Friday, 2026-07-17

opendevreviewMerged openstack/ironic-python-agent stable/2026.1: Fix: Remove legacy agent_token fallback code  https://review.opendev.org/c/openstack/ironic-python-agent/+/99761500:22
opendevreviewSteve Baker proposed openstack/ironic master: Switch ironic-tempest-standalone-advanced to autodetect deploy  https://review.opendev.org/c/openstack/ironic/+/99766601:29
opendevreviewRiccardo Pittau proposed openstack/bifrost master: Use cirros 0.6.3 as default cirros version  https://review.opendev.org/c/openstack/bifrost/+/93217507:54
opendevreviewMerged openstack/ironic-python-agent stable/2025.1: Fix: Remove legacy agent_token fallback code  https://review.opendev.org/c/openstack/ironic-python-agent/+/99762410:26
opendevreviewRiccardo Pittau proposed openstack/bifrost master: Use cirros 0.6.3 as default cirros version  https://review.opendev.org/c/openstack/bifrost/+/93217513:19
TheJuliabrraaaains13:41
dtantsurWHERE???14:01
TheJuliaI has NONE14:04
dtantsurDamn, I thought you intersected a braincell... I've been catching one the whole morning.14:05
TheJuliaI am not an orange.14:06
TheJuliaI'm more void ;)14:06
dtantsur:D14:07
dtantsurIn our pair, my wife is void and I'm orange14:08
TheJuliaheh14:17
TheJuliaspeaking of void and oranges, they defended the bed from the corgi this morning14:28
TheJuliacardoe: would you be available for a specs review jam session next week ?14:29
dtantsurThat's the spirit! How die the overlord endure the rebellion?14:29
cardoeTheJulia: yes. Can I take the call from the lazy river?14:29
TheJuliahe went up front in the RV because the void bops him on the head.... a lot.14:30
dtantsurWork-from-tent is always approved of and encouraged14:30
TheJuliacardoe: .... I mean... I guess so?!14:30
cardoeWell then yes.14:30
TheJuliaI mean, I'm a prime work form RV'er.14:30
TheJuliaAnyway, what day and approximately what time would be good for you ?14:30
TheJuliataking the corgi for his walk before my next meeting14:31
cardoeusually 8am to 9:30am central are terrible for me14:31
cardoeLet's call it 10am14:32
cardoeSo 10am and later any day of the week should be good.14:32
cardoeExcept Monday14:32
* cardoe glares at Outlook for refreshing oddly.14:33
dtantsurShould be 5pm for me, pretty doable (I can normally do 6pm too). Only Thursday already fully booked.14:36
cardoeIf ya want I'll get up earlier.14:37
dtantsurNot a problem for me, I work later hours. Will be an issue if we want e.g. rpittau there.14:38
* dtantsur works smth like 11-7 or even later sometimes14:38
cardoeI usually work 8-5 local time during the summer14:39
dtantsurTimezones, right? A lot of meetings start at 5pm for me (including the Ironic one)14:42
cardoeYep14:43
TheJuliacan we translate it all to UTC to collalesce on a time ?14:49
* TheJulia doesn't know the timezones in people's heads14:49
dtantsurIn the summer, I normally work until 1700 UTC. My RH calendar is up-to-date if you want to check it.14:50
TheJuliaso if we did 10am central, that would be 8 am for me which *should* work... I think.14:51
dtantsurTheJulia: it's the Tuesday's security slot, right? Without Jay, we may just as well take it.14:52
TheJuliaso technically, I'm already triple booked for that slot, but let me see if I can get rid of one of them14:53
dtantsurI can do 1600 UTC on Tue or Wed too. If you ask nicely, I can even do Friday :)14:54
opendevreviewMerged openstack/networking-generic-switch master: Try Two: Split NGS admin docs  https://review.opendev.org/c/openstack/networking-generic-switch/+/99746115:00
dtantsur\o/15:01
opendevreviewMerged openstack/virtualbmc master: Migrate setup configuration to pyproject.toml  https://review.opendev.org/c/openstack/virtualbmc/+/99757915:22
TheJuliameeting invite sent, I got my other meeting I moved15:49
TheJuliaor at least, promised to be move due to the other person having a conflict15:49
