| vanou | good morning ironic | 02:10 |
|---|---|---|
| arne_wiebalck | Good morning Ironic! | 07:31 |
| *** akahat|ruck is now known as akahat|rover|lunch | 07:35 | |
| *** akahat|rover|lunch is now known as akahat|rover | 08:28 | |
| rpittau | good morning ironic! o/ | 09:15 |
| rpittau | FYI switch to jammy is in 2 days, we should be ok, but please keep an eye on https://review.opendev.org/c/openstack/ironic/+/863873 more info at https://etherpad.opendev.org/p/migrate-to-jammy | 09:37 |
| dtantsur | rpittau: I'm fine with keeping an eye on that patch, but I suspect you meant another one ;) | 09:52 |
| rpittau | wellll.... yeah..... :/ | 09:55 |
| rpittau | this is the one! https://review.opendev.org/c/openstack/ironic/+/861325 | 09:56 |
| opendevreview | Arne Wiebalck proposed openstack/ironic master: [doc] Add documentation on SMART test after disk burn-in https://review.opendev.org/c/openstack/ironic/+/864550 | 12:36 |
| arne_wiebalck | rpittau: ^^ (thanks for spotting it!) | 12:37 |
| arne_wiebalck | rpittau: zuul is failing in a network test, this should be unrelated to a string change in the docs :) | 13:40 |
| rpittau | I really hope it is! :) | 13:41 |
| *** akahat|rover is now known as akahat|ruck|afk | 14:03 | |
| TheJulia | Good morning everyone | 14:09 |
| opendevreview | Verification of a change to openstack/ironic master failed: [doc] Add documentation on SMART test after disk burn-in https://review.opendev.org/c/openstack/ironic/+/864550 | 14:11 |
| kubajj | Hello everyone! | 14:14 |
| TheJulia | good morning | 14:15 |
| * TheJulia looks at meeting invite which is now before 6 AM | 14:18 | |
| dtantsur | that's evil | 14:25 |
| dtantsur | also good morning TheJulia, kubajj | 14:25 |
| kubajj | Hi dtantsur | 14:25 |
| TheJulia | dtantsur: Yeah, I've requested it to be moved back just 15 minutes in the past and my request was ignored... soooo *shrug* | 14:26 |
| dtantsur | :( | 14:34 |
| opendevreview | Merged openstack/ironic master: [doc] Add documentation on SMART test after disk burn-in https://review.opendev.org/c/openstack/ironic/+/864550 | 14:45 |
| *** akahat|ruck|afk is now known as akahat|ruck | 15:00 | |
| opendevreview | Harald Jensås proposed openstack/ironic master: Add ports statistics to tools/benchmark scripts https://review.opendev.org/c/openstack/ironic/+/864762 | 16:33 |
| opendevreview | Julia Kreger proposed openstack/ironic-specs master: Add a shard key https://review.opendev.org/c/openstack/ironic-specs/+/861803 | 16:34 |
| hjensas | TheJulia: https://paste.opendev.org/show/boW6bFAIxuh4bMcYsvse <- the via ports API call pattern improvements looks good so good I think I must have made a mistake? | 16:35 |
| TheJulia | wow! | 16:36 |
| JayF | hjensas: or you fixed like, 8 years of mistakes | 16:38 |
| TheJulia | 3533% improvement | 16:40 |
| hjensas | JayF: that would be good. *blink* I can't belive it. | 16:40 |
| TheJulia | that is absurd | 16:40 |
| JayF | hjensas: you're working on a python project that uses a heavy ORM layer. I'm impressed but not in disbelief. | 16:41 |
| JayF | hjensas: python - the real cause of climate change /s | 16:41 |
| TheJulia | heh | 16:41 |
| TheJulia | Yeah, the loop/iteration causes a lot of small db queries and that adds up to be a ton over time | 16:42 |
| TheJulia | so definitely not in land of disbelief, but quite impressed :) | 16:42 |
| TheJulia | hjensas: this is blog worthy, fwiw | 16:43 |
| JayF | hjensas: this is summit-worthy | 16:45 |
| JayF | hjensas: TheJulia: finding performance improvements in frequently hit API endpoints | 16:45 |
| opendevreview | Julia Kreger proposed openstack/ironic-specs master: Add Ironic work items for 2023.1 https://review.opendev.org/c/openstack/ironic-specs/+/862538 | 16:45 |
