| cardoe | libvirt-python installs without fully working often. It’s gotta have python-dev stuff along with pkg-config. | 01:18 |
|---|---|---|
| JayF | I might try a manual pip install of libvirt_python in the virtualbmc setup | 03:59 |
| rpittau | good morning ironic! o/ | 06:35 |
| rpittau | JayF: by experience libvirt python in venv can be problematic because it heavily relies on systemwide components of libvirt, ideally you would want to try using the packaged one | 06:53 |
| rpittau | TheJulia: hi! wondering if you're still working on https://review.opendev.org/c/openstack/ironic/+/896570 and https://review.opendev.org/c/openstack/ironic/+/898010, they're both in merge conflict since a while | 08:28 |
| opendevreview | Merged openstack/networking-generic-switch master: Add Supermicro switches to allow for supported write config https://review.opendev.org/c/openstack/networking-generic-switch/+/926886 | 09:27 |
| opendevreview | Riccardo Pittau proposed openstack/networking-generic-switch master: Force autospec=True in tests and fix unit tests https://review.opendev.org/c/openstack/networking-generic-switch/+/930745 | 09:41 |
| iurygregory | good morning ironic | 11:00 |
| cid | good afternoon ironic. | 11:02 |
| cid | Out of curiosity, what happens when Ironic's minor version gets to v1.99 | 11:02 |
| cid | I assume it likely naturally becomes 2.0? And will the endpoint still serve from /v1? | 11:04 |
| iurygregory | I was thinking it would go to 1.100 | 11:11 |
| iurygregory | but this is indeed a good question | 11:12 |
| iurygregory | I don't have the right answer =) | 11:12 |
| dtantsur | iurygregory, cid, it has to go to 1.100. I wonder how much will break once that happens :D | 12:30 |
| dtantsur | 2.0 will be the next major version of Ironic API, which, realistically, is not going to happen | 12:30 |
| cid | Not going to happen? Why would that be, if you don't mind, dtantsur | 12:32 |
| dtantsur | cid: we simply don't have capacity to work on something so large | 12:34 |
| dtantsur | if we ever do, we have a large backlog of higher priority efforts that we can finish instead | 12:34 |
| iurygregory | 1.100 it will be amazing hehe :D | 12:35 |
| iurygregory | yeah, wondering what will break when it happens | 12:35 |
| dtantsur | We'll know once we get there :D | 12:36 |
| JayF | Now would be a really good time to look at the utilities and sdks though so that if it breaks we can get ahead of it | 13:19 |
| dtantsur | The main thing that can break is sorting versions as strings | 13:24 |
| opendevreview | cid proposed openstack/ironic master: Add support for node name in port creation https://review.opendev.org/c/openstack/ironic/+/930944 | 13:38 |
| JayF | dtantsur: or any clients using a bad string parsing algorithm to get out the version... Like 1.100 being read as 1.10 | 13:44 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-python-agent master: [WIP] A hardware manager call for a full sync before shutdown https://review.opendev.org/c/openstack/ironic-python-agent/+/931017 | 13:49 |
| * TheJulia blinks surprised we *didn't* already have that | 14:05 | |
| TheJulia | oh where oh where did I put my glasses | 14:06 |
| TheJulia | Also, good morning folks! | 14:06 |
| TheJulia | going back the disucssion, we have had a few people in the past year or two raise similar question as to cid's question because they had some adjacent workflow logic where they were doing like two digit string splitting. We've encouraged them to "please don't do that with shell scripting and just have some sort of proper version modeling" | 14:08 |
| opendevreview | Kamil Gustab proposed openstack/sushy master: Adds attributes for IPs for Ethernet Interfaces. https://review.opendev.org/c/openstack/sushy/+/930814 | 14:27 |
| opendevreview | cid proposed openstack/ironic-specs master: [WIP] Kea DHCP backend https://review.opendev.org/c/openstack/ironic-specs/+/931025 | 14:41 |
| TheJulia | It is october, the holloween baking shows have appeared. | 15:37 |
| JayF | It's time for the Great Pumpkin Spice Challenge on BASIC TV | 15:42 |
| JayF | lol | 15:42 |
| JayF | o// | 15:48 |
| JayF | lol I was scrolled up reading backlog and I just waved to someone from days ago | 15:49 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: [WIP] devstack: split into VENVs https://review.opendev.org/c/openstack/ironic/+/930776 | 15:49 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CI: Add a 4k disk CI job https://review.opendev.org/c/openstack/ironic/+/930655 | 15:49 |
| rpittau | good night! o/ | 15:54 |
| JayF | \o | 15:55 |
| TheJulia | So looks like the postgres job is toast | 15:59 |
