| *** pmannidi is now known as pmannidi|Away | 00:03 | |
| opendevreview | Verification of a change to openstack/ironic-lib stable/xena failed: Add platform:base-py[2|3] to bindep https://review.opendev.org/c/openstack/ironic-lib/+/830201 | 00:21 |
|---|---|---|
| arne_wiebalck | Good morning, Ironic! | 07:21 |
| rpittau | good morning ironic! o/ | 07:34 |
| opendevreview | Riccardo Pittau proposed openstack/ironic-python-agent master: Determine codec based on sytem endianness https://review.opendev.org/c/openstack/ironic-python-agent/+/830264 | 07:41 |
| fmuyassarov | Hello folks. We would like to disabled usage of TLSv1_1, and TLSv1_2 in Ironic conductor and instead use TLSv1_3. Setting [ssl]_version to TLSv1_3 doesn't seem to block 1_0 and 1_1. How can we disable TLS versions prior to 1.3? | 07:49 |
| opendevreview | Harald Jensås proposed openstack/ironic-lib master: Remove python version 2 pkgs from bindep https://review.opendev.org/c/openstack/ironic-lib/+/830319 | 07:50 |
| janders | Good Morning arne_wiebalck rpittau fmuyassarov and Ironic o/ | 07:55 |
| rpittau | hey janders :) | 07:57 |
| fmuyassarov | good morning janders | 07:58 |
| opendevreview | Jacob Anders proposed openstack/sushy master: Re-trying InsertMedia call with TransferProtocolType if required https://review.opendev.org/c/openstack/sushy/+/830322 | 08:25 |
| opendevreview | Jacob Anders proposed openstack/sushy master: Re-trying InsertMedia call with TransferProtocolType if required https://review.opendev.org/c/openstack/sushy/+/830322 | 08:28 |
| dtantsur | morning folks | 08:54 |
| dtantsur | fmuyassarov: I wonder if switching to the combined ironic process will make this problem no longer important | 08:55 |
| dtantsur | or do you use a multi-conductor setup? | 08:55 |
| arne_wiebalck | hey janders rpittau fmuyassarov dtantsur o/ | 09:17 |
| rpittau | hey arne_wiebalck :) | 09:18 |
| MahnoorAsghar | Good morning Ironic | 09:48 |
| opendevreview | Merged openstack/tenks master: [CI] Optimise irrelevant files https://review.opendev.org/c/openstack/tenks/+/830272 | 09:58 |
| opendevreview | Aija Jauntēva proposed openstack/ironic master: Update python-dracclient version https://review.opendev.org/c/openstack/ironic/+/830348 | 10:38 |
| iurygregory | good morning Ironic o/ | 10:52 |
| dtantsur | morning iurygregory | 10:56 |
| dtantsur | folks, another fallout from the CI refactoring: https://review.opendev.org/c/openstack/ironic-python-agent/+/829719 (needed to fix the IPA CI) | 10:56 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/xena: Build the new cirros image even when netboot is the default https://review.opendev.org/c/openstack/ironic/+/830352 | 10:58 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/wallaby: Build the new cirros image even when netboot is the default https://review.opendev.org/c/openstack/ironic/+/830353 | 10:58 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/victoria: Build the new cirros image even when netboot is the default https://review.opendev.org/c/openstack/ironic/+/830354 | 10:58 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic bugfix/19.0: Build the new cirros image even when netboot is the default https://review.opendev.org/c/openstack/ironic/+/830355 | 10:58 |
| opendevreview | Merged openstack/ironic-lib stable/xena: Add platform:base-py[2|3] to bindep https://review.opendev.org/c/openstack/ironic-lib/+/830201 | 11:07 |
| arne_wiebalck | is it correct that the sample policy we produce has the same target multiple times? https://docs.openstack.org/ironic/latest/configuration/sample-policy.html | 11:12 |
| * dtantsur has no clue | 11:16 | |
| arne_wiebalck | it seems wrong to have a target specified multiple times as FWIU these will go into a hash and the last one wins | 11:31 |
| arne_wiebalck | TheJulia: ^^ | 11:32 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-python-agent stable/xena: Stop defaulting the standalone job to netboot; use the correct boot mode https://review.opendev.org/c/openstack/ironic-python-agent/+/829721 | 11:32 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-python-agent bugfix/8.4: Stop defaulting the standalone job to netboot; use the correct boot mode https://review.opendev.org/c/openstack/ironic-python-agent/+/830356 | 11:33 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-python-agent bugfix/8.3: Stop defaulting the standalone job to netboot; use the correct boot mode https://review.opendev.org/c/openstack/ironic-python-agent/+/829722 | 11:33 |
