Validation Team Meeting

Agenda

  • Actions from last meeting - plars chasing
  • Engineer progress reports - All team members
  • Work Item progress and team issues discussion - plars and JamieBennett questions to the team

  • Any other business (AOB)

Past Action items

  • JamieBennett to see if freescale can get us a reworked imx53

  • plars or springz to forward imx53 rework info to JamieBennett

People Present

Actions Recorded

  • chithu to discuss scheduler models with plars and mirsad
  • chithu to discuss scheduler models with team


Engineers Reports

Dave Pigott <davepigott>

Highlights

  • Got Orion running and available in the validation lab
  • Continued to act as local sys-admin
  • Putting plan together for reconfig of network
  • General organisation of validation lab

Plans

  • Publish network plan and implement
  • Deploy more boards
  • Start looking at scheduler UI

Issues

  • Orion board doesn't power up when power applied


Spring Zhang (springz)

Highlights

  • Make a quick mp for reducing duplicated code in cmd_install_abrek in abrek.py: lp:~qzhang/lava/refactor-install-abrek.
  • Working on lp:~qzhang/lava/submit-results: get the right scenario and work on serial log attachment, log error detection.
  • Discussion on lava refactor and some future direction.

Plans

  • Finish branch submit_results.
  • Enhance dispatcher blueprint and spec.
  • Beta RC test.

Absence

  • Tomb-sweeping day from Apr.3 to Apr.5, and work on Apr.2 as a substitute


Le Chi Thu (ChiThu)

Highlights

  • Studying the django web framework.
  • Studying the Launch Control source code.
  • Setup the development environment. Ubuntu, Bazaar, etc...

Plans

  • Start to implement the front-end of scheduler

Issues

  • NA


Paul Larson <plars>

Highlights

  • Submitted mp to make some changes on the dispatcher to help the scheduler deal with it easier
  • Submitted proposal for how to handle async calling of multiple dispatcher threads by the daemon
  • code reviews
  • Started thread with ideas about UI in the scheduler
  • Talked to several others about some ideas for changes to Lava
  • Attended off-site meeting with a member
  • Meetings with new members to help get them started

Plans

  • Focus more on the scheduler
  • Continue to think about how we can refactor the code to make it more flexible
  • Work on slides for ELC

Issues

  • Both the orion board, and the imx53 loco board will not automatically power on


Mirsad Vojnikovic <mirsad>

Highlights

  • Proposed an initial version of scheduler daemon for merge, but it lacks some important things, so it will be dropped for now
  • Working on daemon implementation, exchanged some ideas with Paul, discussing implementation details and evaluating different solutions
  • Participated in discussion about scheduler UI, started by Paul
  • Minor discussions about code refactoring within Lava repository
  • Minor mentoring for Chi Thu

Plans

  • Produce initial daemon implementation
  • Update WI list for scheduler BP
  • Discuss various scheduler parts

Issues

  • Slow progress for my internal ticket about bzr access within STE.


Your Name <irc nick>

Highlights

  • Short bullet points of work you've done that week which convey progress and highlights which can be used to report on how the team is progressing as a whole.

Plans

  • Your individual plans for the coming week(s).

