| *** dmellado471725 is now known as dmellado47172 | 00:17 | |
| opendevreview | Merged openstack/networking-baremetal master: Guard rabbit config access in _get_notification_transport_url https://review.opendev.org/c/openstack/networking-baremetal/+/985777 | 00:59 |
|---|---|---|
| opendevreview | Merged openstack/python-ironicclient master: fix: make it possible to unset local-link-connection on baremetal port https://review.opendev.org/c/openstack/python-ironicclient/+/985047 | 01:02 |
| opendevreview | Merged openstack/bifrost master: Add Debian 13 (Trixie) support https://review.opendev.org/c/openstack/bifrost/+/979651 | 01:07 |
| opendevreview | Verification of a change to openstack/networking-generic-switch master failed: Fix SONiC and Arista VXLAN parameter config errors https://review.opendev.org/c/openstack/networking-generic-switch/+/985122 | 01:42 |
| opendevreview | Merged openstack/ironic stable/2026.1: ci: stable: disable metal3 CI jobs https://review.opendev.org/c/openstack/ironic/+/986812 | 02:31 |
| opendevreview | Merged openstack/networking-generic-switch master: SONiC: Create VLAN before adding members/VNI map https://review.opendev.org/c/openstack/networking-generic-switch/+/984648 | 03:24 |
| opendevreview | Steve Baker proposed openstack/ironic master: Graceful browser shutdown for graphical console https://review.opendev.org/c/openstack/ironic/+/987441 | 05:15 |
| opendevreview | Steve Baker proposed openstack/ironic master: DNM re-enable graphical console testing https://review.opendev.org/c/openstack/ironic/+/987442 | 05:15 |
| opendevreview | Kaifeng Wang proposed openstack/ironic master: Document provisioning a riscv machine https://review.opendev.org/c/openstack/ironic/+/987460 | 08:56 |
| opendevreview | Kaifeng Wang proposed openstack/ironic master: Document provisioning a riscv machine https://review.opendev.org/c/openstack/ironic/+/987460 | 08:58 |
| dtantsur | TheJulia, JayF, do we need a release note for operators to change kernel/ramdisk images (back) to aki/ari? | 10:25 |
| opendevreview | Abhishek Bongale proposed openstack/ironic-tempest-plugin master: Add INFO logging to baremetal scenario tests https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/987343 | 10:56 |
| opendevreview | Tero Kauppinen proposed openstack/sushy-tools master: Add --storage-pool CLI option https://review.opendev.org/c/openstack/sushy-tools/+/987486 | 11:02 |
| opendevreview | Verification of a change to openstack/networking-generic-switch master failed: Fix SONiC and Arista VXLAN parameter config errors https://review.opendev.org/c/openstack/networking-generic-switch/+/985122 | 12:04 |
| opendevreview | Takashi Kajinami proposed openstack/ironic-tempest-plugin master: Reuse base tempest code for common config sections https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/987494 | 12:17 |
| TheJulia | I posted a change for that yesterday | 12:38 |
| iurygregory | yay I'm back \o/ | 12:42 |
| dtantsur | ah, found it | 12:48 |
| dtantsur | and welcome back iurygregory :) | 12:48 |
| iurygregory | tks, at least the power is back, but still raining a lot here lol | 12:49 |
| opendevreview | Tero Kauppinen proposed openstack/sushy-tools master: Add --storage-pool CLI option https://review.opendev.org/c/openstack/sushy-tools/+/987486 | 12:54 |
| opendevreview | Ivan Anfimov proposed openstack/ironic-ui master: Update Babel configuration https://review.opendev.org/c/openstack/ironic-ui/+/987506 | 13:52 |
| opendevreview | Ivan Anfimov proposed openstack/ironic-ui master: Update Babel configuration https://review.opendev.org/c/openstack/ironic-ui/+/987506 | 13:54 |
| opendevreview | Julia Kreger proposed openstack/ironic master: security: Use sandbox rendering for jinja2 https://review.opendev.org/c/openstack/ironic/+/987514 | 14:01 |
