Monday, 2025-04-07

rpittaugood morning ironic! o/06:47
rpittaureminder: no regular meeting today, it's vPTG week! :)06:48
rpittautopics and schedule here https://etherpad.opendev.org/p/ironic-ptg-april-202506:48
rpittauwe start at 1400 UTC06:48
kubajjrpittau: 🫡06:50
rpittaulol06:54
kubajjgood morning, btw :D06:58
Ayo[m]Good morning rpittau 07:01
AmarachiOrdor[m]Good Morning! rpittau , good morning everyone!07:04
freemanboss[m]Good morning ironic07:43
opendevreviewSatoshi Shirosaka proposed openstack/ironic-python-agent master: Implement manual cleaning for ContainerHardwareManager  https://review.opendev.org/c/openstack/ironic-python-agent/+/94586211:15
cardoeMorning ironic.12:21
iurygregorygood morning ironic12:26
cardoeMy brain is not processing the network convo from last week JayF and TheJulia. Maybe more coffee.12:32
iurygregorycoffee++12:33
vsaienko_JayF, TheJulia: https://review.opendev.org/c/openstack/ironic/+/900568 - this patch is not needed in order to implement OVN + VTEP support with ironic, what will be the proper way to revert it by taking into account that we did microversion bumps12:50
JayFvsaienko_: can you explain more?12:51
cardoeYeah you'll have to explain some more because that just added those fields as valid fields on local_link_connection12:57
vsaienko_we already have all needed info in local_link_connection information its switch identifier and physical port. One of fields that were added is logical-switch (which is an abstraction of virtual network). This should be created dynamically since we can plug node into different virtual networks. And code that plug port into logical switch (its called bind-ls in ovn terminology) is not implemented on neutron side at all. Please chec13:01
vsaienko_k https://man7.org/linux/man-pages/man7/ovn-architecture.7.html `Life Cycle of a VTEP gateway` paragraph with details on how VTEP works. I have a PoC that adds support to n-g-s will push it later today13:01
TheJuliaAnd here I'm just starting to caffinate13:06
JayFvsaienko_: I was under the impression part of the goal of that change was to take NGS out of the picture?13:07
vsaienko_its not possible13:07
cardoeI'm curious why you say that.13:08
JayFIt seems like the first step would be filing a bug about this being wrong and talking about the right way, and then maybe we can get that guy from juniper to comment in there13:08
TheJuliacat coffee | julia13:10
vsaienko_if you look closer commit that added vtep-logical-switch into neutron was implemented for l2-gtw case, in that case l2gtw creates logical-switch and binds ports13:15
TheJuliaLets please follow bug process since it seems like either bad or incomplete context was presented early on13:17
JayFYeah I mainly just want to get documented how we missed it and see if we can at least get the original requester to put a comment in13:22
vsaienkohttps://bugs.launchpad.net/networking-ovn/+bug/1457569 https://review.opendev.org/c/openstack/networking-ovn/+/22771113:22
vsaienkoand btw, it was never working, https://review.opendev.org/c/openstack/ironic/+/945964 the fields were not passed to neutron at all due to keyError exception13:23
TheJuliaI think we all realize the latter part by now13:54
TheJuliaThe question is how do we fix it and that deserves a bug to track13:54
rpittauironic flamingo PTG starting now https://ptg.opendev.org/ptg.html14:01
JayFhonestly given the previous contributed patch was said to be tested working14:07
JayFI'm wondering what the hell the rest of that story is14:07
JayFbut we may never find out14:07
TheJuliaindeed!14:10
JayFit makes sense if it broke from submitted patch -> commit; we knew that was possible; but vsaienk__ is suggesting the whole idea was flawed from the start14:12
dtantsurcardoe: (inspired by the items I see in the etherpad) do you have any interest in running (Redfish-based) inspection as a service step?14:16
cardoeMaybe.14:16
