope/ cid's change passes it | 18:58 |
| JayF | if so, we'll likely exclude flat network interface from the port active/bound check | 18:59 |
| frickler | JayF: random antsy thing I just stumbled across https://sauropods.win/@futurebird/114344727566425769 | 19:13 |
| JayF | I will tell you it's a lot more disturbing to stumble across a few dozen ants in your (clean!) kitchen lol | 19:14 |
| TheJulia | yeah, looking at the failure, and the discussion, JayF I'm more convinced now that is the right path, which then also helps limit blast scoping | 19:15 |
| * TheJulia waits for results to try and undestand what she got wrong about the metalsmith change | 19:15 | |
| opendevreview | Jay Faulkner proposed openstack/ironic stable/2025.1: Fix mismatch in expected loop functionality https://review.opendev.org/c/openstack/ironic/+/947533 | 19:38 |
| opendevreview | Julia Kreger proposed openstack/metalsmith master: WIP Fix metadata generation https://review.opendev.org/c/openstack/metalsmith/+/947523 | 19:44 |
| opendevreview | cid proposed openstack/ironic master: CI: Coverage for neutron with automated cleaning https://review.opendev.org/c/openstack/ironic/+/947535 | 19:58 |
| opendevreview | cid proposed openstack/ironic master: Make port binding failure (configurably) fatal https://review.opendev.org/c/openstack/ironic/+/699953 | 19:59 |
| cid | JayF, TheJulia: is this ^^ at all what we discussed? :) | 20:00 |
| JayF | lookin | 20:00 |
| JayF | I *think* ideally it wouldn't be an OVN job | 20:00 |
| JayF | but don't change it on that conjecture | 20:01 |
| * cid waits then ... | 20:02 | |
| TheJulia | I think so, I sort of would have expected it to be done together since they are disjointed changes | 20:02 |
| TheJulia | change 699953 needs updating though | 20:05 |
| cid | ++ | 20:37 |
| opendevreview | Julia Kreger proposed openstack/metalsmith master: WIP Fix metadata generation https://review.opendev.org/c/openstack/metalsmith/+/947523 | 20:54 |
| opendevreview | Julia Kreger proposed openstack/metalsmith master: Fix metadata generation https://review.opendev.org/c/openstack/metalsmith/+/947523 | 20:54 |
| TheJulia | Okay, I think that addresses metalsmith. No more on boot cloud-init errors screaming about mac addresses | 20:55 |
| TheJulia | now... .. the fun one | 20:55 |
| TheJulia | oh, wait | 20:55 |
| TheJulia | need to stack it with the other change | 20:55 |
| opendevreview | Julia Kreger proposed openstack/ironic master: provide host_id to neutron early on https://review.opendev.org/c/openstack/ironic/+/946378 | 20:56 |
| TheJulia | okay, thats the true test but it looks like the metalsmith patch was good | 20:56 |
| opendevreview | cid proposed openstack/ironic master: Make port binding failure (configurably) fatal https://review.opendev.org/c/openstack/ironic/+/699953 | 22:10 |
| TheJulia | woot, metalsmith passed | 22:54 |
| JayF | What's the story on indicator API support in the management interface? I noticed it wasn't really hooked up to any clients, and when I tried it against an ILO 6 in the real world, I got an empty list returned when I know that indicators exist in the web UI | 23:05 |
| JayF | The code is 6 years old, I'm wondering if anyone knows if it's a "it probably moved" or "it's not modeled in HPE Redfish" issue (I don't have direct BMC access to poke it) | 23:07 |
| TheJulia | Ilya was working on it before he went into the hospital | 23:11 |
| TheJulia | If memory serves, he modeled it on dell gear for redfish | 23:11 |
| TheJulia | I started to the generic indicator on ipmi at some point maybe 3-4 years ago | 23:11 |
| TheJulia | I don't remember what happened there at this point (doesn't help I have a migraine righ tnow) | 23:11 |
| JayF | it's landed in ironic, it runs without err and finds nothing | 23:15 |
| JayF | I'll do a little recon. | 23:15 |
| JayF | Not super important or urgent but interesting for sure and maybe a quick win | 23:15 |
| JayF | a "add support for this api endpoint to an sdk" task might be a welcome break a month into writing network code | 23:16 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!