15:00:11 <TheJulia> #startmeeting ironic
15:00:11 <openstack> Meeting started Mon Jul 16 15:00:11 2018 UTC and is due to finish in 60 minutes.  The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:12 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:15 <openstack> The meeting name has been set to 'ironic'
15:00:18 <TheJulia> o/
15:00:32 <kaifeng> o/
15:00:34 <bdodd> o/
15:00:37 <rloo> o/
15:00:50 <etingof> o/
15:01:06 <TheJulia> I hope everyone has had a wonderful weekend!
15:01:17 <mjturek> o/
15:01:19 <TheJulia> And now it is time for our weekly meeting!
15:01:26 <stendulker> o/
15:01:27 <jroll> \o
15:01:27 <TheJulia> The agenda can be found on the wiki
15:01:33 <rpioso> o/
15:01:45 <TheJulia> #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting
15:02:34 <TheJulia> #topic Announcements/Reminders
15:03:09 <TheJulia> #info This week is R-6 for the Rocky development cycle
15:03:21 <TheJulia> #info Final non-client libraries must be released this week.
15:04:35 <TheJulia> #info For this week, that means ironic-lib and sushy must be released.
15:05:08 <TheJulia> #info Next week is R-5, which is final release of client libraries, feature freeze, requirements freeze, etc.
15:05:13 <TheJulia> Any questions?
15:06:05 <TheJulia> #info Tomorrow, Tuesday at 12:00 UTC, We will be having a call to try and plan/discuss any last minute details and sort through anything contentious that needs to be discussed before the PTG.
15:06:32 <TheJulia> #link https://etherpad.openstack.org/p/ironic-rocky-midcycle
15:06:49 <TheJulia> Anyone have anything to announce ?
15:07:53 <rloo> thxs to the folks that helped out on the bug day thing last week
15:07:58 * TheJulia secretly replaces everyone's decaf coffee
15:08:43 <TheJulia> rloo: ++
15:08:47 <mjturek> ooh yeah that might be good to announce - next bug day
15:08:49 * dtantsur is still somewhere on the Baltic see :)
15:08:55 <mjturek> August 2nd @ 1:00 PM - 2:00 PM
15:08:56 <rloo> and thanks TheJulia for the awesome email wrt ironic
15:09:14 <mjturek> sorry, 13:00 UTC to 14:00 UTC
15:09:15 <TheJulia> mjturek: those times in UTC?
15:09:33 <TheJulia> mjturek: awesome
15:09:34 <mjturek> TheJulia: yep
15:09:57 <TheJulia> #info Next Bug Day - August 2nd @ 13:00 to 14:00 UTC PM
15:10:07 <TheJulia> Anyone have anything else, or shall we move on?
15:11:02 <TheJulia> Anyway, moving on!
15:11:22 <TheJulia> #topic Review action items from previous meeting
15:11:42 <TheJulia> #link http://eavesdrop.openstack.org/meetings/ironic/2018/ironic.2018-07-09-15.00.log.html
15:12:01 <TheJulia> It looks like we have nothing to discuss here, I believe everything was taken care of
15:12:07 <TheJulia> So... moving on!
15:12:27 <TheJulia> #topic Review subteam status reports
15:12:38 <TheJulia> #link https://etherpad.openstack.org/p/IronicWhiteBoard
15:12:56 <TheJulia> Starting around line 157
15:13:14 <rloo> the bugs stats should have been improved after last week's bug day? (L166ish)
15:13:28 <dtantsur> I expected that as well, but dunno
15:13:34 <dtantsur> maybe we're growing more new bugs?
15:13:43 <TheJulia> Has anyone updated them?
15:14:36 <rloo> the BIOS config stuff; are the TODOs as of today (vs 2 July)?
15:15:10 <TheJulia> rloo: July 2nd, I believe... however I suspect they are still valid
15:15:10 <etingof> just a quick note that this script relies on the `ironic-triaged` tag to be present on the triaged task
15:15:13 <TheJulia> yolanda_: ^^^
15:15:47 <TheJulia> I've not updated the bug numbers, if someone else can, that would be awesome
15:16:07 <dtantsur> wdym by "updating numbers"? applying the tag?
15:16:29 <TheJulia> update the block of text indicating but counts and status on the etherpad
15:16:39 <mjturek> TheJulia: trying to find the numbers, but not sure how? https://storyboard.openstack.org/#!/story/list?status=active&project_group_id=75 according to this is 517
15:17:05 <TheJulia> We need to split this apart from general subteam status report
15:17:06 <dtantsur> TheJulia: yep, I updated it
15:17:16 <TheJulia> dtantsur: ahh, awesome
15:17:29 <dtantsur> mjturek: I generate it from https://github.com/dtantsur/ironic-bug-report
15:17:38 <mjturek> dtantsur: oh cool!