| dtantsur | so many backports to do... | 11:33 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-python-agent master: [WIP] Create mount points before mounting https://review.opendev.org/c/openstack/ironic-python-agent/+/829067 | 11:42 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-python-agent master: [WIP] Create mount points before mounting https://review.opendev.org/c/openstack/ironic-python-agent/+/829067 | 11:43 |
| dtantsur | rpittau: FYI the job change in https://opendev.org/openstack/ironic/commit/be0ebde0107e1addff5b4bb96d038f8bb804224f is wrong, we need to revert it | 11:50 |
| dtantsur | or at least apply something different to ironic, I'll check it | 11:51 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic bugfix/18.1: CI: use a custom cirros partition image instead of the default https://review.opendev.org/c/openstack/ironic/+/829678 | 11:52 |
| dtantsur | rpittau: maybe like this ^^^ | 11:52 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic bugfix/18.1: CI: use a custom cirros partition image instead of the default https://review.opendev.org/c/openstack/ironic/+/829678 | 11:55 |
| rpittau | dtantsur: I'll check that after lunch, wasn't that working before? | 11:56 |
| dtantsur | rpittau: well, it was silently using stable/xena instead of the patch in question :) | 11:56 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-inspector master: DNM test the CI https://review.opendev.org/c/openstack/ironic-inspector/+/830397 | 12:00 |
| rpittau | Applying bugfix 18.1 to ironic should work | 12:01 |
| rpittau | As it's done for ipa | 12:01 |
| rpittau | Anyway, I need a bigger screen for that than my phone's :) | 12:02 |
| dtantsur | :D | 12:06 |
| opendevreview | Jacob Anders proposed openstack/sushy master: Re-trying InsertMedia call with TransferProtocolType if required https://review.opendev.org/c/openstack/sushy/+/830322 | 12:14 |
| opendevreview | Verification of a change to openstack/ironic-python-agent master failed: Stop defaulting the standalone job to netboot; use the correct boot mode https://review.opendev.org/c/openstack/ironic-python-agent/+/829719 | 13:00 |
| iurygregory | really zuul? =( | 13:03 |
| dtantsur | yeaah :( | 13:05 |
| dtantsur | honestly, I would like to stop hacking on the CI at some point this week :D I have job to do | 13:05 |
| opendevreview | Merged openstack/python-ironic-inspector-client master: Ensure tox.ini is ASCII https://review.opendev.org/c/openstack/python-ironic-inspector-client/+/829830 | 13:16 |
| TheJulia | is zuul needing cookies? | 13:40 |
| dtantsur | always | 13:41 |
| TheJulia | :) | 13:41 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder stable/wallaby: Use native Centos 8 Stream for dib images https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/830358 | 14:07 |
| * TheJulia hopes that clears up CI issues for stable/wallaby | 14:14 | |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: [WIP] Use pycdlib to extract deploy iso https://review.opendev.org/c/openstack/ironic/+/819121 | 14:25 |
| arne_wiebalck | TheJulia: is it correct that the sample policy we produce has the same targets multiple times? https://docs.openstack.org/ironic/latest/configuration/sample-policy.html (e.g. "baremetal:node:get") | 14:27 |
| TheJulia | arne_wiebalck: so we added some new policy targets, and the fallback is baremetal:node:get in the deprecated set | 14:46 |
| TheJulia | since there was no *real* way to delineate it for previous usage | 14:47 |
| arne_wiebalck | TheJulia: right, but this is not new targets, it is the same target defined multiple times | 14:47 |
| arne_wiebalck | like "baremetal:node:get" is defined multiple times as a target with several rules | 14:48 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: [WIP] Use pycdlib to extract deploy iso https://review.opendev.org/c/openstack/ironic/+/819121 | 14:48 |
| TheJulia | arne_wiebalck: all deprecated | 14:48 |
| TheJulia | so yes, it does look like it doesn't de-deuplicate deprecated rules | 14:48 |
| arne_wiebalck | the not commented lines in the policy file are used, no? | 14:49 |
| TheJulia | I wonder if it is "parent_rule": "current_rule" sort of relationship | 14:51 |
| TheJulia | same thing occurs with baremetal:node:update\ | 14:51 |
| arne_wiebalck | and with others ... | 14:51 |
| TheJulia | if commented out, it is the default | 14:51 |
| arne_wiebalck | right | 14:51 |
| TheJulia | well, if freshly generated, it is all the default | 14:51 |
| TheJulia | :) | 14:51 |
