| opendevreview | Doug Goldstein proposed openstack/ironic stable/2025.2: Prevent power sync to interfere with FirmwareUpdate https://review.opendev.org/c/openstack/ironic/+/977745 | 03:57 |
|---|---|---|
| opendevreview | OpenStack Proposal Bot proposed openstack/ironic master: Imported Translations from Zanata https://review.opendev.org/c/openstack/ironic/+/977566 | 04:00 |
| opendevreview | Doug Goldstein proposed openstack/ironic bugfix/34.0: Prevent power sync to interfere with FirmwareUpdate https://review.opendev.org/c/openstack/ironic/+/977749 | 04:46 |
| opendevreview | Verification of a change to openstack/networking-generic-switch master failed: l2vni plug case with Cisco NXOS https://review.opendev.org/c/openstack/networking-generic-switch/+/968377 | 05:46 |
| opendevreview | Merged openstack/virtualbmc master: Remove Python 3.9 support https://review.opendev.org/c/openstack/virtualbmc/+/977676 | 06:10 |
| rpittau | good morning ironic! o/ | 09:01 |
| rpittau | final ironicclient release has been proposed | 09:03 |
| rpittau | we have 4 patches to emrge if we want to add the type annotations | 09:03 |
| rpittau | starting chain with https://review.opendev.org/c/openstack/python-ironicclient/+/976885 | 09:03 |
| opendevreview | Merged openstack/ironic bugfix/34.0: Prevent power sync to interfere with FirmwareUpdate https://review.opendev.org/c/openstack/ironic/+/977749 | 09:09 |
| opendevreview | Merged openstack/ironic stable/2025.2: ci: Pin `setuptools` to a range that still ships https://review.opendev.org/c/openstack/ironic/+/977457 | 09:11 |
| opendevreview | nidhi proposed openstack/ironic master: Add storage controller discovery to Redfish inspection https://review.opendev.org/c/openstack/ironic/+/976626 | 09:16 |
| opendevreview | cid proposed openstack/virtualbmc master: Multiprocessing pickling error in vbmcd https://review.opendev.org/c/openstack/virtualbmc/+/977627 | 10:27 |
| opendevreview | Merged openstack/python-ironicclient master: Add type annotations to `ironicclient/common/i18n.py` https://review.opendev.org/c/openstack/python-ironicclient/+/976885 | 10:38 |
| opendevreview | Merged openstack/python-ironicclient master: Add type annotations to `ironicclient/common/filecache.py` https://review.opendev.org/c/openstack/python-ironicclient/+/977496 | 10:47 |
| opendevreview | Merged openstack/virtualbmc master: Multiprocessing pickling error in vbmcd https://review.opendev.org/c/openstack/virtualbmc/+/977627 | 14:30 |
| opendevreview | Julia Kreger proposed openstack/ironic master: ci: Explictly move neutron dhcp config to later https://review.opendev.org/c/openstack/ironic/+/977806 | 14:34 |
| TheJulia | cardoe: that is what I was thinking, I ;eft a comment on it | 14:35 |
| opendevreview | Merged openstack/sushy master: Handle missing RelatedProperties in TransferProtocolType detection https://review.opendev.org/c/openstack/sushy/+/977723 | 14:35 |
| TheJulia | cardoe: left a comment on https://review.opendev.org/c/openstack/python-ironicclient/+/977584/1 | 14:38 |
| TheJulia | So, change https://review.opendev.org/c/openstack/ironic/+/977806 is my next attempt to fix the neutron start race condition, commit message explains why my original change was wrong (because, its a race condition) | 14:39 |
| cardoe | TheJulia: yeah that's why I didn't -1 it. It's just mechanical whatever. I'll +W it. | 14:54 |
| cardoe | TheJulia: https://review.opendev.org/c/openstack/ironic/+/977569 I tried it last night... I removed the touch_provisioning call... it worked.... but it also failed.. might ask ya some questions when we get on. | 14:59 |
| cardoe | rpittau: https://review.opendev.org/c/openstack/python-ironicclient/+/977406 is that okay to backport? | 14:59 |
| rpittau | cardoe: I think so, yeah | 15:00 |
| dtantsur | cardoe, janders, you two are working on the same thing apparently (re https://review.opendev.org/c/openstack/ironic/+/977569) | 15:03 |
| dtantsur | (commented on touch_provisioning) | 15:04 |
| cardoe | hrm I spoke to janders previously and didn't think he was working on the BIOS settings without IPA. | 15:05 |
| dtantsur | he was not, now he is :) not your fault, just making sure he sees your work | 15:06 |
