17:00:19 <TheJulia> #startmeeting ironic
17:00:20 <openstack> Meeting started Mon Dec 12 17:00:19 2016 UTC and is due to finish in 60 minutes.  The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:21 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:23 <openstack> The meeting name has been set to 'ironic'
17:00:25 <jlvillal> o/
17:00:25 <mariojv> o/
17:00:27 <xhku> o/
17:00:27 <aslezil> o/
17:00:29 <lucasagomes> o/
17:00:31 <joanna> o/
17:00:32 <TheJulia> #chair jroll
17:00:32 <bfournie> o/
17:00:32 <openstack> Current chairs: TheJulia jroll
17:00:36 <rama_y_> o/
17:00:37 <jlvillal> All hail our new chair-person!
17:00:41 <rloo> o/
17:00:43 <krtaylor> o/
17:00:44 <rloo> HAIL :)
17:00:45 <TheJulia> jlvillal: lol
17:00:47 <vdrok> o/
17:00:47 <TheJulia> The agenda for today’s meeting:
17:00:49 <Michael-zte2> o/
17:00:49 <yuriyz|2> o/
17:00:52 <davidlenwell> o/
17:00:53 <TheJulia> #link https://wiki.openstack.org/wiki/Meetings Ironic#Agenda_for_next_meeting
17:00:59 <TheJulia> doh
17:01:01 <TheJulia> #undo
17:01:02 <openstack> Removing item from minutes: <ircmeeting.items.Link object at 0x7fa25e74e490>
17:01:16 <Michael-zte2> Hello everyone
17:01:26 <TheJulia> #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting
17:01:33 <Michael-zte2> Glad to join this project
17:01:34 <dtantsur> o/
17:01:35 <milan> o/
17:01:44 <TheJulia> #topic Announcements and Reminders
17:01:47 <rloo> welcome Michael-zte2
17:02:03 * jlvillal can't tell difference between the two links...
17:02:13 <TheJulia> So other than jroll is out for the majority of the week, I have nothing :)
17:02:15 <xavierr> o/
17:02:20 <Michael-zte2> :)  <rloo>
17:02:30 <TheJulia> jlvillal: Meetings/Ironic :)
17:02:31 <mjturek1> o/
17:02:34 <joanna> jlvillal: lack of '/' ;)
17:02:34 <rloo> deadline for travel support program is this week: http://lists.openstack.org/pipermail/openstack-dev/2016-December/108645.html
17:02:42 <rloo> for PTG ^^
17:02:43 <jlvillal> Ah
17:03:01 * dtantsur will be on PTO starting next week and til January
17:03:10 <rajinir> o/
17:03:37 * jlvillal envies dtantsur
17:03:48 <TheJulia> #info PTG travel support deadline is this week http://lists.openstack.org/pipermail/openstack-dev/2016-December/108645.html
17:03:58 <jlvillal> dtantsur: Enjoy Moscow! :)
17:04:20 <rpioso> o/
17:04:21 <dtantsur> thanks :)
17:04:26 <TheJulia> Any other announcements ?
17:04:38 <rloo> oh, maybe mention, seems like some gates are broken?
17:04:39 <jlvillal> Where do priorities get mentioned?
17:05:08 <mariojv> is it just IPA gate or all gates?
17:05:13 <rloo> python-ironicclient gate has been broken since last Thurs.
17:05:29 <rloo> fix for ^ is https://review.openstack.org/#/c/408205/
17:05:46 <rloo> IPA gate is broken... not sure what the fix is yet, is that right?
17:05:48 <TheJulia> IPA is now broken due to tiny core linux being down
17:05:53 <vdrok> yup, fix on review, but now i started seeing races in set_console/get_console in tempest api tests :(
17:06:06 <TheJulia> specifically, tiny core's mirror is down.
17:06:09 <rloo> vdrok: is that for ironic patches?
17:06:17 * jlvillal remembers that we really need to get TinyCore linux mirror setup on OpenStack infrastructure
17:06:30 <TheJulia> jlvillal: I believe that was just proposed :)
17:06:35 <vdrok> rloo: both ironic and client, the jobs run the same api tests everywhere
17:06:53 <TheJulia> jlvillal: that or I'm going crazy
17:07:05 <pas-ha> TheJulia: I think we already asked for TC mirror and were refused
17:07:10 <rloo> vdrok: sigh, so a different breakage? please update our whiteboard with that :-(
17:07:20 <jlvillal> TheJulia: I remember it being discussed before. I think sambetts_ was looking into it, or at least thinking about it.