| * Sandzwerg[m] uploaded an image: (130KiB) < https://matrix.org/oftc/media/v1/media/download/AVCa3PHlLn-jw1MLEoB0cyHkogShJrXSobBOuA74a6eQ2e8eyYxxPNyL6JsQoPi3xUzXGtZdyLTDS7Mh4kvNo0JCeeRbGs_wAG1hdHJpeC53YXJwem9uZS5tcy93WUZmQ3puQ1VOY3ZzdVpqdWxOdndqeGQ > | 14:02 | |
| TheJulia | Sandzwerg[m]: uhhhh ;) | 14:03 |
| Sandzwerg[m] | Yeah that was a wrong chat. Sorry. But can't remove it from irc 😅 | 14:03 |
| TheJulia | it is what it is :( | 14:04 |
| Sandzwerg[m] | 🤷🏻 don't sent me money via wero? | 14:07 |
| Sandzwerg[m] | But happens to the best of us. | 14:07 |
| TheJulia | lol, yeah | 14:07 |
| dtantsur | Sandzwerg[m]: at least you have wero *side eyes DKB* | 14:08 |
| TheJulia | dtantsur: corgi's provide the best side eye | 14:09 |
| TheJulia | SideEye as a Service | 14:09 |
| TheJulia | ... although my corgi looks tired and is not allocating side-eye right now | 14:09 |
| dtantsur | Oh, they're awesome at that. If corgis could get an account at my bank, it would have wero already. | 14:09 |
| Sandzwerg[m] | Yeah, the one advantage my otherwise questionable bank has. But this was actually to test if it leaks your full name, and yeah it seems it does if you sent the money you see the full name while the recipient might only see the first name. | 14:10 |
| Sandzwerg[m] | I'd argue with more Corgis in position of power the world would be better | 14:10 |
| dtantsur | Sigh, great | 14:11 |
| dtantsur | but you found something good about Postbank, congrats! :D (my wife has it and hates passionately) | 14:11 |
| Sandzwerg[m] | I know someone who has a association account there and by now they are probably on first name basis with the support. For me it works okayish but yeah something else would be preferred. Puts it on the big pile of to-do | 14:15 |
| TheJulia | wait, if corgis are in positions of power.... oh my. | 14:17 |
| rpittau | cid, JayF: I've been made aware that you were planning to cut a bugfix branch for ironic soon? what's the reason behind that? the next bugfix branch should not come before end of may/beginning of June | 14:19 |
| TheJulia | My wife points out that corgi's are hearding dogs and by their very nature would be authoritarian as "rules enforcement" are in their very DNA... albeit cute authoritarians who demand endless food. | 14:23 |
| dtantsur | :D | 14:30 |
| cid | rpittau, a bugfix branch just landed. I don't think there are plans for any others (yet) | 14:31 |
| dtantsur | not created yet though? | 14:32 |
| rpittau | cid: alnded where ? | 14:32 |
| dtantsur | I see tag 36.0 but not the branch (yet?) | 14:33 |
| rpittau | cid: the plans for bugfix branches are in the whiteboard, I don't know of any others | 14:33 |
| dtantsur | I'm also confused, didn't we discuss that it's not until next week at least? | 14:33 |
| cid | Maybe I am mixing things up. by branching you do not mean a release do you? | 14:33 |
| dtantsur | cid: any relaese has a branch, except for patch releases from branches | 14:33 |
| cid | The releases on the stable branches are patch releases, I believe. And the 36.0.0 Hibiscus was just tagged. | 14:35 |
| dtantsur | cid: yeah, which is a bit early and bugfix/36.0 is still missing | 14:36 |
| cid | Not branched, do those go hand in hand? | 14:36 |
| rpittau | cid: I know that, I approved it :) | 14:37 |
| rpittau | I guess I should've asked/thought about it before approving it | 14:37 |
| rpittau | in any case, major releases during cycle usually go with a bugfix branch, and it's still early for that | 14:37 |
| dtantsur | cid: yes, per our release policy. also, rpittau, you literally approved it... | 14:37 |
| rpittau | yep | 14:37 |
| rpittau | I counted 3-4 weeks before bugfix, that's why | 14:38 |