| cardoe | So I ran it locally without touch_provisioning() and it worked. But it reported failure. | 15:06 |
| TheJulia | That is likely the caller still thinking it is needed, I guess | 15:07 |
| cardoe | So where it went sideways... it didn't actually "complete" until the box attempted to boot off the disk or PXE (not sure which cause of timing). Not sure if we want to wait that long or if I should instead wait for the HW_INIT_COMPLETE state instead... | 15:07 |
| dtantsur | It's needed only in one odd case: when the process can run longer than the CLEANWAIT timeout | 15:07 |
| opendevreview | Merged openstack/python-ironicclient master: Add type annotations to `ironicclient/common/apiclient/exceptions.py` https://review.opendev.org/c/openstack/python-ironicclient/+/977584 | 15:08 |
| opendevreview | Merged openstack/python-ironicclient master: Add type annotations to `ironicclient/common/apiclient/base.py` https://review.opendev.org/c/openstack/python-ironicclient/+/977588 | 15:08 |
| TheJulia | I'd be worried its just not working if it is taking *that* long to change settings, fwiw. | 15:08 |
| cardoe | So touch_provisioning() is a weird name and the docs say that its recording the box as provisioning... which its not... | 15:08 |
| cardoe | So Ironic then checked if the BIOS settings had actually changed and it reported they had NOT changed and so it marked it as clean failed. | 15:08 |
| dtantsur | Yeah, it used to mean a much simpler thing when relies on IPA alone | 15:08 |
| * TheJulia takes the corgi overlord for his walk so he can judge everything and give me endless sideeye | 15:09 | |
| cardoe | But then the box moved back to manageable (not sure why) and I queried the BIOS settings via Ironic CLI and they had been updated. | 15:09 |
| dtantsur | cardoe: it could be extremely valuable for you to sync with janders in your evening/his morning: he has gained a ton of experience with these issues recently | 15:10 |
| dtantsur | (and I'm just out of PTO and cannot remember a thing :D) | 15:10 |
| cardoe | https://paste.opendev.org/show/b4YLAhf4ay9xOS1R0hwJ/ that's the log from me trying that change. | 15:14 |
| cardoe | What's jander's email so I can CC him on that patch | 15:16 |
| dtantsur | pm | 15:22 |
| cardoe | Thank you. | 15:22 |
| opendevreview | Merged openstack/ironic master: fix: missing staticmethod decorator on inspection_rules func https://review.opendev.org/c/openstack/ironic/+/977498 | 15:22 |
| dtantsur | rpittau: hey, have you seen [release][horizon][ironic] cycle-with-intermediary deliverables without any release yet in 2026.1 Gazpacho? | 15:23 |
| rpittau | dtantsur: yeah, I actually thought we had ngs and networking-baremetal, remembered wrong | 15:24 |
| TheJulia | I'm hoping we can get the needful reviews for networking-generic-switch and networking-baremetal this week | 15:30 |
| TheJulia | ironic-ui is good to go I think | 15:31 |
| rpittau | TheJulia: I will submit release requests for them just to have a placeholder | 15:37 |
| rpittau | you probably wnat to check this https://review.opendev.org/c/openstack/bifrost/+/968416/18 before releasing bifrost? | 15:37 |
| opendevreview | Doug Goldstein proposed openstack/ironic master: re-enable some pycodestyle checks not covered by ruff https://review.opendev.org/c/openstack/ironic/+/970918 | 15:38 |
| cardoe | alright folks... if ya want those pycodestyle checks re-enabled... ^ needs to get merged before anything else lands... the delta keeps getting bigger and bigger and bigger... | 15:39 |
| opendevreview | Doug Goldstein proposed openstack/ironic stable/2025.2: Prevent power sync to interfere with FirmwareUpdate https://review.opendev.org/c/openstack/ironic/+/977745 | 15:42 |
| TheJulia | I haven't had time to cycle back to it | 15:43 |
| rpittau | we need to fix ironic-ui ci | 15:47 |
| rpittau | I'm preparing a patch | 15:47 |
| TheJulia | k | 15:48 |
| opendevreview | Verification of a change to openstack/networking-baremetal master failed: Add conductor group sharding support https://review.opendev.org/c/openstack/networking-baremetal/+/948432 | 15:52 |