17:07:22 <TheJulia> Any other items for announcement or shall we proceed?
17:07:39 <pas-ha> as TC itself is not CDN-ed they do not want to mirror that
17:07:47 <lucasagomes> pas-ha, :-/ would be great to have a mirror for it in OS
17:08:35 <pas-ha> lucasagomes: surely - TC down would simply become not ours problem, but infras
17:08:59 <pas-ha> and surprise - they are not happy :)
17:09:21 <lucasagomes> heh I figure
17:09:26 <TheJulia> #topic subteam status reports
17:09:42 <TheJulia> #link https://etherpad.openstack.org/p/IronicWhiteBoard
17:10:24 <TheJulia> Starting around line 88
17:11:03 <rloo> jlvillal: are you adding the testing stuff for rolling upgrades, would be good if you could consolidate with the grenade stuff under 'status as of most recent weekly meeting'
17:11:18 <jlvillal> Ah, okay.
17:12:57 <mgould> o/
17:13:30 <rloo> vdrok, vsaienk0: what is the order wrt reviewing code for portgroups & attach/detach.
17:13:48 <vdrok> rloo: no particular order, atm they are independent
17:13:52 <jlvillal> TheJulia: Are the priorities the same this week?
17:14:07 <rloo> vdrok: ok, so we need to get both into ironic, before we can get the nova-related ones ?
17:14:10 <jlvillal> rloo: Should Line 85 be deleted?
17:14:11 <vdrok> attach/detach seems the most important tho :)
17:14:21 <rloo> jlvillal: yup
17:14:25 <TheJulia> jlvillal: they appear to be in need of updating, One of the listed priorities has already landed
17:14:38 <vdrok> rloo: yes, attach/detach and portgoup mode/properties for sure
17:14:42 <rloo> vdrok: if we get attach/detach (and none of the portgroup), can we then do the nova one?
17:14:44 * jlvillal goes to delete Line 85
17:15:01 <TheJulia> side question, is anyone else getting disconnected from etherpad?
17:15:07 <vdrok> rloo: not fully, we need the one with portgroup mode for config drive generation
17:15:09 <rloo> TheJulia: i think you/we need to update those priorities based on what we see wrt subteam reports etc
17:15:09 <jlvillal> TheJulia: Solid for me
17:15:16 <xavierr> TheJulia: nope
17:15:22 <vdrok> TheJulia: I'm OK
17:15:25 <lucasagomes> TheJulia, it's working here
17:15:30 <rloo> vdrok: if you had to choose one -- could you put in priorities, which patch to review?
17:15:39 <vdrok> rloo: OK, will do now
17:15:43 <jlvillal> vdrok: Do you think attach/detach should be #1 priority? Instead of portgroups?
17:16:00 <vdrok> jlvillal: yes, tho client side of portgroups is also needed
17:16:11 <vdrok> but it should be an easier review
17:16:32 <jlvillal> I'll let TheJulia change priorities as chair
17:16:36 <rloo> are folks ok if i switch the order and move driver comp up to #3? (cuz dtantsur is away after this week)
17:16:43 <rloo> wrt the priorities ^^
17:16:56 <jlvillal> I'm okay with that.
17:17:16 <vdrok> rloo: are you ok with merging portgroups and attach/detach into one group?
17:17:19 <TheJulia> rloo: I'm good with that, I was thinking the same
17:17:31 <vdrok> rloo: or I can just put portgroups under attach
17:17:32 <jlvillal> dtantsur: As a question, any idea how many more driver composition patches until completion?