| JayF | Do we have a policy to never cut an ironic release unless it has a bug fix branch? Honestly, it makes sense as a policy but it would be my first time hearing that said out loud | 14:39 |
| rpittau | so next bugfix will be a minor release | 14:39 |
| rpittau | unless we land soem breaking change in the mneantime | 14:39 |
| cid | dtantsur, hmm. okay. so, bugfix/36.0 should exist but it can't exist as soon as the release for the new cycle is cut but it can't because the release is too early. | 14:39 |
| dtantsur | JayF: in all fairness, we don't | 14:39 |
| rpittau | JayF: never said that :) | 14:39 |
| JayF | I'm just trying to say that there is a perceptional difference between a bug fix Branch release and just a release. | 14:39 |
| dtantsur | so maybe we skip bugfix/36.0 and soon afterwards create, say, 36.1 and bugfix/36.1? | 14:39 |
| cid | *should exist as soon as the release for the new cycle is cut but it can't because the release is too early. | 14:40 |
| rpittau | dtantsur: exactly | 14:40 |
| dtantsur | where "soon afterwards" is whenever we were supposed to create a bugfix branch | 14:40 |
| rpittau | JayF, cid, I was just asking if there was a plan for a new bugfix branch soon, that's it | 14:40 |
| rpittau | since it's early | 14:40 |
| JayF | I think the only reason cid maybe looking at this is that I asked him to make sure that we got a release out with all of the security fixes... I may not have properly scoped that request to stable | 14:40 |
| dtantsur | yeah, I guess we leave 36.0 alone and just do the full release thing when it's due? with a branch and so on? | 14:41 |
| rpittau | yep | 14:41 |
| cid | ++ | 14:43 |
| cid | . I was also a little curious. The patch was originally a feature release, but the review stated that it had to be a major release. | 14:43 |
| cid | Question | 14:43 |
| cid | rpittau, why would it be a breaking change if we branched early Just for curiosity sake | 14:44 |
| rpittau | cid: I meant that fi we want another major release we need a breaking change in ironic to land :) | 14:44 |
| JayF | Well, I have good news | 14:45 |
| rpittau | lol | 14:45 |
| JayF | https://review.opendev.org/c/openstack/ironic/+/987439 does this break enough? | 14:46 |
| JayF | or do you need me to break harder | 14:46 |
| rpittau | that's probably enough :D | 14:46 |
| rpittau | cid, JayF, thanks and sorry for the confusion :) | 14:51 |
| dtantsur | Note: we don't *have* to have the next release as major :) 35.1 is fine too | 15:13 |
| TheJulia | did molds get ripped? | 15:15 |
| JayF | The only real thing we have to ensure is that no bugfix or cycle release is just a x.y.+1 | 15:15 |
| TheJulia | yet?! | 15:15 |
| JayF | because then we lose a spot to release between | 15:15 |
| TheJulia | I think it rechecked yesterday? But metal3-integration was unahppy I think | 15:15 |
| * TheJulia goes back to her git hammer | 15:15 | |
| JayF | https://review.opendev.org/c/openstack/ironic/+/986703 landed yesterday afternoon | 15:15 |
| TheJulia | shoot, I missed it | 15:15 |
| JayF | hat-tip to cid for the recheck | 15:16 |
| TheJulia | thank you cid! | 15:16 |
| cid | \o/. that's partly why I even delayed updating the release :D | 15:17 |
| cid | *release patch that just landed. | 15:18 |
| cardoe | TheJulia: what do ya want me to do with https://review.opendev.org/c/openstack/ironic/+/987366 ? | 15:43 |
| cardoe | want me to get a patch against global-requirements? | 15:43 |
| TheJulia | I just wanted others to see it | 15:44 |
| TheJulia | we're not really supposed to change the requirements.txt once branched, but this is sort of one of those weird cases where it also kind of makes sense, and we're just doing a minor version exclusion there | 15:44 |