| opendevreview | Riccardo Pittau proposed openstack/ironic-ui master: Add setuptools as dep for docs test https://review.opendev.org/c/openstack/ironic-ui/+/977829 | 15:52 |
| rpittau | this ^ and then we can merge the other one | 15:52 |
| dtantsur | cardoe: thanks, +2 | 16:00 |
| opendevreview | Julia Kreger proposed openstack/bifrost master: Upload a disk image to OCI https://review.opendev.org/c/openstack/bifrost/+/968416 | 16:02 |
| opendevreview | Ivan Anfimov proposed openstack/ironic master: Change tenant_network_types to project_network_types https://review.opendev.org/c/openstack/ironic/+/977834 | 16:21 |
| opendevreview | Ivan Anfimov proposed openstack/ironic master: Change tenant_network_types to project_network_types https://review.opendev.org/c/openstack/ironic/+/977834 | 16:21 |
| opendevreview | Ivan Anfimov proposed openstack/networking-baremetal master: Change tenant_network_types to project_network_types https://review.opendev.org/c/openstack/networking-baremetal/+/977836 | 16:25 |
| opendevreview | Ivan Anfimov proposed openstack/networking-baremetal master: Change tenant_network_types to project_network_types https://review.opendev.org/c/openstack/networking-baremetal/+/977836 | 16:27 |
| opendevreview | Ivan Anfimov proposed openstack/networking-baremetal master: Change tenant_network_types to project_network_types https://review.opendev.org/c/openstack/networking-baremetal/+/977836 | 16:28 |
| *** gmaan is now known as gmaan_afk | 16:29 | |
| TheJulia | I guess those are all bot driven changes :\ | 16:38 |
| opendevreview | Ivan Anfimov proposed openstack/ironic master: Change tenant_network_types to project_network_types https://review.opendev.org/c/openstack/ironic/+/977834 | 16:40 |
| opendevreview | Ivan Anfimov proposed openstack/networking-baremetal master: Change tenant_network_types to project_network_types https://review.opendev.org/c/openstack/networking-baremetal/+/977836 | 16:40 |
| opendevreview | Merged openstack/ironic-ui master: Add setuptools as dep for docs test https://review.opendev.org/c/openstack/ironic-ui/+/977829 | 16:56 |
| TheJulia | cardoe: review jam! | 17:02 |
| cardoe | I'm coming. | 17:03 |
| opendevreview | Merged openstack/ironic master: ci: Explictly move neutron dhcp config to later https://review.opendev.org/c/openstack/ironic/+/977806 | 17:18 |
| opendevreview | Merged openstack/ironic-ui master: Remove MANIFEST.in https://review.opendev.org/c/openstack/ironic-ui/+/975982 | 17:24 |
| opendevreview | Verification of a change to openstack/ironic-prometheus-exporter master failed: Refactor redfish parser to eliminate code duplication https://review.opendev.org/c/openstack/ironic-prometheus-exporter/+/975891 | 17:32 |
| *** gmaan_afk is now known as gmaan | 17:37 | |
| opendevreview | Verification of a change to openstack/networking-baremetal master failed: Trunk port reconciliation for L2VNI attachments https://review.opendev.org/c/openstack/networking-baremetal/+/974619 | 18:29 |
| cardoe | well I dunno if I got kicked again or what. | 19:07 |
| cardoe | looks like I did. | 19:09 |
| TheJulia | are y'all still talking? | 19:11 |
| JayF | we can free your room up lol | 19:11 |
| TheJulia | oh, its no worries | 19:11 |
| TheJulia | I'm on a zoom for the next hour and its dynamic in gmeet | 19:12 |
| cardoe | As far as reviews go, unrelated to the networking bits I was gonna ask folks about feedback on https://review.opendev.org/c/openstack/ironic/+/976626 The goal there is for us to fetch the information about the storage controllers so that we can put together a RAID configuration for the hardware. jamesdenton_ might look at something like TBN in the future for dynamically crafting RAID configs based on the image being | 19:14 |
| cardoe | deployed. | 19:14 |
| TheJulia | so further extending deploy_steps? | 19:15 |
| cardoe | It's a departure from what IPA does today because IPA is only inspecting and gathering the virtual disks that it sees. But to create the RAID layout with Ironic you need to know information about the storage controller and disks already so this is making it possible to fetch that. | 19:15 |
| TheJulia | okay, I feel like I need a diagram, but I sort of grok it | 19:16 |