17:17:35 <TheJulia> other than driver comp until he is away, it looks like we could make decent progress on notifications
17:17:35 <lucasagomes> rloo, seems ok
17:17:36 <rloo> vdrok: nope, links etc are diff, so would be confusing i think
17:17:39 <dtantsur> jlvillal, no so few
17:17:50 <vdrok> ok
17:17:52 <rloo> TheJulia: notifications is really close
17:17:52 <dtantsur> I'll try to put up as many as possible this week
17:18:08 <jlvillal> dtantsur: Sorry I got confused by "no so few"
17:18:36 <mariojv> yep, notifications just needs a rebase and reviews on the CRUD patch
17:18:43 <mariojv> after that, all the other notification patches have 2 +2s
17:18:50 <dtantsur> jlvillal, s/no/not/. maybe 4, 5, 6 more
17:18:52 <yuriyz|2> yes will do tomorrow
17:18:59 <mariojv> thanks yuriyz|2
17:19:04 <jlvillal> dtantsur: Ah, okay. A fair amount. Thanks!
17:19:09 <yuriyz|2> comments are resolved
17:19:26 <TheJulia> I moved notifications up above boot from volume since I need to rebase all the BFV stuff and make updates
17:19:51 <jlvillal> dtantsur: I added comment to the driver composition section. Feel free to correct it.
17:19:58 <rloo> TheJulia: guess that's ok, although looks like both need to be rebased and boot-from-volume is more important i think.
17:20:26 <rloo> dtantsur: wrt bugs, we have 3 critical ones?
17:20:36 <dtantsur> rloo, CI breakages
17:20:56 <rloo> dtantsur: the only one with a patch is the one for the client? should we put that as a priority?
17:21:28 <TheJulia> It looked close essentially ready to land when I last glanced at it about an hour ago
17:21:36 <rloo> TheJulia: ok, then no need
17:21:49 <dtantsur> rloo, likely yes
17:22:24 <rloo> ++ thx JayF for bug triaging
17:22:58 <rloo> and nice that security groups is DONE
17:23:06 <TheJulia> With that, I Think we're good for the sub team update and priorities.  We don't have any stuck specs that I'm aware of that require discussion, nor any discussion topics, so I guess if there are no objections, time for RFE review?
17:23:07 <lucasagomes> ++
17:23:39 <JayF> o/
17:23:42 <lucasagomes> (the ++ was to the security groups) but I'm fine to move on the RFE review as well
17:23:51 <rloo> sec, wrt snmp driver. anyone know who the contact person might be?
17:24:12 <rloo> that is now non-deprecated :)
17:24:25 <rloo> supported i mean (forgot the word for a sec)
17:24:28 <JayF> rloo: probably mat or someone else from internap
17:24:34 <JayF> rloo: they did all the work to get it back going
17:24:45 <rloo> JayF: ok, i'll put mat's name down for now
17:24:51 <xhku> +1
17:25:49 <TheJulia> #topic RFE review
17:26:07 <rloo> https://bugs.launchpad.net/ironic/+bug/1642719
17:26:07 <openstack> Launchpad bug 1642719 in Ironic "[RFE] RADOSGW authentication mechanism for storing config drive in Ceph" [Wishlist,In progress] - Assigned to Anup (anup-d-navare)
17:26:25 <rloo> i think that's actually ready to be approved w/o a spec but thought I'd bring it up
17:26:57 <JayF> I'm OK with approving it, it's sad that Anup isn't here though.
17:27:25 <TheJulia> I looked at that earlier and I feel like a spec would be a good idea, but not absolutely required IMHO.
17:27:38 <rloo> JayF: yes, true, but i've brought up other rfe's w/o the person around
17:27:58 <rloo> TheJulia: anything in particular missing, or just want the spec so you can see that things are addressed (ie, as a checklist)
17:27:58 <lucasagomes> JayF, if we fix the problem with generating the configdrive at rebuild, do we need to store it at all ?
17:28:06 <mariojv> i think it's sufficiently detailed
17:28:14 <mariojv> given that ceph can use same APIs as swift, from what i understand
17:28:17 <JayF> lucasagomes: no ; but that bug is going to take a lot of work to fix, that neither nova nor ironic has prioritized
17:28:20 <lucasagomes> but other than that yeah, the RFE seems straight forward
17:28:54 <lucasagomes> JayF, yeah indeed. Plus backward compat
17:29:12 <rloo> so are we good to approve? (or only good not requiring a spec?)
17:29:23 <TheJulia> rloo: Checklist is a nice to have, but it feels like it could be built upon, we just need full scope with regards to credentials and the like, but still good with approving.