15:18:07 <rloo> mkrai: wrt your graphical console interface; i updated it. please just summarize the latest status; i don't think we want to carry weekly statuses on this; otherwise all of them will get way too long.
15:18:34 <TheJulia> looks like hshinna is also not on :\
15:18:53 <rloo> are we done with management interface refactoring? L231ish?
15:18:57 <TheJulia> (regardingg bios interface)
15:19:36 <TheJulia> rloo: I believe we are functionally done. etingof is there anything else on your plate for that?
15:19:36 <etingof> quick note on the new bug panopticon -- https://storyboard.openstack.org/#!/board/83
15:19:53 <TheJulia> Regarding rescue: the nova patch landed over the weekend. \o/
15:20:00 <rloo> dtantsur: wrt classic driver removal, is that the latest (today) status? L252ish
15:20:08 <rloo> WHEEE, great news about rescue!!!
15:20:08 <TheJulia> so anything we have held up in ironic pending that, we can merge
15:20:19 <jroll> \o/
15:20:20 <dtantsur> rloo: it is
15:20:22 <rloo> TheJulia: just the CI patch i believe!
15:20:45 <rloo> dtantsur: OH. it is still july, sorry i thought it was august. moar caffeine...
15:20:51 <dtantsur> :)
15:21:04 <dtantsur> TheJulia: I remember approving something about rescue CI today
15:21:05 <etingof> TheJulia, I believe the mgmt interface implementation is done, we may want to run some more tests in the CI perhaps
15:21:34 <TheJulia> Also, with rescue, we landed a little bit of code in nova to gracefully handle the prior API version to enable the upgrade to be more graceful
15:21:54 <TheJulia> so \o/
15:22:16 <rloo> That was probably TheJulia's favorite part! :D
15:22:20 <dtantsur> :)
15:22:41 * dtantsur is still too happy the classic drivers are gone - great release!
15:22:50 <jroll> ++
15:23:20 <TheJulia> I think that largely means based on current statuses, we only have the conductor locality work and anything of high impact that we would like to see.
15:23:59 * TheJulia feels like we're in a REALLY good place for releasing without any headaches and having a huge improvement this cycle
15:24:28 <TheJulia> Anyway, I'm good to move on? Everyone else ready to move on to proprities?
15:24:31 <TheJulia> priorities
15:24:32 <mjturek> this is probably a stretch (and not sure it's the right place to bring up) but I think I'm finished with the implementation here https://storyboard.openstack.org/#!/story/1749057
15:24:47 <mjturek> if it's too late, I understand
15:25:22 <TheJulia> mjturek: at least the ironic-lib change didn't seem major so I dont see why not?!?
15:25:53 <mjturek> TheJulia okay would appreciate some reviews if everyone's not already too swamped
15:27:06 <rloo> wrt priorities. we don
15:27:16 <rloo> 't need deploy steps there?
15:27:32 <TheJulia> rloo: I was thinking for doc items or follow ups?
15:27:46 <rloo> TheJulia: yeah, but i don't know that they are priorities for this week.
15:27:47 <dtantsur> I'd like https://storyboard.openstack.org/#!/story/2002868 on priorities
15:27:48 <TheJulia> #topic Priorities for the week
15:28:02 <TheJulia> rloo: ack
15:28:05 <rloo> TheJulia: but i am fine if we leave it there.
15:28:13 <rloo> just thinking we want to get any features in
15:28:15 <TheJulia> Anyone aware of sushy patches that need reviews this week?
15:28:31 <rloo> is there anything in nova-land that we need/want to get in for their FF?
15:29:15 <dtantsur> TheJulia: there are a few open, I'm not sure how close to landing they are
15:29:41 <TheJulia> dtantsur: agreed :\
15:29:51 <TheJulia> bdodd: are you aware of anything that is needed in sushy for this release?
15:30:01 <rloo> also, our client freeze/release is next week. is conductor locality the only one that has a client part?
15:30:25 <TheJulia> rloo: I'm not actively aware of anything nova side that must land for us this cycle
15:30:33 <TheJulia> rloo: I believe that is correct
15:30:44 <rloo> we're good then :)
15:30:52 * dtantsur pokes etingof
15:31:06 <rloo> dtantsur: what about your reset-interfaces stuff?