| TheJulia | s/diagram/flow chart/ | 19:16 |
| cardoe | TheJulia: Could be. He wants dynamic RAID layouts. So I've just super duper roughly floated out that TBN can make port groups. Maybe a step like that to generate the RAID layout. | 19:17 |
| cardoe | TheJulia: so flow chart would come with the spec. | 19:17 |
| cardoe | Today for you to use the "apply_configuration" step from the "raid" interface, you need to know details about your storage controller and your disk. There's no inspection way to gather those details today. | 19:18 |
| TheJulia | yeah, the flow on deploy today attempts to match based upon flavor traits, but to make it even more dynamic.... | 19:18 |
| TheJulia | queue: "aieeeeee" | 19:18 |
| cardoe | Well the partitions need to be created before deploy starts. | 19:19 |
| TheJulia | before write_image ? | 19:20 |
| TheJulia | I mean, obviously, but deloy_steps executing can include raid being triggered today, its the definition though | 19:21 |
| cardoe | okay yeah fair | 19:21 |
| cardoe | In theory. | 19:21 |
| cardoe | So the RAID configuration is fixed is what I'm saying. | 19:22 |
| TheJulia | yeah, it is, but today on flavor matching, there is really nothing saying "hey, we need to do additional raid config determination" as a step and then making magic happen | 19:22 |
| TheJulia | Queue magic-8 ball resulting in raid configs | 19:22 |
| jamesdenton_ | I just need you to read my mind. | 19:23 |
| TheJulia | Nobody wants inside my mind, that is FOR SURE | 19:23 |
| cardoe | It would just be dynamically injecting "target_raid_config" rather than from a fixed JSON blob. | 19:23 |
| TheJulia | yeah | 19:24 |
| cardoe | Actually we've already got a patch for this. | 19:26 |
| cardoe | And yes it's adding a step. | 19:26 |
| opendevreview | Merged openstack/ironic master: convert trait based networking internal classes to dataclasses https://review.opendev.org/c/openstack/ironic/+/974524 | 19:26 |
| cardoe | If the image had trait X, we set target_raid_config to X RAID config. If the image had trait Y then we set target_raid_config to Y. And we're running apply_configuration as a deploy_step | 19:27 |
| cardoe | The steps just reads the inventory to select the right storage controller and disks | 19:28 |
| cardoe | Anyway, I'm gonna make those folks write a spec and a diagram and a flow. | 19:28 |
| cardoe | But I had Nidhi at least look at the Dell Redfish code to extract the info. | 19:31 |
| cardoe | https://docs.openstack.org/ironic/latest/admin/drivers/idrac.html#examples and https://docs.openstack.org/ironic/latest/admin/raid.html#backing-physical-disks in our own docs say that the way to configure RAID is machine and system dependent. Go use your vendor's BMC to look up what the info needs to be. | 19:31 |
| cardoe | Her patch lets the redfish interface fetch that data and put it into inventory. | 19:31 |
| cardoe | Anytime our docs say "use the vendor tools outside of Ironic", I'm trying to fix that. | 19:33 |
| TheJulia | That is a good thing to fix | 19:52 |
| opendevreview | Merged openstack/ironic master: Redfish VirtualMedia NFS and SMB/CIFS Support https://review.opendev.org/c/openstack/ironic/+/958505 | 20:00 |
| opendevreview | Merged openstack/ironic stable/2025.2: Prevent power sync to interfere with FirmwareUpdate https://review.opendev.org/c/openstack/ironic/+/977745 | 20:05 |
| opendevreview | Merged openstack/ironic bugfix/34.0: Fix Redfish boot device setting for full boot request vendors https://review.opendev.org/c/openstack/ironic/+/976751 | 20:05 |
| opendevreview | Verification of a change to openstack/ironic master failed: re-enable some pycodestyle checks not covered by ruff https://review.opendev.org/c/openstack/ironic/+/970918 | 20:05 |
| cardoe | anndddd we failed. | 20:10 |
| TheJulia | NOOOOOOOOOOO | 20:11 |
| TheJulia | :( | 20:11 |
| TheJulia | ugh, CI, why you hate me | 20:26 |
| opendevreview | Julia Kreger proposed openstack/ironic master: vxlan: Add docs to Ironic to cover the VXLAN use case https://review.opendev.org/c/openstack/ironic/+/977102 | 20:59 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM: Debug devstack plugin https://review.opendev.org/c/openstack/ironic/+/977875 | 21:04 |