| TheJulia | the true true true right way is via g-r | 15:44 |
| cardoe | I also don't get why https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/942220 keeps failing | 15:46 |
| TheJulia | Offhand, dunno | 16:29 |
| cardoe | cid: Didn't you have a series to fix "Service RPCService is not picklable with spawn; falling back to fork. Please make the service spawn-safe to avoid this fallback."? | 16:37 |
| cardoe | I'm seeing that in stable/2026.1 | 16:37 |
| cid | That change has now landed in both oslo.service and in Ironic but I am not sure it was part of Gazpacho | 16:41 |
| cid | https://review.opendev.org/c/openstack/ironic/+/977266 | 16:45 |
| cid | cardoe, I don't think it's backportable since it's a feature-ish? | 16:46 |
| cid | Maybe it is | 16:46 |
| cardoe | dang it's the very first commit after 2026.1 | 16:49 |
| cardoe | or I should say 35.0.0 | 16:50 |
| cardoe | but okay.. I won't file a bug | 16:50 |
| JayF | I'm not sure I'd call that a feature. It doesn't really provide any new functionality | 16:50 |
| JayF | Does it have any negative side effects? | 16:50 |
| JayF | I can't see any obvious reason we *wouldn't* want to backport this change, but the *obvious* there is doing some work :D | 16:50 |
| opendevreview | cid proposed openstack/ironic stable/2026.1: Fix pickle error when oslo.service uses spawn mode https://review.opendev.org/c/openstack/ironic/+/987550 | 16:53 |
| TheJulia | I'm not entirely sure of the impact/footprint, but one less error is always for the best | 16:56 |
| opendevreview | Merged openstack/networking-generic-switch master: Fix SONiC and Arista VXLAN parameter config errors https://review.opendev.org/c/openstack/networking-generic-switch/+/985122 | 17:11 |
| TheJulia | is launchpad working for folks? | 17:17 |
| TheJulia | oh, there it goes | 17:17 |
| TheJulia | err, maybe not | 17:17 |
| JayF | it's been pretty off-and-on for days, I always wait and retry as needed | 17:17 |
| JayF | took me three tries to get a comment in yesterday | 17:18 |
| opendevreview | Julia Kreger proposed openstack/ironic master: security: Use sandbox rendering for jinja2 https://review.opendev.org/c/openstack/ironic/+/987514 | 17:21 |
| TheJulia | That should fix the reno issue on that. | 17:22 |
| JayF | +23 | 17:22 |
| JayF | +2** | 17:22 |
| TheJulia | Okay, where was my brain | 17:22 |
| TheJulia | !jay, !dmitry, but other ironic-cores: it would be super awesome if you could take a look at https://review.opendev.org/c/openstack/ironic-specs/+/954612 so we we can begin to build some velocity there since so much rooted back into "we really need this functionality" at the ptg. Current idea is /v1/jobs, love it or hate it, please comment! | 17:23 |
| opendevmeet | TheJulia: Error: "jay," is not a valid command. | 17:23 |
| TheJulia | heh | 17:24 |
| * TheJulia laughs evilly | 17:24 | |
| opendevreview | Nicholas Kuechler proposed openstack/ironic master: Fix redfish sensor data crash when redfish_system_id is None https://review.opendev.org/c/openstack/ironic/+/987557 | 17:37 |
| TheJulia | cardoe: by chance, have you looked at https://github.com/scrapli ? | 18:05 |
| cardoe | I haven't | 18:06 |
| TheJulia | I asked claude to come up with some crazy ideas to be able to excise netmiko/paramiko, and it came up with that | 18:06 |
| TheJulia | cardoe: https://paste.opendev.org/show/bpGJi4XauaLIVoghIbh7/ | 18:15 |
| * cardoe pushes some yaks to try and make em move out of the way. | 18:17 | |
| TheJulia | Electric yak shavers? | 18:17 |
| JayF | scrapli, afaict, works somewhat via screen scraping | 18:18 |
| JayF | which is a potential red flag for cross-distro compatability | 18:18 |
| TheJulia | but if its a launched pty, that might be okay I guess | 18:31 |
| TheJulia | ... might not be the happiest dpending on io | 18:31 |
