Tuesday, 2025-10-14

JayFiurygregory: there's no need from my perspective to chat about inspector. We've already merged changes to retire it.00:14
*** gursahibpreetsingh is now known as hargunkaur00:57
iurygregoryJayF, I had this feeling, but since the topic was still in the etherpad I added,I will remove, tks!01:12
hargunkaurHi everyone! đź‘‹01:17
hargunkaurI’m Hargun Kaur, an Outreachy applicant interested in the “Extend Ironic node deployment history API” project. I’ve joined via WeeChat and started exploring the Ironic and Bifrost documentation. Excited to learn and contribute!01:17
rpittaugood morning ironic! o/07:04
rpittauwelcome hargunkaur :)07:04
rpittauiurygregory: besides removing the inspector talk as mentioned by JayF , it looks good, thanks for that!07:06
opendevreviewMerged openstack/ironic-prometheus-exporter master: [IPE] Support iDRAC driver metrics  https://review.opendev.org/c/openstack/ironic-prometheus-exporter/+/95487007:51
opendevreviewStephen Finucane proposed openstack/ironic stable/2025.2: api: Catch correct exception on schema validation error  https://review.opendev.org/c/openstack/ironic/+/96393810:17
opendevreviewStephen Finucane proposed openstack/ironic stable/2025.2: api: Allow bios values to be None  https://review.opendev.org/c/openstack/ironic/+/96394010:19
Ayo[m]hi everyone... (full message at <https://matrix.org/oftc/media/v1/media/download/ASf_ewJ6A0M-Fcb-VXwXNr7H-rMbw9X7FOl8QWz6RpgTPf5_3QNgKVLXcZiXjMDae7LwgLmFmHXazDQfc-E3HABCeaKmNHUAAG1hdHJpeC5vcmcvVVJqakRQWldRR1B2VHdyVXFqakVacmdY>)10:35
FreemanBoss[m]Ayo: what are the commands you ran,?10:42
Ayo[m]the initial command i ran10:44
Ayo[m]./bifrost-cli testenv10:44
Ayo[m]then "./bifrost-cli --testenv" command that gave the error10:44
FreemanBoss[m]The second command won't work without the install parameter10:55
FreemanBoss[m]./bifrost-cli install --testenv10:55
Pointbr8kerHi everyone... My name is David and I'm an outreachy applicant. I'd really appreciate any guidance on getting started. I’m eager to explore and contribute meaningfully to the project. Looking forward to learning from the community and collaborating with you all!11:50
iurygregoryrpittau, tks!11:55
rpittauPointbr8ker: welcome! You can find all the info in the project page :)11:56
Pointbr8kerThanks rpittau!12:06
opendevreviewMahnoor Asghar proposed openstack/ironic master: Fix cleaning example in docs that sets the BMC clock to use correct arg  https://review.opendev.org/c/openstack/ironic/+/96395112:19
*** AB is now known as Guest2898913:20
TheJuliagood morning13:22
*** AB is now known as Guest2899915:14
*** gmaan_pto is now known as gmaan16:00
cardoeSo looking at that patch above... if I wanted to set the clock in a runbook... I couldn't really do that cause I couldn't dynamically get the right date/time in a runbook right?16:08
TheJulia... yeah, we should extend that so you can just say "set the current time of the conductor"16:12
cardoeYeah that's what I want. Is that possible?16:35
*** AB is now known as Guest2900816:56
alegacyTheJulia: looks like openstack/requirements is consulted when validating test-requirements.txt because my addition of NGS to it has caused a build failure.  17:04
alegacyis adding NGS to the global requirements an option to consider, or is that another can of worms?17:05
TheJuliacardoe: dunno off hand, but seems like a reasonable thing to do.17:05
TheJuliaalegacy: can you point me to a job log/change diff ?17:05
alegacyhttps://zuul.opendev.org/t/openstack/build/2b2069d462c24c0eb5355fc2f9e3ffca from this change: https://review.opendev.org/c/openstack/ironic/+/96249317:06
TheJuliaI... guess that could be reasonable to do17:07
TheJuliafwiw, I thought driver-reqs was skipped on that check17:09
alegacyI put it in both driver-requirements.txt and test-requirements.txt since it is used from unit tests... the test-requirements.txt seems to be the one that was flagged17:12
alegacyI could use importutils.try_import('networking-generic-switch') to get the test skipped if it is not present, but wasn't sure if that was appropriate17:13
alegacyI noticed that was done for some of the proliant utils17:13
JayFTheJulia: driver-reqs have always been subject to that check; it's why our dusty broken SNMP driver is holding back the rest of the project :( 17:14
TheJuliaahh17:14
TheJuliaalegacy: for soft dependencies, that is an appropriate path17:14
alegacywhat's your definition of "soft dependency" from a unit test perspective?  Having tox skip tests seems like a risk to having someone accidentally break them without realizing.17:16
TheJuliaThat is indeed the risk with that path17:17
TheJuliaUgh, putting it in g-r as a plugin also prevents release adoption17:19
TheJuliaWhy don’t we break n-g-s into two pieces?17:20