| opendevreview | Julia Kreger proposed openstack/networking-baremetal master: DNM: Test https://review.opendev.org/c/openstack/networking-baremetal/+/977876 | 21:06 |
| opendevreview | Doug Goldstein proposed openstack/ironic master: re-enable some pycodestyle checks not covered by ruff https://review.opendev.org/c/openstack/ironic/+/970918 | 21:16 |
| cardoe | TheJulia, JayF, dtantsur, clif: ^ do your thing again if ya wanna see that land before something else breaks it. | 21:16 |
| TheJulia | already did | 21:17 |
| * TheJulia puts away the "APPROVED" stamp | 21:17 | |
| cardoe | It was literally 1 line changed... :/ | 21:17 |
| TheJulia | now I just want to understand why the original networking-baremetal job just hates me so much | 21:18 |
| cardoe | You offended the corgi and therefore it won't pass. | 21:18 |
| TheJulia | NOOOOOO | 21:18 |
| * TheJulia gives the corgi scritches to be hopefully forbidden | 21:19 | |
| cardoe | Sooo... https://review.opendev.org/c/openstack/ironic/+/977569 remove touch_provisioning()? | 21:23 |
| TheJulia | I would *think* so | 21:23 |
| TheJulia | it just feels wrong to me | 21:29 |
| cardoe | I'm also wondering if ya think https://opendev.org/openstack/sushy/src/branch/master/sushy/resources/system/constants.py#L190 is good enough | 21:32 |
| opendevreview | Doug Goldstein proposed openstack/ironic bugfix/34.0: fix: missing staticmethod decorator on inspection_rules func https://review.opendev.org/c/openstack/ironic/+/977880 | 21:33 |
| cardoe | What was the bug that you had commented all this on? | 21:34 |
| opendevreview | Doug Goldstein proposed openstack/ironic stable/2025.2: fix: missing staticmethod decorator on inspection_rules func https://review.opendev.org/c/openstack/ironic/+/977882 | 21:34 |
| TheJulia | https://bugs.launchpad.net/ironic/+bug/2075980 | 21:36 |
| cardoe | ugh I see what dtantsur says about touch_provisioning.... he's not wrong. | 21:39 |
| cardoe | The hope is we won't hit the timeout... but hopes are always bad with hardware. | 21:39 |
| cardoe | So I'm gonna leave it. | 21:39 |
| cardoe | I'll put a comment similar to what's in the firmware update code. | 21:40 |
| TheJulia | yeah, he is not unfortunately | 21:40 |
| TheJulia | I'm just... "alarmed", is alarmed the right word? | 21:40 |
| TheJulia | that you *can* even time that out | 21:40 |
| TheJulia | that just seems HORRIBLE | 21:40 |
| TheJulia | Please ship me the hardware vendor which says this is good, the corgi will bark at them. | 21:41 |
| cardoe | Well the situation he's speaking of is that you're in IPA and then you issue a step to update BIOS settings and that has to shutdown and apply them and then boot back into the IPA image. | 21:48 |
| cardoe | Technically the old way used the stock method of "I want to reboot the box and come back into IPA". In that case we set the state to CLEANWAIT and it would actually call touch_provisioning() for us until it left CLEANWAIT. | 21:50 |
| cardoe | So if I remove it then it's actually a behavior regression. | 21:50 |
| TheJulia | then put that as text in-line then :) | 21:50 |
| cardoe | Yeah I'm writing this wall of text there. | 21:50 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM: Debug devstack plugin https://review.opendev.org/c/openstack/ironic/+/977875 | 22:33 |
| TheJulia | I'm sort of feeling like networking-baremetal's CI has long been grumpy, so now I get to fix it. | 22:37 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM: Debug devstack plugin https://review.opendev.org/c/openstack/ironic/+/977875 | 23:13 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Trivial: CI: fix conditional logic around vxlan/vlan/etc. https://review.opendev.org/c/openstack/ironic/+/977896 | 23:28 |
| opendevreview | Julia Kreger proposed openstack/networking-baremetal master: DNM: Test https://review.opendev.org/c/openstack/networking-baremetal/+/977876 | 23:28 |
| opendevreview | Merged openstack/ironic master: re-enable some pycodestyle checks not covered by ruff https://review.opendev.org/c/openstack/ironic/+/970918 | 23:58 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!