| TheJulia | https://7d12977942ed9fbb75e4-56d2480851ef91de1634aa83a3abd804.ssl.cf1.rackcdn.com/930655/3/check/ironic-tempest-pxe_ipmitool-postgres/1c245b1/controller/logs/screen-keystone.txt | 15:59 |
| TheJulia | There is a possibility I might move the 4k disk job to just use LVM as well, that way it is sort of "as complex as possible" in that regards, but hopefully I have a starting point | 16:03 |
| JayF | well, that was a best effort thing for a long time | 16:22 |
| JayF | we should at least post to the ML, then remove support with fire | 16:22 |
| * dtantsur is very sad that the OpenStack in general decided to neglect PostgreSQL | 16:26 | |
| TheJulia | Yeah, I feel like we can’t be the only ones, but reality is… we really are the only ones who run it. | 16:40 |
| JayF | I will say, if it's failing for keystone and not ironic | 16:41 |
| JayF | we could put the job in standalone | 16:41 |
| dtantsur | Doesn't our standalone job use Keystone? | 16:44 |
| dtantsur | I thought it was essentially Nova-less | 16:45 |
| JayF | I guess I assumed at some level we have a standalone-standlone ironic job that's actually standalone | 16:45 |
| TheJulia | It does | 16:45 |
| TheJulia | Use keystone | 16:45 |
| dtantsur | JayF: this is bifrost :) | 16:45 |
| JayF | honestly it seems like a gap in our CI to not have any noauth setup | 16:45 |
| JayF | ah yeah, so can we hook a bifrost job up to postgres if we care about keeping it? | 16:45 |
| JayF | I personally don't but am trying to find a path if others do :) | 16:45 |
| dtantsur | We could but there is no support for Postgres there now | 16:46 |
| dtantsur | so it's more complicated than just flipping some configuration options | 16:46 |
| JayF | is anyone actually using it in prod? | 16:46 |
| * dtantsur no clue | 16:46 | |
| JayF | I guess the question I wanna ask: would RH patch postgres support back in if upstream dropped it | 16:46 |
| JayF | I can imagine you all having customers with strong opinions about db backends :) | 16:47 |
| dtantsur | Not on my side of RH | 16:47 |
| dtantsur | and I don't think OSP ever supported Postgres either | 16:47 |
| JayF | so who wants to make the -x,xxx commit then? :D | 16:47 |
| dtantsur | Not me, I should be packing for my vacation right now already :) | 16:48 |
| JayF | where are you going? | 16:48 |
| JayF | my bro's wedding got moved to Winston-Salem so my trip is still on | 16:48 |
| dtantsur | a small bungalow in the middle of nowhere in Schwarzwald | 16:49 |
| JayF | nice | 16:49 |
| JayF | I've been having :( seeing the news all week, one of the towns that is gone in NC, Spruce Pine, is basically where my family vacationed for a lot of my childhood | 16:50 |
| JayF | in cabins up there that likely no longer exist | 16:50 |
| dtantsur | Sigh. I'm sorry to hear. We had this sort of a nightmare not way away from Düsseldorf 3(?) years ago. | 16:50 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: [WIP] devstack: split into VENVs https://review.opendev.org/c/openstack/ironic/+/930776 | 16:53 |
| JayF | I heard someone describe it as like katrina-like because the valleys naturally form bowls to hold in flood waters, but because it's in the mountains it also added a lot of velocity to the water | 16:53 |
| JayF | it's crazy to see because you'd think in the mountains you're safe from hurricanes | 16:54 |
| dtantsur | oh no, it's a bad scenario | 16:54 |
| dtantsur | especially after a certain dry time, valleys can turn into a fire hose | 16:54 |
| JayF | yeah and that area is extremely rural in ways that are hard to understand unless you've been there -- meaning no power/cell phone/etc is keeping some of the worst information out still | 16:55 |
| JayF | definitionally the worst places have no ways to communicate how bad it is :| | 16:55 |
| dtantsur | what a nightmare :( | 16:55 |
| dtantsur | Anyway, time to pack indeed. See you next Tuesday! | 17:04 |
| JayF | o/ | 17:04 |
| TheJulia | Really not a fan of the new zuul ui :( | 17:15 |
| JayF | the changes to zuul.opendev.org? | 17:16 |
| JayF | like when looking up in progress changes? | 17:16 |
| JayF | +9001 to your assessment if so | 17:16 |
| TheJulia | It is almost like the screen which was at the perfect level for "people in the middle" is now a zoomed out view | 17:16 |
| TheJulia | so now I got to click a bunch of times and the search is not generalized, I have to select the project I want to match and perfectly do the project name | 17:17 |
| JayF | apparently globbing does work | 17:17 |
| JayF | it's just no longer an implied 'ending-in-*' | 17:17 |
| TheJulia | oh, I tried 'ironic' and it showed an empty list | 17:19 |
| TheJulia | so now I have to type 'openstack/ironic' | 17:19 |
| TheJulia | didn't try globbing though | 17:19 |
| JayF | *ironic* would work | 17:19 |