| JayF | there's your title, go write the paper | 16:46 |
| TheJulia | Actually yes, it *is* summit worthy | 16:46 |
| TheJulia | i broke some minds in berlin talking about the ports because people just didn't seem to grok just how much a performance impact spreading out that loop creates | 16:46 |
| JayF | seriously; about node and ports | 16:46 |
| JayF | I bet there are other projects with similar benefits to make | 16:46 |
| TheJulia | I did w/r/t nodes, but there is even another piece in nodes which can be improved, chassis_uuid.... | 16:47 |
| TheJulia | and that would be awesome to revisit | 16:47 |
| JayF | heh | 16:47 |
| TheJulia | btw, JayF, I rebased the work items, clean minus the index, re-applied my +2 | 16:48 |
| JayF | I figured; ty | 16:48 |
| TheJulia | np | 16:49 |
| TheJulia | I also fixed the shard key validation error | 16:49 |
| rpittau | good night! o/ | 17:05 |
| TheJulia | goodnight rpittau | 17:07 |
| TheJulia | https://review.opendev.org/c/openstack/ironic-specs/+/861803 wants reviews so it is not sitting off in a corner feeling like an abandoned spec | 17:10 |
| * TheJulia wonders if we can corgi-gif specs.... | 17:11 | |
| TheJulia | stevebaker[m]: umm.. w/r/t the failure on :\ https://review.opendev.org/c/openstack/ironic-python-agent/+/862656 | 18:02 |
| stevebaker[m] | Hmm? | 18:05 |
| JayF | TheJulia: recheck that before digging | 18:07 |
| JayF | TheJulia: I suspect some of those jobs may be fixed by gmann's patch to tag our branches for ironic-tempest-plugin properly | 18:07 |
| TheJulia | ahh | 18:07 |
| TheJulia | it failed with the dib build though | 18:08 |
| TheJulia | not tempest | 18:08 |
| JayF | aw hell | 18:08 |
| opendevreview | Merged openstack/ironic-lib stable/victoria: CI: Various fixes https://review.opendev.org/c/openstack/ironic-lib/+/860175 | 18:12 |
| opendevreview | Julia Kreger proposed openstack/ironic-inspector master: Use declarative reader/writer with enginefacade https://review.opendev.org/c/openstack/ironic-inspector/+/463768 | 18:38 |
| opendevreview | Julia Kreger proposed openstack/ironic-inspector master: SQLAlchemy 2.0 prep https://review.opendev.org/c/openstack/ironic-inspector/+/860731 | 18:38 |
| TheJulia | hjensas: and wow, those are barebones ports objects, no llc data | 18:40 |
| TheJulia | hjensas: https://review.opendev.org/c/openstack/ironic/+/862933 is awesome, thanks! | 18:48 |
| TheJulia | kubajj: o/ so, I'm not sure why your second patch is failing | 19:01 |
| hjensas | TheJulia: thanks, I'll look for other places where we could use the same pattern. | 19:02 |
| kubajj | TheJulia: I am not really sure either. I did not change much. | 19:02 |
| TheJulia | things *look* like it seems to be working, but I guess I'm wondering what the overall goal of the second change is in the grand scheme of things. Maybe just bridge/duplicate until the rest of the code is moved in? | 19:03 |
| TheJulia | hjensas: ++ chassis_uuid ;) | 19:03 |
| hjensas | TheJulia: yes, probably portgoup and allocations as well. | 19:04 |
| TheJulia | ++ | 19:04 |
| * TheJulia is feeling really low energy as the day is going on | 19:05 | |
| TheJulia | kubajj: Anyhow, I would consier continuing with the other data model stuffs so you at least have that foundation laid. That one in particular, I'm wondering if it is transient, or if we're swallowing errors in an exception catch someplace | 19:06 |
| TheJulia | kubajj: in other words, debug logging might help | 19:06 |
| JayF | kubajj: also, if you wanna put an early review on https://review.opendev.org/c/openstack/ironic/+/864236 I wouldn't hate it | 19:06 |
| * TheJulia runs to the market around the corner to get something to boost the energy. | 19:07 | |