opendevreviewVasyl Saienko proposed openstack/networking-generic-switch master: WIP: Add support of OVS VTEP devicesy  https://review.opendev.org/c/openstack/networking-generic-switch/+/94655814:30
opendevreviewNicolas Belouin proposed openstack/ironic-python-agent master: netutils: Use ethtool ioctl to get permanent mac address  https://review.opendev.org/c/openstack/ironic-python-agent/+/94656214:42
vsaienkoTheJulia: let me create a bug and describe in details why/how VTEP with ironic does not work and we will decide what to do next14:57
kulsoomsrpittau, masghar: Hi mentors, please let me know if submitting a patch to the Ironic project is also required (along with the summary and report) to be accepted as an intern?15:00
satoshikulsooms: The third task is highly encouraged but optional. 15:48
masgharkulsooms: satoshi is correct, its highly encouraged, but optional :)15:57
JayFI know other outreachy candidates have submitted changes.15:57
JayFI also don't know how much that impacts decisions as I'm not a decider.15:57
rpittauwell, it is indeed highly encouraged :D16:00
opendevreviewJay Faulkner proposed openstack/ironic-python-agent-builder master: Science: Remove ipmitool from tinyipa  https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/94657316:41
freemanboss[m]rpittau: masghar please I just launched this patch, help me check it out please. https://storyboard.openstack.org/#!/story/201139816:43
rpittaufreemanboss[m]: thanks for that, you should report that in launchpad though https://launchpad.net/bifrost we don't use storyboard anymore16:45
freemanboss[m]Please I will be happy if I can get a review as soon as possible so I can know if I am on the right path cause I still have some patch I16:45
freemanboss[m]rpittau: oh alright. I will go with that now16:46
freemanboss[m]rpittau: please I am not getting the icon or link to click on to create a bug16:52
opendevreviewSatoshi Shirosaka proposed openstack/ironic master: WIP Add allow_ironic_project_image_access  https://review.opendev.org/c/openstack/ironic/+/94657516:56
freemanboss[m]https://bugs.launchpad.net/bifrost/+bug/210641617:07
freemanboss[m]rpittau: masghar this is it now17:07
opendevreviewHarald JensÃ¥s proposed openstack/networking-baremetal master: Stop using eventlet for notification listener  https://review.opendev.org/c/openstack/networking-baremetal/+/94657617:42
* JayF jumping into https://meetpad.opendev.org/ironic-ptg-followup-cidocs17:58
iurygregorycardoe, you added the "Firmware Updates via Push" to the ptg etherpad right? do you have links about it I can take a look? :D18:04
opendevreviewJay Faulkner proposed openstack/ironic-python-agent master: Science: base standalone job on redfish standalone  https://review.opendev.org/c/openstack/ironic-python-agent/+/94658018:12
JayFironic-tempest-ramdisk-bios-snmp-pxe18:34
cardoeiurygregory: I did add it. Uhhh lemme find some.18:37
iurygregorytks18:38
opendevreviewJulia Kreger proposed openstack/ironic-tempest-plugin master: remove duplicatd rbac jobs  https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/94658218:51
opendevreviewJay Faulkner proposed openstack/ironic master: PTG: Move CI job descriptions into yaml  https://review.opendev.org/c/openstack/ironic/+/94658419:20
kulsoomssatoshi, masghar, JayF, rpittau: ok thank you 19:22
JayFTheJulia: https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/2PME5HIGOSCOWPX3EFYOYNFCGSPVKVM6/ as discussed in our breakout earlier19:39
opendevreviewQueensly Kyerewaa Acheampongmaa proposed openstack/bifrost master: Clarify usage of master branch and SSH key types in testenv.rst  https://review.opendev.org/c/openstack/bifrost/+/94611619:45
JayFis it somehow possible for our hash ring to get so absolutely bamboozled that a node is perm never working with a conductor?20:10
JayFe.g. Apr 07 20:06:50 ip-172-31-94-37 devstack@ir-api.service[4096727]: DEBUG ironic.api.controllers.v1.node [None req-53c704b3-caff-45ce-89a9-8df66f59d0c2 None admin] Currently there is no conductor servicing node e2601873-0399-44f1-ada5-c3a6485d37c4. {{(pid=4096727) node_convert_with_links /opt/stack/ironic/ironic/api/controllers/v1/node.py:1662}} in the logs20:12