| JayF | "networking-generic-switch not working on nixos with openssh [newer than rhel], please help?" | 18:32 |
| JayF | Is more what I'm thinking/worried about | 18:32 |
| JayF | and I'm not even going to mention dropbear :D | 18:33 |
| cardoe | what's a CLI that'll target paste.opendev.org? | 18:33 |
| TheJulia | gah https://zuul.opendev.org/t/openstack/build/052b9da59bb94f1383a5f6b0f3c59c5f | 18:42 |
| cardoe | #openstack-infra has said the flavor we are targetting is flaky | 18:42 |
| TheJulia | yeah | 18:43 |
| TheJulia | JayF: dropbears requires us to all travel to Sydney for meetings. | 18:43 |
| cardoe | TheJulia: https://gist.github.com/cardoe/2972dc791f3aaea02796f4dea75fe8bd since we're linking things at each other. | 18:46 |
| TheJulia | cardoe: to post, dunno | 18:46 |
| * TheJulia blinks | 18:47 | |
| cardoe | https://www.shutterstock.com/image-photo/vigilant-black-cat-piercing-yellow-600nw-2724357767.jpg | 18:53 |
| TheJulia | clif: comment added to https://review.opendev.org/c/openstack/ironic-python-agent/+/987391 | 18:54 |
| TheJulia | I could be okay with the approach, but other cores chiming in would be good | 18:56 |
| JayF | I thought in the sync chat you, me, and dtantsur that he had a use case for the UEFI partition images that he requested we not break? | 18:58 |
| JayF | That's why I thought we discussed reorienting the fix... | 18:58 |
| TheJulia | Well, there are users, and without that, it may be breaking to some folks | 18:59 |
| TheJulia | I guess I'm also prompting out of some sense of inner urgency to get things sorted but its never that simple :) | 19:01 |
| JayF | I read a blogpost yesterday which encouraged me to take my time to find the right solution even if there are pressures to go fast :D | 19:12 |
| TheJulia | JayF: ++ | 19:14 |
| TheJulia | cardoe: so, Interesting idea regarding traits and all. I guess the thing I'm mentally struggling with, and maybe I'm not makign the connection is your doing a VNI range with a physical network, and then when a user requests a network, that implies based upon the segment possible, right? | 19:15 |
| TheJulia | so you end up with, for example, you have FAB_A and FAB_B, 1-100,000, and 100,001-200,000 respective and they are diverse network fabrics | 19:16 |
| TheJulia | the only thing your missing in the middle is a possible need to bridge between the two, but if its on the original segment, then so be it | 19:16 |
| TheJulia | you just potentiallly end up with hypervisors which may not be attached to one of the Fabrics | 19:16 |
| cardoe | TheJulia: So remember Helen's type 5 routing spec? | 19:16 |
| TheJulia | or they miht be | 19:16 |
| TheJulia | yeah | 19:16 |
| TheJulia | at least, last time I looked at it | 19:17 |
| cardoe | Oh absolutely I expect to have boxes hooked to one fabric vs the other. | 19:17 |
| cardoe | There was someone on the ML that recently responded with that as well. | 19:17 |
| cardoe | Saying they have Ironic nodes on two separate fabrics and they'd like to use the rest of OpenStack tools and have it do the right thing. | 19:17 |
| TheJulia | in a physcial world, totally reasonable. Not so much in the neutron world but bringing together context is double++ good. | 19:18 |
| cardoe | Right now Neutron assumes a network is an overlay and it exists everywhere. | 19:18 |
| TheJulia | yup | 19:18 |
| cardoe | So if you recall I said I made the custom vrf router plugin? | 19:18 |
| TheJulia | yup | 19:18 |
| cardoe | So now I've carved out a pool of VNIs that can be my magical L3VNI pool. | 19:19 |
| TheJulia | super reasonable | 19:19 |
| cardoe | Helen's spec needs the same and they're just making a config file entry with a range. | 19:19 |