| * JayF needs to unanger the linter | 19:07 | |
| opendevreview | Merged openstack/bifrost master: Switch bifrost to use the dnsmasq DHCP driver https://review.opendev.org/c/openstack/bifrost/+/851682 | 19:24 |
| opendevreview | Merged openstack/ironic-python-agent stable/yoga: Drop python2 from bindep.txt https://review.opendev.org/c/openstack/ironic-python-agent/+/862653 | 19:24 |
| TheJulia | joy linter fun | 19:25 |
| opendevreview | Merged openstack/ironic-python-agent stable/xena: Drop python2 from bindep.txt https://review.opendev.org/c/openstack/ironic-python-agent/+/862654 | 19:30 |
| TheJulia | is review suddenly super slow? | 19:31 |
| TheJulia | for anyone else that is | 19:31 |
| JayF | that link opened pretty quickly for me ^ | 19:31 |
| stevebaker[m] | TheJulia: ok I think that failure will stop with newer diskimage-builder, I'll recheck https://review.opendev.org/c/openstack/requirements/+/862651 | 19:33 |
| TheJulia | ack | 19:33 |
| TheJulia | oh wow it is just crawling for me | 19:39 |
| TheJulia | some stuff works just fine | 19:39 |
| TheJulia | only other slow thing was bugzilla... and arguably... that can be normal | 19:39 |
| opendevreview | Harald Jensås proposed openstack/ironic master: Use association_proxy for port groups node_uuid https://review.opendev.org/c/openstack/ironic/+/864781 | 19:43 |
| TheJulia | JayF: I htink if you fix the import order and take care of the pep8 stuffs, you'll be good to go | 19:45 |
| TheJulia | Also, raised a compound index question | 19:46 |
| JayF | do I need to test the object layer stuff? | 19:46 |
| TheJulia | it may not matter in reality | 19:46 |
| JayF | I have that as a todo but looking at the tests not sure there's anything to test | 19:46 |
| TheJulia | I'd just have a test that ensures we get a list from the db | 19:46 |
| TheJulia | you'll be adding more to ensure it works as you go | 19:46 |
| JayF | ack; will-do | 19:47 |
| JayF | how close is kubajj's first patch to landing? | 19:47 |
| JayF | I guess it shouldn't matter; none of this should merge until the Ironic+Nova specs are seen as OK | 19:47 |
| TheJulia | I +2'ed it | 19:47 |
| TheJulia | I think it is good to go from a data storage point of view | 19:48 |
| JayF | so if I review it, you think it's OK to land that piece now? | 19:48 |
| TheJulia | normally we would want to see more, but I think he is going to end up with ~3 more tables in the grand scheme of things | 19:48 |
| TheJulia | yeah, I think so | 19:48 |
| JayF | even though bugfix releases are cut very, very soon? | 19:48 |
| TheJulia | it wouldn't be harmful | 19:48 |
| TheJulia | yeah, I think it is fine | 19:48 |
| TheJulia | I don't see us needing to be hesitant for bugfixes since we can always backport to them to fix them if there is a huge issue | 19:49 |
| opendevreview | Merged openstack/ironic-python-agent stable/wallaby: Drop python2 from bindep.txt https://review.opendev.org/c/openstack/ironic-python-agent/+/862655 | 19:49 |
| JayF | that's mostly my thought; but with my oper hat on I'd probably be mildly steamed at having to run a ~useless migration | 19:49 |
| kubajj | JayF: what is a shard? | 19:56 |
| TheJulia | you have to run it on any upgrade, at least that is the standard pattern | 19:56 |
| TheJulia | that being said, I do want to go ahead and make us do auto-schema upgrades | 19:56 |
| TheJulia | or at least, support doing it | 19:56 |
| JayF | kubajj: so sharding is a generic term for when you split a big group of things into smaller groups to make them easy to handle; for Ironic; we're adding a shard_key which external services (like nova or networking-baremetal) can use to shard with assistance from Ironic's info | 20:01 |