| JayF | change dropdown to project-> *ironic* works | 17:20 |
| JayF | but am I crazy or did you used to not even need to change the dropdown? | 17:20 |
| TheJulia | but then I need to click to expand out each area | 17:20 |
| JayF | I noticed it when putting the 6 digit code in wasn't enough anymore, I had to change the dropdown (and expand each area) | 17:20 |
| TheJulia | JayF: you didn't have to | 17:20 |
| TheJulia | I can totally see the zoomout view value, just... ouch | 17:21 |
| TheJulia | :( | 17:21 |
| TheJulia | (maybe not valuable necessarily for OpenStack as an aggregate, but smaller more tightly integrated things, sure) | 17:21 |
| clarkb | the default is pipeline whcih I've suggested be chagned to something else because you can already click on a pipelien and get a filter for it | 17:21 |
| clarkb | if you do expand things you can also click the magnifying glass to add filters for you | 17:22 |
| clarkb | JayF: correct the old filter didn't require you to select input types but it was more limited in what things it could filter over | 17:23 |
| JayF | FWIW not putting this in #opendev or pointing at you all beacuse I assume this is all zuul upstream | 17:24 |
| JayF | which the new UI appears to be sensible for a less distributed project | 17:24 |
| clarkb | it is, but we also have a lot of input. we have an etherpad going and dtantsur already provided the feedback on globbing behavior for project filters that is in there | 17:24 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CI: Add a 4k disk CI job https://review.opendev.org/c/openstack/ironic/+/930655 | 17:24 |
| clarkb | I suspect that BMW is actually more distributed fwiw | 17:25 |
| clarkb | (and they were driving a lot of the work to change the UI to better meet the needs of varying use cases) | 17:25 |
| TheJulia | maybe a "please just start with a zoomed in view" cookie could make me happy | 17:26 |
| clarkb | all of the filters should be encoded in your url | 17:27 |
| clarkb | I think you should be able to bookmark the zoomed in view that you like to see and have that work | 17:27 |
| clarkb | though maybe not the expansion of jobs? | 17:28 |
| * clarkb makes a note on saving that state in the feedback doc | 17:28 | |
| TheJulia | I guess it is just rough, 7 steps to find the ironic job status is semi-owie when I could jsut reach the page and control-f it | 17:29 |
| TheJulia | oh where did I put my brain | 17:30 |
| JayF | https://review.opendev.org/c/openstack/virtualpdu/+/928823 oh wow | 17:31 |
| JayF | CI on this is making it clear | 17:31 |
| JayF | virtualpdu devstack jobs | 17:32 |
| JayF | do not install virtualpdu from git | 17:32 |
| JayF | wheeee | 17:32 |
| TheJulia | doh | 17:32 |
| JayF | wait, no, what | 17:32 |
| JayF | I might have been looking at the wrong job, please hold | 17:32 |
| JayF | even still, I think at best it'd mean we're running ironic CI against released virtualpdu | 17:33 |
| JayF | which is non-ideal as well | 17:33 |
| JayF | so close to getting all the pieces in place for the vpdu + venv stuff | 17:38 |
| opendevreview | Jay Faulkner proposed openstack/virtualpdu master: Vendor pysnmp v5.0.33 into virtualpdu https://review.opendev.org/c/openstack/virtualpdu/+/928823 | 17:38 |
| clarkb | fwiw providing feedback on zuul things in the zuul matrix room or on the zuul-discuss mailing list is always fair game. If there is interest in providing thoughts on the new dashboard more directly so we don't have to play telephone that may help people writing the js understand needs better | 17:44 |
| clarkb | other ideas: if you click on a pipeline to get the pipeline view you get a toggle for expanding all jobs. You could potentially do that if it is a specific pipeline you care about. That toggle is persisted in local storage | 17:59 |
| clarkb | the idea that is being bounced around is having that toggle on the top level dashboard too (to basically restore the old dashboard behavior of showing you all the things at once) | 17:59 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: [WIP] devstack: split into VENVs https://review.opendev.org/c/openstack/ironic/+/930776 | 18:00 |
| JayF | clarkb: if that setting was cookie-saved, that would work. If it's a button to click it's probably not worth the effort | 18:01 |
| clarkb | JayF: its local storage saved | 18:01 |
| JayF | I think that'd be OK but not awesome | 18:01 |
| JayF | but I use more different computers/browsers than average :D | 18:01 |
| clarkb | not sure I understand that last statement. It should functionally be the same as a cookie | 18:01 |
| clarkb | except that info isn't sent back to the server so we don't have privacy concerns | 18:01 |
| clarkb | yes you have to set it per browser but that is no different than with a cookie | 18:02 |