| arne_wiebalck | isn't https://docs.openstack.org/ironic/latest/configuration/sample-policy.html what would be generated? | 14:52 |
| arne_wiebalck | or where does that come from? | 14:52 |
| TheJulia | it is generated from the code | 14:52 |
| arne_wiebalck | right | 14:53 |
| * TheJulia wonders if the deprecated rule enumeration just renders confusing output | 14:53 | |
| arne_wiebalck | is it only confusing or actually wrong? | 14:54 |
| opendevreview | Riccardo Pittau proposed openstack/ironic-python-agent master: Use utf-16-le if BOM not present https://review.opendev.org/c/openstack/ironic-python-agent/+/830264 | 14:54 |
| arne_wiebalck | oslo does not complain about a target defined multiple times | 14:55 |
| TheJulia | I'm starting to think the rendered output is actually wrong | 14:55 |
| TheJulia | in part because we had to add rules | 14:56 |
| TheJulia | arne_wiebalck: when your generating, are you passing any options and do have enforce_scope and enforce_new_defaults in your config? | 14:56 |
| arne_wiebalck | TheJulia: we started with a policy file generated with tox IIRC | 15:00 |
| TheJulia | okay, that gives me a starting point | 15:00 |
| arne_wiebalck | it is basically the policy sample we have in the docs | 15:01 |
| opendevreview | Merged openstack/ironic stable/xena: Build the new cirros image even when netboot is the default https://review.opendev.org/c/openstack/ironic/+/830352 | 15:17 |
| TheJulia | arne_wiebalck: https://bugs.launchpad.net/oslo.policy/+bug/1945336 sound familiar? | 16:21 |
| arne_wiebalck | TheJulia: boom | 16:21 |
| arne_wiebalck | TheJulia: I think that is it. | 16:22 |
| TheJulia | https://bugs.launchpad.net/oslo.policy/+bug/1870187 makes me sad | 16:22 |
| arne_wiebalck | uh, that would have been nice to know :) | 16:25 |
| arne_wiebalck | instead of starting from the (buggy) sample file, we could have had one with our overrides | 16:27 |
| TheJulia | yeah, so tripleo is just starting with outright overrides, fwiw | 16:27 |
| TheJulia | but we have the expertise to handle that | 16:27 |
| * TheJulia tries to trigger an internal discussion regarding oslo policy | 16:28 | |
| arne_wiebalck | I wonder if the policy generator has the same bug as the sample generator. | 16:29 |
| arne_wiebalck | it seems likely | 16:29 |
| TheJulia | yeah | 16:29 |
| opendevreview | Merged openstack/ironic bugfix/19.0: Build the new cirros image even when netboot is the default https://review.opendev.org/c/openstack/ironic/+/830355 | 16:31 |
| TheJulia | arne_wiebalck: yeah, same code, just slightly different entry points | 16:33 |
| arne_wiebalck | not surprising | 16:33 |
| arne_wiebalck | the sample policy file has 1400 lines and I guess many will start from it | 16:34 |
| TheJulia | yeah | 16:35 |
| TheJulia | I'm reading the generator code now | 16:35 |
| opendevreview | Merged openstack/ironic bugfix/18.1: CI: use a custom cirros partition image instead of the default https://review.opendev.org/c/openstack/ironic/+/829678 | 16:41 |
| TheJulia | oh wow, the way it displays deprecated rules is awful | 16:41 |
| TheJulia | it is in format of "old_rule_name": "new_rule_name" | 16:43 |
| TheJulia | talk about confusing | 16:44 |
| arne_wiebalck | heh | 16:48 |
| arne_wiebalck | there is another thing: unless I define some of the *default* rules explicitly, e.g. for "baremetal:node:list" I get deprecation warnings | 16:52 |
| arne_wiebalck | but if I add them, I get "this rule is redundant" warnings | 16:52 |
| TheJulia | yeah, I'm staring at the code trying to think how to cleanly solve this | 16:54 |
| TheJulia | I *think* the first thing is to just strip deprecated rules by default | 16:56 |
| arne_wiebalck | you're looking at the oslo generator code? | 16:57 |
| TheJulia | yes | 16:58 |
| arne_wiebalck | yeah ... one thing I did at some point was to remove all comments, so I could actually read the policy file | 17:01 |
| TheJulia | yeah | 17:01 |
| TheJulia | I'm working on a change to by default comment this stuff out | 17:02 |
| TheJulia | since it is inherently wrong if the names have changed | 17:02 |
| TheJulia | arne_wiebalck: how does https://paste.openstack.org/show/812868/ look? | 17:06 |
| arne_wiebalck | TheJulia: thanks! There seems a 'defined' or 'introduced' missing. Also: how about not using the 'target':'rule' format for a rule replacement ... or is there any reason for this? | 17:10 |
| * arne_wiebalck will step away for a moment | 17:12 | |
| TheJulia | so, I don't know the reason behind target:rule | 17:13 |