15:31:07 <bdodd> TheJulia: Not that I currently know of. Still testing/debugging my Redfish BIOSInterface code. But I haven't hit any issues with the sushy code.
15:31:08 <dtantsur> rloo: the reset_interfaces work will need a client change
15:31:13 <dtantsur> will post tomorrow hopefully
15:31:18 * dtantsur is fighting with unit tests
15:31:47 <rloo> ok, i think dtantsur's stuff is a priority. otherwise, it'll make it hard for our users to change hw types
15:32:07 <jroll> well, it's already hard :)
15:32:20 <rloo> true, but we had classic drivers before.
15:32:25 <jroll> I guess we are dropping drivers though, let's make it a priority
15:32:43 <TheJulia> ++
15:32:48 * etingof does not see anything important to be merged with sushy atm
15:32:56 <TheJulia> I just added an entry for the client only side patch that dmitry posted
15:33:08 <jroll> well, the API needs to land too
15:33:13 <TheJulia> bdodd: awesome!
15:33:16 <rloo> we're still missing the 'get me hw-type driver properties for non-default interfaces' i think but ...
15:33:35 * etingof shamelessly suggests merging this little sushy-tools patch of his own though -- https://review.openstack.org/#/c/568636/
15:33:36 <patchbot> patch 568636 - sushy-tools - Add unittests for OpenStack nova driver
15:33:45 <TheJulia> etingof: :)
15:33:54 <TheJulia> etingof: we don't need to cut sushy-tools this week :)
15:34:05 * etingof /o\
15:34:09 <TheJulia> sorry :(
15:35:09 <TheJulia> I think the priorities are basically good at this point, sans having the reset interface patches from dmitry. We can add that later in the week if we're in agreement
15:35:57 <rloo> we might re-eval during mid-cycle tomorrow anyway. (hopefully it won't change)
15:36:03 <TheJulia> ++
15:36:11 <TheJulia> Anyway, seems like we can move on
15:36:55 <TheJulia> Everyone good with proceeding?
15:37:34 <rloo> the bios interface support stuff is questionable (seems like it includes status stuff) but anyway, let's move on :)
15:38:12 <TheJulia> slightly less statusy now
15:38:32 <TheJulia> Anyway! Moving on!
15:38:38 <TheJulia> #topic RFE Review
15:39:03 <TheJulia> I have an RFE that I wanted to bring up for discussion or raise awareness of because it is an interesting problem and it is something we should consider addressing.
15:39:06 <TheJulia> #link https://storyboard.openstack.org/#!/story/1546949
15:39:21 <openstackgerrit> Dmitry Tantsur proposed openstack/ironic master: Add reset_interfaces parameter to node's PATCH  https://review.openstack.org/582951
15:39:55 <TheJulia> Any thoughts?
15:40:14 <jroll> "why not" is my only thought :)
15:40:26 <mjturek> implementation is here https://review.openstack.org/#/c/346723/
15:40:27 <patchbot> patch 346723 - ironic-python-agent - Parallel erase disk devices
15:40:47 <TheJulia> Anyone up for taking that on?
15:40:48 <rloo> is that an rfe or a bug?
15:40:54 <TheJulia> definitely an RFE
15:41:44 * TheJulia wonders if we have sufficient consensus to mark it approved and move on?
15:42:17 <rloo> so looking at the patch and the -1 there. the question or a question is whether to make it configurable?
15:42:33 <rloo> oh wait, maybe they did make it configurable.
15:42:46 <rloo> anyway, if it is an rfe, the description needs to be updated
15:42:57 <rloo> other than that, i'm good with it being approved.
15:42:57 <TheJulia> they did not
15:43:15 <TheJulia> I'm good as well, and it seems like jroll is, so I feel like we have consensus
15:43:19 <rloo> oh, no config? i'm only looking at the comments, and saw 'done' but dunno what was done.
15:43:19 * TheJulia goes and updates it
15:43:42 <rloo> sorry, is this configurable or not? (or do we want it configurable?)
15:43:51 <TheJulia> yeah, it looks not configurable. I think it might be better to not to... but we're also dealing with hardware that could have quarks
15:43:58 <TheJulia> so maybe configurable would be best?
15:43:58 <jroll> I'm fine with it with or without a config - I don't think the config is necessary, but for whatever reason we make everything configurable, so can deal with it
15:44:39 <rloo> it is an IPA config, not ironic config
15:44:59 <jroll> would need to be able to configure it via ironic somehow, though
15:45:12 <TheJulia> yeah, I think that is an implementation detail we can work through
15:45:22 <kaifeng> i remember ipa know options from command line..