17:29:46 <JayF> I also did talk to Anup, IDK if it's mentioned in that RFE, but he volunteered to make sure this was documented
17:29:46 <jlvillal> lucasagomes: Yeah, I wasn't sure how difficult it would be to fix the bug about rebuild
17:29:47 <vdrok> I'm good with approving, seems straightforward
17:29:56 <JayF> even if it meant we had to add docs for the configdrive support for swift
17:29:56 <rloo> ok, if no objections, will approve
17:29:59 <JayF> :)
17:30:05 <TheJulia> None here. :)
17:30:06 <jlvillal> I did ping Anup on he is attempting to join the meeting.
17:30:10 <rloo> JayF: ++ saw your comment about docs
17:30:13 <jlvillal> s/on he/and he/
17:30:24 <rloo> jlvillal: thx, honestly, no need for him to join the meeting.
17:30:27 * lucasagomes is ok approving as well w/o a spec
17:30:30 <rloo> i'm moving on to the next
17:30:32 <rloo> https://bugs.launchpad.net/ironic/+bug/1647813
17:30:32 <openstack> Launchpad bug 1647813 in Ironic "[RFE] CI for using nova with resource classes and new-style flavors" [Wishlist,In progress] - Assigned to Jim Rollenhagen (jim-rollenhagen)
17:30:42 <rloo> he better not be in this meeting :)
17:30:44 <lucasagomes> jlvillal, yeah it will be hard, it requires some changes to the API + Nova
17:30:46 <TheJulia> Heh
17:31:03 <JayF> I'm sorta meh on the idea this is an RFE in general
17:31:07 <mariojv> given that's a new ci job and not a big ci rework
17:31:11 <JayF> basically it's work we've more or less committed to doing already
17:31:12 <mariojv> i think it's fine to approve
17:31:23 <TheJulia> I feel like this does not require a spec, that it is more an ongoing compatibility issue and that just needs to be addressed. tl;dr approve!
17:31:24 <rloo> i'm good with +A that.
17:31:26 <JayF> and jim just wanted a bug to put his ci-related patches towards
17:31:32 <rloo> JayF: well, jroll tagged it as rfe :)
17:31:34 <JayF> so I think it's an easy +A
17:31:38 <lucasagomes> ++
17:31:42 <JayF> rloo: I know, I argued with him about it afterwards :P
17:31:47 <rloo> any objections to not approving? going once ...
17:31:55 <rloo> twice ... <waits>...
17:32:04 <rloo> DONE! :)
17:32:15 <rloo> last but not least: https://bugs.launchpad.net/ironic/+bug/1648448
17:32:15 <openstack> Launchpad bug 1648448 in Ironic "[RFE]HPSUM firmware update as cleaning step for iLO drivers " [Wishlist,In progress] - Assigned to Aparna (aparnavtce)
17:32:18 <rpioso> I would prefer an approach that applies to all in-tree drivers.  Is there such a thing for firmware update?  Also, I would like support of this functionality as an out-of-band cleaning step, not just in band.
17:33:36 <mariojv> since that's in-band (for now), that's only going to be happening with the agent_ilo driver right?
17:33:38 <TheJulia> Truthfully I feel like this is something better suited inside a hardware manager, not across a management network as virtual media.
17:33:47 <mariojv> if so, i'm wondering... yup, exactly what TheJulia pointed out
17:33:53 <JayF> TheJulia: I disagree vehemently
17:34:10 <JayF> TheJulia: if a BMC supports updating firmware out of band, not allowing that support to be used is borderline silly
17:34:24 <JayF> AIUI it hooks in an ISO with the firmware to upgrade to in it
17:34:28 <TheJulia> JayF: how so, I see this as the hardware manager knowing it can mount an ISO to get files in later cleaning steps?
17:34:29 <JayF> then triggers the firmware upgrade via BMC
17:34:31 <mariojv> this RFE is explicitly for an inband clean step, though
17:34:33 <rpioso> +1 to oob!
17:34:37 <JayF> oh really/
17:34:39 <mariojv> oob would be cool too
17:34:40 <JayF> did I misread it?
17:34:42 <mariojv> but not part of RFE
17:34:46 <mariojv> "This RFE proposes the HPSUM(HP Smart Update Manager) based firmware update as a inband clean step for iLO drivers.
17:34:48 <mariojv> "
17:35:01 <TheJulia> JayF: Totally with you for OOB :)
17:35:12 <JayF> oh, so they want to hook up an ISO, and use that HP ISO to upgrade firmware?
