Earlier  
Posted Nick Remark
#openstack-cyborg - 2026-07-07
14:01:42 rlandy Courtesy ping: sean-k-mooney bogdando jgilaber skovili
14:01:46 jgilaber o/
14:02:08 sean-k-mooney o/
14:03:00 rlandy k - let's begin
14:03:26 rlandy pls add any additional topics/reviews/bugs to https://etherpad.opendev.org/p/openstack-cyborg-irc-meeting
14:03:33 melwitt o/
14:03:53 skovili hi
14:04:18 rlandy we have the crowd expected ... starting topics
14:04:34 rlandy #topic: Cyborg documentation review and restructure
14:04:42 rlandy that is my topic
14:04:56 rlandy I started looking at the current Cyborg documentation
14:05:23 rlandy it appears that there are some sections we may want to rewrite completely
14:05:32 rlandy ans some we may want to review and update
14:06:01 sean-k-mooney yep i have been incremently impoving them over time
14:06:16 rlandy The first question was if we want to keep the current structure - and the answer in the etherpad is yes
14:06:19 sean-k-mooney the overall stcuture of the docs is more or less standarised across services
14:06:25 rlandy understood
14:06:28 sean-k-mooney i.e. the user guide contibutor guide admin guide
14:06:44 sean-k-mooney so we dont want to entrilly change the sturue but rather build out the content
14:07:06 rlandy the next question is where to start tracking the content the team wants to change
14:07:47 chandankumar we can use launchpad itself. Like this one https://bugs.launchpad.net/openstack-cyborg/+bug/2150086
14:07:48 sean-k-mooney a full rework fo the docs to build them out is really a multi cycle effort but we coudl use a blueprint or wishlist bug
14:08:11 melwitt could make an etherpad for it too
14:08:20 sean-k-mooney yep i was about to say the sam
14:08:20 rlandy well, do we want a full rework?
14:08:34 sean-k-mooney i think an ether pad is likely the better direction
14:08:34 rlandy minimally we would need a new introduction
14:08:42 rlandy and then a review of the reference sections
14:08:54 rlandy including API, support matric, install, config
14:09:03 jgilaber +1, an etherpad will be more flexible for larger tasks like this
14:09:39 skovili +1
14:09:46 sean-k-mooney the intoduction in the user guide is honestly one of the least impoatn section as that doc in its current form coudl honestly be deleted
14:10:08 sean-k-mooney https://docs.openstack.org/cyborg/latest/index.html is actully ment to be the main docs landing page
14:10:18 sean-k-mooney the intoduciton of what the service is
14:10:23 rlandy except we would want to clarify the purpose of the service
14:10:34 rlandy since that seems to confuse people
14:10:41 sean-k-mooney is soming i wanted to eveutlly handel vai the scope of the project doc and the archterure doc
14:11:11 sean-k-mooney yes so i started workign on https://docs.openstack.org/nova/latest/contributor/project-scope.html for cybrog twice this week and stopped
14:11:28 sean-k-mooney a similar do to that is what i want to replace the intoduction with
14:11:46 rlandy +1
14:12:34 rlandy ok - so I think the team is agreed an etherpad would be best to start dividing and prioritizing the sections to rewrite and review/update, correct?
14:12:41 melwitt +1 agree having an introduction (overview of what the service is) is important
14:13:00 sean-k-mooney it is but i dont think its sould be in the user guide
14:13:08 sean-k-mooney more at the top level
14:13:15 melwitt yes agree top level
14:13:51 sean-k-mooney so in the doc index we shoudl have a what is cyborg section like https://docs.openstack.org/nova/latest/index.html#what-is-nova
14:14:22 sean-k-mooney then you can drill down into the differnt docs seciosn for more detail depenign on what you want
14:15:34 rlandy ok - so going with next steps here:
14:16:07 rlandy #action: rlandy to put together etherpad for the team to start working on doc review/additions
14:16:45 rlandy team can comment/vote on which sections to prioritize/pick tasks from there
14:17:18 rlandy do we have a target date for this work? like the next PTG?
14:17:41 sean-k-mooney i dont want to have that level of process or planning honestly.