opendevreviewTakashi Kajinami proposed openstack/python-ironicclient master: Add openstack-python3-next-jobs job  https://review.opendev.org/c/openstack/python-ironicclient/+/99775415:53
opendevreviewTakashi Kajinami proposed openstack/python-ironicclient master: Remove Python 3.10 support  https://review.opendev.org/c/openstack/python-ironicclient/+/99775515:53
opendevreviewTakashi Kajinami proposed openstack/ironic master: Remove Python 3.10 support  https://review.opendev.org/c/openstack/ironic/+/99775615:55
opendevreviewMerged openstack/virtualbmc master: Update supported python versions  https://review.opendev.org/c/openstack/virtualbmc/+/99758016:40
TheJuliaAny demand/interest for risc-v out there?16:40
opendevreviewTakashi Kajinami proposed openstack/sushy master: Add openstack-python3-next-jobs  https://review.opendev.org/c/openstack/sushy/+/99776916:44
opendevreviewTakashi Kajinami proposed openstack/sushy master: Remove Python 3.10 support  https://review.opendev.org/c/openstack/sushy/+/99777016:44
opendevreviewVerification of a change to openstack/tenks master failed: Add openstack.cloud collection to requirements  https://review.opendev.org/c/openstack/tenks/+/99306017:02
opendevreviewJulia Kreger proposed openstack/ironic-specs master: Libssh based switch configuration drivers?  https://review.opendev.org/c/openstack/ironic-specs/+/99765317:39
JayFTheJulia: in a "that's cool" sense, sure. I don't see it getting much traction in larger orgs (yet?)17:45
opendevreviewJay Faulkner proposed openstack/ironic-python-agent master: Allow limiting what config can be set in mDNS  https://review.opendev.org/c/openstack/ironic-python-agent/+/99763817:48
opendevreviewJulia Kreger proposed openstack/networking-generic-switch master: Add Juniper Junos multicast BUM replication support  https://review.opendev.org/c/openstack/networking-generic-switch/+/98130719:21
rm_work[m]hey do you guys know/remember why Ironic RPC addresses conductors by embedding the hostname into the topic on the client side (get_topic_for -> ironic.conductor_manager.<host>) instead of passing it as server= on prepare()? the conductor binds with `Target(server=self.host)`, but the client never sets server, even though it even parses the host back out of the topic in _prepare_call. is that just historical, or does something19:52
rm_work[m]depend on it being in the topic string?19:52
rm_work[m]I ran into this in the http-rpc work and worked around it, but i'm at the point where i'm going through and trying to clean up and validate all my assumptions and this still stands out to me as an odd one-off19:52
TheJuliaso the jsonrpc stuffs was largely for just single api and single conductor. The messaging model is fanout mdoel because the client doesn't lock to a specific conductor, they are addressed based upon the destination, and each conductor is bound to their specific topic for the api to address to it19:57
TheJulia(and conductor itself)19:58
opendevreviewJulia Kreger proposed openstack/networking-generic-switch master: Add opt-in MTU management for switch ports  https://review.opendev.org/c/openstack/networking-generic-switch/+/98483220:00
opendevreviewJulia Kreger proposed openstack/networking-generic-switch master: Add opt-in port carrier bounce on bind  https://review.opendev.org/c/openstack/networking-generic-switch/+/98483320:00
opendevreviewJulia Kreger proposed openstack/networking-generic-switch master: Add opt-in STP edge port and BPDU guard support  https://review.opendev.org/c/openstack/networking-generic-switch/+/98484720:00
rm_work[m]hmm ok, I think I'm going to make a kind of garbage CR just so I can visualize and explain the issue and you can tell me why i'm dumb maybe? XD20:05
rm_work[m]with AI these days I find it easier/quicker just to do that with code than be vague trying to badly explain what i'm trying to figure out lol20:08
opendevreviewMathieu Parent proposed openstack/ironic-python-agent master: Allow auto_tls_allowed_clock_skew via kernel cmdline or vmedia  https://review.opendev.org/c/openstack/ironic-python-agent/+/99778820:20