17:35:13 <JayF> Hmm.
17:35:20 <TheJulia> yup
17:35:22 * JayF votes for needs-spec
17:35:25 <mariojv> this actually says something about hardware manager too though
17:35:33 <mariojv> last sentence of the rfe
17:35:54 <TheJulia> For use by the hardware manager, so it seems like the conductor would be orchestrating virtual media attachments for the ISO.  I'm also in the needs-spec category
17:35:56 <JayF> Yeah, given there seems to be confusion around what this is trying to do
17:36:00 <JayF> we should totally request a spec
17:36:08 <mariojv> agreed
17:36:13 <rloo> thx, needs spec it is :)
17:36:21 <rloo> TheJulia: I'm done, over to you!
17:36:26 <TheJulia> Thank you rloo!
17:36:35 <TheJulia> #topic Open Discussion
17:37:11 * jlvillal listens to the crickets...
17:37:19 <TheJulia> Anyone have anything to discuss? Patches that need reviewing? etc? :)
17:37:52 <rloo> oh, do we want to discuss the party we're going to have now that jroll is away? :)
17:37:55 <gabriel-bezerra> Hi folks, this xavierr 's patch is out https://review.openstack.org/408298 addressing the power-off before set-boot-device issue we discussed a couple weeks ago
17:37:55 <TheJulia> jlvillal: I wonder if the crickets can play other music :)
17:38:02 <jlvillal> heh :)
17:38:17 <gabriel-bezerra> thanks vdrok and mgould for the reviews
17:38:29 <TheJulia> rloo: What sort of party? ;)
17:38:33 <NobodyCam> :)
17:38:37 <xavierr> yes, ty mgould and vdrok :)
17:38:40 <gabriel-bezerra> one of the jobs of our ci is broken while this patch doesn't land
17:39:05 <vdrok> rloo: portgroups client patch review party?
17:39:33 <Michael-zte2> Where do I get the latest information on this project?  From here, https://etherpad.openstack.org/p/IronicWhiteBoard ?
17:40:01 <jlvillal> Michael-zte2: That is a good resource on what is currently happening as far as issues and progress.
17:40:05 <vdrok> depends on what information you want :) that is more of a status thing
17:40:16 <TheJulia> gabriel-bezerra: It look xavierr received some feedback this morning :)
17:40:25 <xavierr> Michael-zte2: White board + IRC :)
17:40:39 <jlvillal> Michael-zte2: That is not documentation though.
17:40:46 <xavierr> TheJulia: haha yeah o/
17:40:52 <Michael-zte2> jlvillal:thank you
17:41:16 <rpioso> Michael-zte2: https://trello.com/b/ROTxmGIc/ironic-ocata-priorities
17:42:02 <TheJulia> xavierr: gabriel-bezerra: It would likely help reviewer context to know that this unbreaks your gate in the commit message :)
17:42:23 <Michael-zte2> xavierr: thank you. I want to start with the code review, is this ok?
17:42:40 <xavierr> TheJulia: indeed
17:42:46 <gabriel-bezerra> TheJulia: yes, he did. I'd just like to grab attention to it as it is important for the CI, and seems like a simple patch.
17:42:48 <TheJulia> Michael-zte2: In that case, the whiteboard is a good starting place for revisions to take a look at.
17:43:03 <gabriel-bezerra> TheJulia: thanks for the tip
17:43:22 <xavierr> Michael-zte2: sure :)
17:43:41 <Michael-zte2> TheJulia: OK. Thank you very much
17:43:55 <TheJulia> Michael-zte2: and feel free to ask any questions in #openstack-ironic, especially if you need additional context.
17:44:38 <TheJulia> jlvillal: The question is now, do the crickets return ;)
17:44:46 <Michael-zte2> TheJulia:Thank you for your very useful information
17:45:02 <jlvillal> TheJulia: If so, we get ~10 minutes back :)
17:45:10 * rloo hears crickets
17:45:13 <NobodyCam> :)
17:45:25 <TheJulia> Well, if there is nothing else, thank you everyone!
17:45:34 <NobodyCam> thank you
17:45:35 <rloo> thank you TheJulia! :)
17:45:37 <lucasagomes> ty
17:45:40 <TheJulia> #endmeeting