15:45:35 <TheJulia> I have updated the rfe accordingly
15:46:21 <rloo> TheJulia: it doesn't say anything about configuration?
15:46:48 <TheJulia> rloo: I just put a comment on it stating that configurability is an implementation detail that can be figured out
15:46:51 <rloo> if we don't care, should we mention that?
15:47:02 <rloo> ok, so we want it.
15:47:17 <TheJulia> yeah, I think so
15:47:37 <TheJulia> Time to proceed to Open Discussion?
15:47:41 <rloo> ok, i'll update the description then. thx.
15:47:49 <TheJulia> rloo: k
15:47:59 * TheJulia wonders if it iddn't save
15:48:44 <TheJulia> #topic Open Discussion
15:49:55 <mjturek> need to hammer out any details for tomorrow's call?
15:50:06 <rloo> TheJulia: are we trying to get https://review.openstack.org/#/c/576717/ for rocky?
15:50:06 <patchbot> patch 576717 - ironic-specs - Add Boot from Ramdisk spec
15:50:30 <rloo> or just the spec approved?
15:50:33 <TheJulia> So, one thing worth noting during open discussion. I've had some arm folks reach out about deploying arm gear in noop node. I've already got a grub related knob to help enable that, there are some outstanding questions, but I might post a patch or two more to make things easier/simpler
15:50:49 <TheJulia> rloo: I'd like to get it merged, as-is I think it is only like 270-something lines
15:51:07 <TheJulia> And it would be a nice thing to provide the scientific community
15:51:09 <openstackgerrit> Dmitry Tantsur proposed openstack/ironic-tempest-plugin master: Add API tests for the reset_interfaces parameter  https://review.openstack.org/582985
15:51:21 <rloo> TheJulia: i haven't actually read it; does it have a client component? (I'm guessing not)
15:51:22 <TheJulia> since they would be the ones jumping on it and trying to kick the tires as it were
15:51:30 <TheJulia> rloo: no, no client component
15:51:38 <rloo> TheJulia: ok thx.
15:52:27 <TheJulia> Oh, BTW, Get your summit submissions in!
15:53:27 <TheJulia> I have nothing else to discuss or mention today
15:53:35 <TheJulia> Anyone have anything?
15:53:42 <kaifeng> if time permits, I'd like to know opinions about https://storyboard.openstack.org/#!/story/2002977
15:53:46 <TheJulia> Oh, 6 minues remaining
15:53:51 <kaifeng> :)
15:54:13 <TheJulia> kaifeng: Ohh, interesting
15:54:34 <kaifeng> mainly, the boot mode timeout option, I'm not sure if it's mandatory or optional in the IPMI standard
15:55:06 <kaifeng> but I encountered several models didn't support this option
15:55:20 <TheJulia> kaifeng: what sort of hardware did you encounter this on?
15:55:33 <TheJulia> I _think_ support is optional
15:56:09 <TheJulia> kaifeng: was it that we never successful set it for long enough, or we never actually set it?
15:56:49 <kaifeng> actually the raw command never succeeds, so the pxe mode is not set
15:57:48 <rloo> based on the comment, it looks like if the cmd doesn't succeed, it'll time out in 60s.
15:58:24 <rloo> eiddccfiultrehgnfhtrjnjfglketkleicvvlvdjibjc
15:58:26 <kaifeng> right, but the timeout is an edge case
15:58:46 * TheJulia wonders if rloo just turned on encryption :)
15:58:57 <TheJulia> or a cat decided to capture her keyboard
15:59:00 <rloo> oops :D
15:59:05 <kaifeng> in most cases, it would make it within timeout limits
16:00:10 <TheJulia> seems like something we ought to fix, exact details on how I personally have no opinion on at present
16:00:13 <rloo> kaifeng: do you mean 'timeout limits = 60s' or some other timeout limit?
16:01:06 <kaifeng> My thought is if this bmc option is optional, we can just wrap this raw command into a try catch, log a warning if not bmc doesn't support it
16:01:29 <kaifeng> rloo: I mean the power action timeout (60s)
16:01:40 <rloo> kaifeng: that won't solve your issue though? you want the inspector to stop, not continue?
16:01:56 <TheJulia> rloo: I'm thinking the same thing
16:02:12 <TheJulia> I'm going to end the meeting, but we can keep discussing this
16:02:15 <TheJulia> Thanks everyone!
16:02:30 <rloo> a demain!
16:02:35 <dtantsur> thanks!
16:02:50 <jroll> thanks \o
16:02:57 <TheJulia> rloo: you've got me confused :)
16:03:03 <TheJulia> #endmeeting