14:17:55 sean-k-mooney what i woudl prefer is for us to create a list of the current gaps
14:18:09 sean-k-mooney adn organise around reviews in gerrit
14:18:16 sean-k-mooney and then more review the progress at the next ptg
14:18:20 sean-k-mooney then wait for it
14:18:43 sean-k-mooney as we develop feature and close bugs i hope we will impvoe the docs as we go
14:19:11 sean-k-mooney for example im execting the new drivers to be acocmpianed by a driver specific docuemt that describs how to confugre and use them
14:19:30 sean-k-mooney melwitt: you have started to add api sample tests
14:19:47 sean-k-mooney oen of the request i had was to add a followup to the contibutros docs that explains how they work
14:20:18 melwitt ++ agree that would be good. I will do it
14:21:24 rlandy any other comments or suggestions on the documentation topic?
14:21:39 melwitt yeah for the docs work, I think we could agree as a team we are looking to get it done in the near term, if ppl like the idea. +1 to organize around reviews on gerrit
14:21:42 sean-k-mooney i woudl be include to start with https://docs.openstack.org/cyborg/latest/reference/support-matrix.html adn refactor it to add teh supprot level or work on the main docs landing page first https://docs.openstack.org/cyborg/latest/index.html
14:22:46 sean-k-mooney what we could do is sue a cybrog-docs-update or similar topic to idenfiy related patches
14:23:09 sean-k-mooney it woudl be nice ot maintian a list of those patches and there are in teh ether pad as well
14:23:57 melwitt yes
14:24:46 rlandy ok - let's start with the etherpad and shared gerrit topic: cybrog-docs-update
14:24:55 rlandy we can see how that progress goes
14:26:11 rlandy thank you all for the feedback
14:26:49 rlandy moving on ...
14:27:10 rlandy that's the end of listed topics - so we are moving on to reviews
14:27:17 rlandy #topic: reviews
14:28:02 rlandy There are no reviews listed on the meeting etherpad. Does anyone have any reviews to raise here?
14:28:52 sean-k-mooney i guess a quick update form me on that front
14:29:19 sean-k-mooney i have merge the change to adapt cybrog to the removal fo enforce_scope in oslo.policy
14:29:45 sean-k-mooney beacuse of that i will need to rework the srbac serise slightly
14:30:12 sean-k-mooney i have also merged some clean up patches form takashi removing nova code that was never used form our context object
14:30:20 sean-k-mooney i had to do the same in watcher
14:30:51 sean-k-mooney the other thng to not is the requirements-check job now fails if a project does not have a vailid pyproject.toml
14:31:03 sean-k-mooney so i had to quickly fix that in the cyborg-tempest-plugin repo
14:31:20 sean-k-mooney i had a patch for that so i just brougth it to the front of my srbac serise
14:31:31 sean-k-mooney those are the main changes since the last meeting
14:31:47 sean-k-mooney but i did start reviewing melwitt's api sample test serise this morning
14:32:56 sean-k-mooney that about it for reviewd form me
14:33:02 sean-k-mooney did anyoen want to raise any?
14:33:40 chandankumar https://review.opendev.org/c/openstack/cyborg/+/995935 adds cs10 support in pci-sim but i will rebase on top of sean pci-sim validation patch https://review.opendev.org/c/openstack/cyborg/+/996101
14:33:48 chandankumar Just wanted to highlight here
14:34:24 sean-k-mooney oh ok whiel we proably could merge my patch that was an inital version
14:35:00 sean-k-mooney for context i started working on some linting for the kernel module inclduign a port of the kernesl checkpatch script and adding clang format for code formating
14:35:18 chandankumar one more thing, can we drop depends on from https://review.opendev.org/c/openstack/cyborg-tempest-plugin/+/994580 so that we can merge it independently
14:35:20 sean-k-mooney what is not done yet in my patch is a zuul job
14:35:29 chandankumar Add Cyborg PCI driver lifecycle scenario tests
14:35:57 sean-k-mooney technially yes
14:36:13 sean-k-mooney the depend on was there becasue to use the uart you need the managed mode flag
14:36:26 sean-k-mooney without it ti will be in the guest but you cant write to it and read back the data
14:36:38 sean-k-mooney but technially i coudl drop it since the tests dont assert that currently
14:36:54 sean-k-mooney it just does an ssh to the vm and runs lspci
14:37:35 sean-k-mooney well not quite but basiclly https://review.opendev.org/c/openstack/cyborg-tempest-plugin/+/994580/1/cyborg_tempest_plugin/tests/scenario/test_pci_driver_lifecycle.py#101
14:38:14 sean-k-mooney the lspci testing was in your follow up https://review.opendev.org/c/openstack/cyborg-tempest-plugin/+/995328/3/cyborg_tempest_plugin/tests/scenario/test_pci_driver_lifecycle.py#103

Earlier   Later