| JayF | kubajj: how we're doing that for Ironic is specified here https://review.opendev.org/c/openstack/ironic-specs/+/861803 | 20:01 |
| JayF | kubajj: so, as described in that document, the shard table will be used to hold rough, cached information about how many nodes per shard | 20:02 |
| JayF | kubajj: essentially a cached output of `SELECT count(*), shard_key from nodes group by shard_key;` | 20:03 |
| JayF | (that sql may be bad, it's off the top of the head) | 20:03 |
| opendevreview | Steve Baker proposed openstack/bifrost master: Copy shim and grub into tftp and http directories https://review.opendev.org/c/openstack/bifrost/+/849247 | 20:06 |
| opendevreview | Steve Baker proposed openstack/bifrost master: Remove enable_uefi_ipxe https://review.opendev.org/c/openstack/bifrost/+/849248 | 20:06 |
| opendevreview | Steve Baker proposed openstack/bifrost master: Support PXE network boot with grub https://review.opendev.org/c/openstack/bifrost/+/807220 | 20:06 |
| opendevreview | Steve Baker proposed openstack/bifrost master: Restore discovery for dnsmasq dhcp provider https://review.opendev.org/c/openstack/bifrost/+/864787 | 20:36 |
| opendevreview | Merged openstack/ironic master: Implements node inventory: database https://review.opendev.org/c/openstack/ironic/+/862569 | 21:12 |
| hjensas | TheJulia: this test - https://opendev.org/openstack/ironic/src/branch/master/ironic/tests/unit/api/controllers/v1/test_node.py#L603 | 21:54 |
| hjensas | We don't allow deleting a chassis if there are nodes - https://opendev.org/openstack/ironic/src/branch/master/ironic/db/sqlalchemy/api.py#L1222 | 21:54 |
| TheJulia | there is a possible race condition if your mid-transaction on one | 21:55 |
| TheJulia | or if you update one and not the other | 21:55 |
| TheJulia | it is stupid weird, but turned out to be a legit sort of issue | 21:56 |
| hjensas | ok, the actual user pattern would be unset chassis on node(s) then delete chassis? | 21:57 |
| TheJulia | I think actually delete node and chassis and be in-flight on the nodes table | 21:58 |
| TheJulia | the tl;dr is the test can get deleted in the end if a join is used | 21:58 |
| hjensas | ok, yeah seems this could only happen with the _get_chassis_uuid pattern in the api controller. | 21:59 |
| TheJulia | yup | 22:00 |
| TheJulia | specifically more possible if you have a read replica in the mix | 22:00 |
| TheJulia | it hurt my head when I came across the bug because I think someone actuallly had logs of it happening | 22:01 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: WIP: DB implementation of shards / shard_key https://review.opendev.org/c/openstack/ironic/+/864236 | 22:09 |
| TheJulia | well https://zuul.opendev.org/t/openstack/build/8f87a555cf9c45fdb87b6966558440c1 feels new | 22:12 |
| opendevreview | Julia Kreger proposed openstack/ironic-inspector master: Use declarative reader/writer with enginefacade https://review.opendev.org/c/openstack/ironic-inspector/+/463768 | 22:32 |
| TheJulia | lets see if that gets us back to 90%... | 22:33 |
| opendevreview | Julia Kreger proposed openstack/ironic-inspector master: SQLAlchemy 2.0 prep https://review.opendev.org/c/openstack/ironic-inspector/+/860731 | 22:38 |
| TheJulia | much better | 22:39 |
| opendevreview | Julia Kreger proposed openstack/ironic-inspector master: SQLAlchemy 2.0 prep https://review.opendev.org/c/openstack/ironic-inspector/+/860731 | 22:57 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: DB & Object layer for of shards & node.shard_key https://review.opendev.org/c/openstack/ironic/+/864236 | 23:28 |
| JayF | ^^ should be code-good, but can't land until the spec docs (and arguably shouldn't until the nova spec lands) | 23:30 |
| JayF | s/docs/lands/ | 23:32 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!