TheJuliaLike a lib-genericswitch17:20
TheJuliaAnd ironic/ngs separately import lib-generic switch to do their needful actions17:20
JayFThis feels like the other side of the cycle we hit with ironic-lib and sushy of not wanting to pay the lib price. Is there any way to integrate the lib-y parts into Ironic directly so NGS takes an ironic dep rather than other way around?17:21
TheJuliaNo, we keep trying to dance to find a way to avoid the lift and we footgun ourselves on every path]17:21
JayFwhat did you mean by "prevents release adoption"17:21
JayFI'm still trying to sort out what you meant by that17:21
TheJuliaSo, basically, we won’t be able to do an end-of-release n-g-s version anymore, it will need to be earlier, but since it is also an ml2 plugin, it needs a late in cycle before release dependency change with neutron-lib17:22
TheJuliaIt’s the same reason we hit the breaks with ironic and dependency mapping, just a different shape17:23
JayFmakes sense17:24
JayFwhat's the code in question? Assuming it's more than just something like object definitions we could copy around in the short term17:24
alegacyisn't this a problem for all libraries?17:24
JayFalegacy: we've been on a recent push to move away from internal libraries where possible because of the massive pain they cause when security issues happen17:25
TheJuliaMight be easier to just talk through first, but would it be okay to make a fresh cup of coffee before starting a call?17:26
JayFI'm OK with a call, I'm OK with keeping it async, I'm OK with hallway tracking it in 3 days17:26
alegacyyes, caffeinate at will17:26
* TheJulia goes and makes coffeeeeeeeeeee17:27
TheJuliahttps://meet.google.com/mmv-hanj-cnc?authuser=017:34
TheJuliaJayF: https://review.opendev.org/c/openstack/networking-generic-switch/+/96203718:07
JayFsummary from call: it isn't a bad thing to put NGS in g-r, we'll just have to exercise more discipline about hitting release deadlines (which is a good thing anyway)18:15
TheJulia++++18:15
JayFclif: it'd be ideal if you could make sure to get eyes on alegacy's patches in NGS and Ironic for standalone networking. Not really urgent since your reviews aren't blocking, but it'd be a good thing to get familiar with.18:17
opendevreviewJulia Kreger proposed openstack/networking-generic-switch master: remove hard requirement on neutron  https://review.opendev.org/c/openstack/networking-generic-switch/+/96330218:20
TheJuliaJayF: I think that should resolve it18:21
clifJayF: ack18:21
JayFTheJulia: alegacy: https://review.opendev.org/c/openstack/networking-generic-switch/+/962037 soft +1 with a question and an ask. Just making sure I understand (and agreement to add the requested test case as a follow up) before upgrading to +218:28
alegacyJayF: thanks, will update.18:30
*** AB is now known as Guest2901018:33
opendevreviewJulia Kreger proposed openstack/networking-generic-switch master: tox: Enable codespell by default and limit it to relevant files  https://review.opendev.org/c/openstack/networking-generic-switch/+/95815418:58
opendevreviewMerged openstack/ironic master: docs: add context around bmc password changes  https://review.opendev.org/c/openstack/ironic/+/96344119:05
JayFupgraded ^^ that review to a +2 with the assumption the test will be a followup if it's landed19:05
alegacyJayF: could you check out this test case "test_delete_port_with_per_port_default_vlan" ... I think it covers the case you asked about (i.e., global config + api parameter)19:11
JayFalegacy: I don't see an NGS config mocked in there with a default vlan set?19:17
JayFam I misreading something?19:17
alegacyOh, I've got 2 tests named the same way in 2 different files... the one in test_netmiko_base.py does this:  switch = self._make_switch_device({'ngs_port_default_vlan': '20'})19:24
JayFack sgtm19:26
JayFI read again and thought that might be what I missed19:26
JayFI don't love the decoder ring of yet-more-layered-config but that's not your fault, not anyone's fault really19:26
JayFclouds be complex :( 19:26
alegacy:)19:38
opendevreviewClif Houck proposed openstack/ironic master: Add a new 'category' field to the Portgroup object  https://review.opendev.org/c/openstack/ironic/+/95571319:57
opendevreviewVerification of a change to openstack/ironic-tempest-plugin master failed: CI: Set the neutron network interface for dhcpless runs  https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/96283720:22
opendevreviewVerification of a change to openstack/networking-generic-switch master failed: Add default vlan to driver interfaces  https://review.opendev.org/c/openstack/networking-generic-switch/+/96203721:39
TheJuliaHmm, that multitenancy job is not happy21:50

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!