| TheJulia | or in this case, deprecated target:rule | 17:13 |
| TheJulia | I think I'm going to have to document some stuff based upon code state | 17:13 |
| rpittau | good night! o/ | 17:13 |
| opendevreview | Verification of a change to openstack/ironic stable/wallaby failed: Build the new cirros image even when netboot is the default https://review.opendev.org/c/openstack/ironic/+/830353 | 17:14 |
| arne_wiebalck | target:rule is fine, I just find that deprecated target:rule does not make any sense ... or does it? | 17:17 |
| * arne_wiebalck really steps away :) | 17:17 | |
| opendevreview | Merged openstack/ironic-python-agent-builder stable/xena: Create random character devices so systemd does not crash https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/830200 | 17:19 |
| TheJulia | oh, I think I see what is going on | 17:19 |
| TheJulia | artifact of someone trying to keep explicit policy files a thing after policy in code was a thinh | 17:19 |
| * TheJulia seeks clarifcation and sees the person's irc client has been idle for over 64 hours | 17:20 | |
| dtantsur | see you tomorrow folks o/ | 17:33 |
| arne_wiebalck | TheJulia: this is for the above bug about deprecated rules marked active ? | 17:59 |
| TheJulia | well, I'm trying to decipher intent | 18:00 |
| TheJulia | but basically yes as well | 18:00 |
| arne_wiebalck | right, it seems intertwined | 18:00 |
| TheJulia | Indeed | 18:01 |
| TheJulia | I *feel* like the intent was more to help map for version compatability | 18:01 |
| TheJulia | but at the same time, we shouldn't be publishing a bunch of deprecated things in latest | 18:02 |
| opendevreview | Merged openstack/ironic-python-agent master: Stop defaulting the standalone job to netboot; use the correct boot mode https://review.opendev.org/c/openstack/ironic-python-agent/+/829719 | 18:04 |
| arne_wiebalck | yes, the idea was to help understand what changed | 18:04 |
| arne_wiebalck | fixing the missing comment but should have priority imo | 18:06 |
| TheJulia | I'm just going to do it as an all in one change and see what the reviewer ssay | 18:07 |
| arne_wiebalck | comments are easily ignored ... but with this rules active, it gets really confusing :-D | 18:07 |
| TheJulia | since this is a bit confusing/insane | 18:07 |
| arne_wiebalck | totally | 18:07 |
| arne_wiebalck | thanks for picking this up, TheJulia | 18:08 |
| arne_wiebalck | bye everyone o/ | 18:37 |
| opendevreview | Merged openstack/ironic stable/victoria: Build the new cirros image even when netboot is the default https://review.opendev.org/c/openstack/ironic/+/830354 | 18:43 |
| TheJulia | arne_wiebalck: change posted to oslo.policy | 19:26 |
| stevebaker[m] | good morning | 20:01 |
| arne_wiebalck | stevebaker[m]: good morning! | 20:09 |
| arne_wiebalck | TheJulia: awesome, thanks | 20:09 |
| arne_wiebalck | TheJulia: there was an option already to have the deprecated rules added? | 20:10 |
| TheJulia | arne_wiebalck: not one that could be toggled | 20:10 |
| TheJulia | good morning stevebaker[m] | 20:10 |
| arne_wiebalck | TheJulia: right, but in the _yield_yaml function | 20:10 |
| arne_wiebalck | TheJulia: the change looks good to me | 20:11 |
| arne_wiebalck | TheJulia: and imo it has the potential to save a lot of operators a lot of hassle :) | 20:12 |
| TheJulia | arne_wiebalck: thanks! | 20:24 |
| TheJulia | stevebaker[m]: if you wouldn't mind, a quick review on https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/830358 would be appreciatec | 20:24 |
| TheJulia | appreciated | 20:24 |
| arne_wiebalck | TheJulia: I will see if I can try the change tomorrow and then leave a review. | 20:25 |
| arne_wiebalck | bye everyone o/ | 20:25 |
| stevebaker[m] | TheJulia: done! | 20:29 |
| TheJulia | stevebaker[m]: thanks! | 20:32 |
| TheJulia | arne_wiebalck: that would be much appreciated | 20:32 |
| opendevreview | Merged openstack/ironic-python-agent-builder stable/wallaby: Use native Centos 8 Stream for dib images https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/830358 | 21:59 |
| iurygregory | I've pushed the patch with the Cycle Highlights for our project https://review.opendev.org/c/openstack/releases/+/830530 feel free to provide feedback =) | 22:48 |
| opendevreview | Jacob Anders proposed openstack/sushy master: Re-trying InsertMedia call with TransferProtocolType if required https://review.opendev.org/c/openstack/sushy/+/830322 | 23:17 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!