| JayF | what I'm saying is I use like, 3 machines and two browsers on each | 18:02 |
| JayF | which means 6 button presses everytime my data is cleared | 18:02 |
| JayF | which is probably 5 more than most people :D | 18:03 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CI: Add a 4k disk CI job https://review.opendev.org/c/openstack/ironic/+/930655 | 18:26 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: Remove postgresql testing and documentation https://review.opendev.org/c/openstack/ironic/+/931055 | 18:34 |
| JayF | -128 +50 ... not nearly as satisfying as I might have hoped :D | 18:37 |
| TheJulia | JayF: was my call to action, a call to action? :) | 18:44 |
| JayF | I already said in here I was gonna do it | 18:44 |
| JayF | it was mentioned in the TC meeting | 18:44 |
| TheJulia | :) | 18:44 |
| TheJulia | ahh, okay | 18:44 |
| JayF | I basically said we had no desire at all to maintain it | 18:44 |
| TheJulia | same | 18:44 |
| JayF | manila apparently has pgsql jobs too | 18:44 |
| JayF | due to some known large deployments using it | 18:44 |
| JayF | my reponse was to invite those people to the ironic community to help maintain it if desired :D | 18:45 |
| TheJulia | that works | 18:45 |
| JayF | also like, this is early cycle | 18:46 |
| JayF | this is /exactly/ when to merge removal like this | 18:46 |
| JayF | the earliest days of the slurp release | 18:46 |
| * JayF notes re: his post to the list that we appear to use ironic-* groups throughout LP, so we should be good w/r/t ironic-coresec | 19:39 | |
| JayF | based onthe actions taken by fungi and I when enabling VMT for Ironic recently | 19:39 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CI: Add a 4k disk CI job https://review.opendev.org/c/openstack/ironic/+/930655 | 20:30 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: CI: Change 4k job to use LVM https://review.opendev.org/c/openstack/ironic/+/931061 | 20:30 |
| iurygregory | TheJulia, you will be at the OpenInfra Days NA, right? | 21:03 |
| TheJulia | Yes | 21:03 |
| iurygregory | can you give the pxe boot sticker to guilhermesp? since Amy forgot to bring to Suwon D: | 21:04 |
| TheJulia | I forgot to give some to Amy | 21:07 |
| iurygregory | no worries :D | 21:08 |
| JayF | I'll note I am always happy to post stickers to anyone | 21:11 |
| JayF | and have budget to make more exist if I run out :) | 21:11 |
| iurygregory | JayF, shipping costs to Brazil are a bit high I would say :D | 21:12 |
| JayF | that would /not/ even be close to the most remote place I've mailed stickers | 21:12 |
| JayF | sometimes it takes 2 or 3 tries to get them into some international places | 21:13 |
| * JayF needs to try to ship cid some stickers again | 21:13 | |
| iurygregory | yup it can happen! | 21:13 |
| clarkb | usps says $18. A lot for a sticker but not for the sentiment :) | 21:36 |
| clarkb | if it fits in a letter it would be cheaper | 21:36 |
| TheJulia | I offered to mail some of the car ones out :) | 21:37 |
| TheJulia | They are also in my desert lair | 21:38 |
| iurygregory | clarkb, yep and here in Brazil they would make me pay taxes based on the shipment and price of the product lol | 21:39 |
| clarkb | iurygregory: fun | 21:40 |
| iurygregory | so it's cheaper to give to someone at the conference so I can pick up here in Brazil XD | 21:40 |
| iurygregory | please bring the car ones to the OID NA :D | 21:41 |
| JayF | clarkb: I have international forever stamps, when I bought them they were $1 | 21:50 |
| JayF | clarkb: anything <1 oz in an envelope, it's just one of those | 21:50 |
| JayF | which is why I don't mind sending it a few times to get it thru, even if it takes 3 times that's like, $15 in stickers, postage, envelopes :) | 21:51 |
| opendevreview | Julia Kreger proposed openstack/ironic-tempest-plugin master: Fallback to SSH if ping fails https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/931069 | 23:02 |
| opendevreview | Julia Kreger proposed openstack/ironic-tempest-plugin master: Fallback to SSH if ping fails https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/931069 | 23:04 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CI: Add a 4k disk CI job https://review.opendev.org/c/openstack/ironic/+/930655 | 23:09 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: CI: Change 4k job to use LVM https://review.opendev.org/c/openstack/ironic/+/931061 | 23:09 |
| TheJulia | clarkb: did the logic around running stacked jobs change at all? | 23:09 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: CI: Change 4k job to use LVM https://review.opendev.org/c/openstack/ironic/+/931061 | 23:47 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!