JayFcc satoshi 20:12
JayFnode basically goes to cleaning and just parks there indefinately20:12
JayFI suspect this is somewhat environmentally driven but it's too strange to not try and suss out20:13
JayFnode.reservation in the DB is null, and a restart of the services makes no hcange20:15
satoshihttps://review.opendev.org/c/openstack/ironic/+/946575 This is the patch I'm trying to test.20:16
opendevreviewJay Faulkner proposed openstack/ironic master: Added a FAQ on CI debugging, starts with network  https://review.opendev.org/c/openstack/ironic/+/94658920:29
JayFwell, we managed to cause that issue to occur after reverting to master branch + restarting20:29
JayFso I don't /think/ it's related to your code, but who knows?!20:29
JayF(I also don't think this is indicative of any ironic break; I'm just curious how folks would dig into something like this)20:45
TheJuliaENOBRAIN21:03
TheJuliaJayF: that would be a new one, knowing the algorithm model under the hood I'm not sure how that would have occured21:05
opendevreviewJay Faulkner proposed openstack/ironic-python-agent master: Base standalone job on redfish standalone  https://review.opendev.org/c/openstack/ironic-python-agent/+/94658021:11
opendevreviewJulia Kreger proposed openstack/ironic-tempest-plugin master: remove duplicated rbac jobs  https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/94658221:18
opendevreviewJulia Kreger proposed openstack/ironic-tempest-plugin master: Remove legacy standalone jobs from test execution  https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/94659221:18
opendevreviewJulia Kreger proposed openstack/ironic-tempest-plugin master: CI: Increment jobs for 2025.1 release  https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/94659321:18
TheJuliaJayF: ^21:18
opendevreviewHarald JensÃ¥s proposed openstack/networking-baremetal master: Fix tempest - enable neutron-trunk service plugin  https://review.opendev.org/c/openstack/networking-baremetal/+/94659421:18
JayFTheJulia: https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/946593/1#message-3a1e3b04d9cbe9272526e2ee346540ead42fd8a821:19
TheJuliagood catch, meant to remove htem, thanks21:20
opendevreviewJulia Kreger proposed openstack/ironic-tempest-plugin master: CI: Increment jobs for 2025.1 release  https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/94659321:21
JayF+221:22
hjensasTheJulia: ^ on the topic of what we discussed in PTG. The trunk tests added to ngs broke networking-baremetal because trunk service_plugin not enabled in neutron. (I think). If I wanted to disable it instead, could I set something like tempest_baremetal_feature_enabled_trunks_supported: False in the zuul yaml?21:24
TheJulianot really21:25
TheJuliayou'd have to format it as an entire thing21:25
TheJuliaI think the cleanest path is just "don't turn it on by default"21:25
TheJuliauhhh.21:25
TheJuliahttps://review.opendev.org/c/openstack/networking-generic-switch/+/94608921:25
hjensasoh, yes - that would do it.21:27
JayFhjensas: if you are +1 to that, put the vote on there and I'll single core land it21:28
* TheJulia wonders if there are any stray threads which still need attention today21:29
JayFmy brain is leaking, personally ;)21:29
JayFbetween ptg and wtf'ing at satoshi's devstack21:30
opendevreviewMerged openstack/ironic master: Create Support Resources Page  https://review.opendev.org/c/openstack/ironic/+/94129321:58
opendevreviewMerged openstack/ironic master: Ironic Container Config for IPA Integration  https://review.opendev.org/c/openstack/ironic/+/94479622:39
opendevreviewMerged openstack/networking-generic-switch master: CI: Fix trunks enabled by default  https://review.opendev.org/c/openstack/networking-generic-switch/+/94608923:27

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!