TheJuliafair, I guess :)20:30
TheJuliaI REALLY think we should default out to a larger clock skew window20:34
opendevreviewAdam Harwell proposed openstack/ironic master: Address conductors by Target.server, not host-in-topic  https://review.opendev.org/c/openstack/ironic/+/99778920:39
rm_work[m]TheJulia: ^^20:39
rm_work[m]AI seems to think that wouldn't immediately break everything but like, I do not trust that assessment XD20:39
TheJuliaso, that is not that different20:41
TheJuliafwiw20:41
rm_work[m]not that different, as in... it would be fine?20:42
TheJuliait might be20:42
rm_work[m]I am spinning up a job right now (regardless of what happens with this) that tests Ironic with my HTTP-RPC stuff, so I could just Depends-On this CR and see what happens20:43
rm_work[m]it removes a very fragile bit where we have to dot-walk the topic to get the host on the oslo.messaging side20:43
rm_work[m]and just... makes Ironic match literally every other service20:43
TheJuliayeah, I'm curious how CI takes it20:44
rm_work[m]the other oddity I'm working around is that ironic includes the service name in the topic (ironic.conductor_manager)20:45
TheJuliaThat would be a super breaking change20:45
rm_work[m]but it's... fine, mostly just cosmetic I think20:45
TheJuliayeah20:45
rm_work[m]yeah I figured that part out :D20:45
rm_work[m]I have just had to special-case it20:46
rm_work[m]HTTP-RPC is super duper cool BTW20:48
rm_work[m]RMQ can go fly a kite20:48
TheJuliaheh, yeah20:49
rm_work[m]we completely removed RMQ from our prod clouds20:49
rm_work[m]it is glorious20:49
rm_work[m]went from an incident a week that was RMQ related to ... nothing20:49
TheJulianice20:57
rm_work[m]i kinda want to go scream from rooftops that everyone needs to immediately go Rabbit hunting and purge it from the ecosystem :D21:00
rm_work[m]but I think a lot of folks likely just don't have the time or interest in trying something experimental T_T21:00
TheJuliayeah, and I think it sort of gets miss-used due to the nature21:03
TheJuliaand its harder to tease all that apart21:03
rm_work[m]kafka works brilliantly for notification transport, and http-rpc just drops in to replace rabbit, very easy, but I am sure there are edge cases21:04
rm_work[m]though it has worked with every service we run and that is quite a few non-standard21:04
opendevreviewJulia Kreger proposed openstack/networking-generic-switch master: Add ngs_ssh_allowed_algorithms for SSH cipher control  https://review.opendev.org/c/openstack/networking-generic-switch/+/99122521:05
TheJuliayeah, I've only been through a few project's rpc code and its a bit of a blur to me21:06
TheJuliaI just know I had a broken deployment once when we lost the message bus contents21:06
TheJuliabut that was also a super duper long time ago21:06
rm_work[m]if you ever get a wild hair and just want to see what a better system looks like, hit me up :P21:07
rm_work[m]I have lots of devstack jobs that spin it up you could look at too21:07
rm_work[m]though realistically I realize that'll never happen because ya'll are so busy lol21:07
opendevreviewMerged openstack/ironic-python-agent stable/2025.2: Fix: Remove legacy agent_token fallback code  https://review.opendev.org/c/openstack/ironic-python-agent/+/99744921:11
TheJuliaalways busy, unfortunately22:44
rm_workhttps://review.opendev.org/c/openstack/oslo.messaging/+/997783?tab=change-view-tab-header-zuul-results-summary23:10
rm_workSo that job tests ironic with my original workaround23:11
rm_workI’m going to set it to use the CR I posted earlier and we’ll see23:11
rm_workIf it passes, do you think you’d consider approving? I mean not without possibly some cleanup and thorough review, but I mean conceptually no hard blockers?23:13
TheJuliaeh, quite likely, but I would need to sit down and take a good look at it just to make sure.23:50

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!