| TheJulia | yeah, I *REALLY* am not a fan about config file management | 19:19 |
| cardoe | Me neither. | 19:20 |
| cardoe | So my approach is using network_segment_range | 19:20 |
| TheJulia | SGTM | 19:20 |
| cardoe | openstack network segment range create --type vxlan --minimum 2000 --maximum 3000 --physical-network my-l3vni-pool | 19:20 |
| cardoe | and it doesn't affect openstack network create because that'll still use --physical-network None in all the cases. | 19:21 |
| cardoe | So this is where there is a little config file mumbo jumbo. | 19:21 |
| cardoe | I'm using the networking-bgpvpn extension as well so an admin can create a bgpvpn and define the AS | 19:22 |
| TheJulia | cool cool | 19:23 |
| cardoe | My backend plugin adds a "fabric" field as well and the admin creates it with that fabric value | 19:23 |
| cardoe | I've added validation in the "routes-to" field as well to only allow other fabric values. | 19:24 |
| cardoe | So the admin makes these bgpvpn objects per-fabric and makes them public. | 19:25 |
| cardoe | then a tenant/user can do openstack router create --flavor vrf blah | 19:25 |
| cardoe | openstack bgpvpn router association create fabricA-bgpvpn blah | 19:27 |
| TheJulia | interesting! | 19:27 |
| cardoe | The config file mumbo jumbo is that the vrf router flavor reads "pool_prefix" which in my case is "l3vni-" | 19:28 |
| cardoe | And above where I called the pool "my-l3vni-pool" I actually called it "l3vni-fabricA" for example. | 19:29 |
| cardoe | And so it automatically looks it up for you. | 19:29 |
| cardoe | I have a TODO to use policy or permission checks to allow an admin to select the VNI directly. | 19:31 |
| cardoe | Nothing is working yet. It's just all in a dummy devstack | 19:31 |
| cardoe | and I haven't figured out the nova scheduling pieces | 19:32 |
| cardoe | But knowing what configs are necessary on the switches to make all this work. And knowing the hooks I have available to me. And knowing the data I have in the database. I _SHOULD_ be able to push the right config to the switches. | 19:33 |
| cardoe | I've done ONE by hand with a networking person typing commands and me looking at debug prints from my plugins. | 19:35 |
| cardoe | The last thorn I have for the requirement is that we have to do BGP auth / encryption which requires each network VNI have its own key. So I was looking at getting barbican involved. | 19:40 |
| cardoe | But I'll likely cheat for now and just stuff something as another extension field. | 19:40 |
| estherdomfeh[m] | Hi everyone. I'm estherd, an Outreachy intern for the Ironic node history API project. Mentors are hroy and rpittau. Looking forward to working with everyone! | 20:04 |
| *** estherdomfeh[m] is now known as estherd[m] | 20:04 | |
| TheJulia | cardoe: cool cool, sorry I'm heads down right now, but sounds neat | 20:39 |
| TheJulia | estherd[m]: welcome! | 20:40 |
| opendevreview | Harald Jensås proposed openstack/networking-generic-switch master: GenericSwitchNetmikoConfigError error details https://review.opendev.org/c/openstack/networking-generic-switch/+/985029 | 21:31 |
| *** jroll01 is now known as jroll0 | 21:32 | |
| cardoe | hjensas: ah I didn't mean we had to get rid of it. We can include the details in the exception. I was more saying we should be consistent. Cause in some places did we generate a message and then log it and include it in the exception. In other places we generate a detailed message and just log it and then raise the exception with just a vague comment. | 22:12 |
| opendevreview | Armin M proposed openstack/ironic master: fix: oci image service handling webserver_verify_ca when it's string https://review.opendev.org/c/openstack/ironic/+/986668 | 23:06 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