Issues

  • Things that might affect/are affecting your ongoing work


IRC logs

Started logging meeting in #linaro-meeting
[13:06:59] <plars> https://wiki.linaro.org/Platform/Validation/Meetings/2011-03-31
[13:07:09] <plars> [LINK] https://wiki.linaro.org/Platform/Validation/Meetings/2011-03-31
[13:07:28] <plars> [TOPIC] Actions from last meeting
[13:07:42] <plars> JamieBennett to see if freescale can get us a reworked imx53
[13:07:59] <JamieBennett> Still waiting on a reply
[13:08:01] <plars> I think this is still in progress right?
[13:08:02] <plars> yeah
[13:08:11] <plars> I also talked to Yves directly about it on Monday
[13:08:31] <plars> and now we have the same problem with the orion boards as well
[13:08:35] <JamieBennett> right
[13:08:36] <plars> which is also still in progress
[13:08:39] <JamieBennett> right
[13:08:48] <davepigott> plars: Who's following up on the orion?
[13:08:49] <JamieBennett> what did Yves have to say on this?
[13:08:57] <JamieBennett> davepigott: I'm chasing
[13:09:03] <davepigott> thank
[13:09:05] <davepigott> *thanks
[13:09:07] <plars> JamieBennett: just that he would see what he could do
[13:09:13] <JamieBennett> OK
[13:09:33] <plars> I really don't think we should take on doing board rework
[13:09:41] <davepigott> plars: Agreed
[13:09:42] <plars> not ourselves
[13:09:43] * JamieBennett agrees
[13:09:55] <plars> so hopefully they will have a good solution
[13:10:17] <plars> the only other AR was for me to forward JamieBennett the info I had on it, which is done
[13:10:34] <plars> [TOPIC] Engineer progress reports
[13:10:57] <plars> zyga is still on vacation, should be back tomorrow iirc?
[13:11:03] <plars> learning spanish hopefully :)
[13:11:13] <davepigott> :)
[13:11:30] <plars> For me, I submitted a few merge proposals for lava, one still in flight
[13:11:33] <JamieBennett> yes, zyga is due back tomorrow
[13:11:50] <plars> submitted some example code for discussion for the daemon async job handler portion
[13:12:25] <plars> kicked off the UI thread for scheduler with some ideas, had a lot of discussions about possible changes/tweaks in how lava works
[13:12:35] <plars> those are the main things
[13:12:53] <plars> mirsad: anything to highlight?
[13:13:10] <mirsad> well, basically I'm working on daemon part
[13:13:43] <mirsad> pushed a proposal on Monday, but it's not good, so it should be dropped
[13:14:07] <mirsad> looking at your proposal now, have like 10 different ideas to try
[13:14:27] <plars> mirsad: oh good, so it helped?
[13:14:59] <mirsad> well, at some parts I'm confused, but we can take it outside this conversation
[13:15:12] <plars> let's narrow it down quickly then... find something that works well, and move on. We may have some time to experiment more later
[13:15:13] <plars> ok
[13:15:38] <plars> anything else?
[13:16:01] <mirsad> various discussions etc... about scheduler, nothing else
[13:16:07] <plars> thanks
[13:16:19] <plars> ChiThu: anything to highlight this week?
[13:16:44] <ChiThu> yes,
[13:17:15] <ChiThu> I study the Django web framework and the Launch Control code
[13:17:41] <ChiThu> Discussed with plars around with tasks I can start with.
[13:18:20] <ChiThu> It seems like I can help out with the front end part (UI) of the scheduler.
[13:18:36] <ChiThu> right plars ?
[13:18:42] <plars> ChiThu: yes
[13:19:15] <plars> ChiThu: do you feel like you have what you need to get started? You seemed like you had some good ideas about the UI for the scheduler in the discussion over email
[13:19:56] <ChiThu> I would like to discuss more about the models for Scheduler (TestJob, TestSuite...) offline.
[13:20:45] <plars> [action] chithu to discuss scheduler models with plars and mirsad
[13:20:54] <plars> mirsad should be part of that as well
[13:21:02] <plars> ok, anything else?
[13:21:06] * davepigott would like to be included
[13:21:08] <ChiThu> and there are many code from Launch Control can be shared with scheduler.
[13:21:30] <plars> [action] chithu to discuss scheduler models with team
[13:21:44] <ChiThu> Should I copy & past the code from Launch control ? or ?
[13:22:10] <plars> ChiThu: that's an option, but also you should consider that zyga broke out a lot of those reusable pieces into libraries
[13:22:18] <plars> ChiThu: so if it's one of those, just use it
[13:23:00] <ChiThu> authentication, test result model, device..., look and feel...
[13:23:43] <plars> thematically, I think it's a good fit, but UI interaction I think will be quite different
[13:24:01] <plars> test result model is irrelevant, since we don't store results in the scheduler
[13:24:21] <plars> anything else?
[13:24:34] <ChiThu> ok. I can start with copy the files first and we can refactory out the common parts later.
[13:24:41] <ChiThu> nothing more.
[13:24:48] <plars> ok
[13:24:54] <plars> davepigott
[13:25:14] <davepigott> Highlight of the week was getting the Orion up and connected
[13:25:34] <plars> \o/
[13:25:40] <plars> mostly :)
[13:25:46] <davepigott> Also installed VirtualBox/Ubuntu on my Mac and used that for the build and deployment
[13:26:01] <davepigott> Nice to have such a clean integrated environment
[13:26:23] <davepigott> Yes. As you noted, problem with board not booting up on power up
[13:26:57] <davepigott> Still awaiting cable order. Supplier has apologised.
[13:27:09] <plars> so what do we have for now?
[13:27:22] <davepigott> Direct to mains power, not through PDU
[13:27:27] <davepigott> That's the only thing missing
[13:28:06] <plars> that's sensible for now, especially since the pdu would just power it off
[13:28:08] <davepigott> Got a huge cable order in that we can refresh at any time and should mean that as a board comes in we can pretty much install it
[13:28:21] <plars> excellent
[13:28:51] <plars> and on the development side? have you had much opportunity yet to look into the UI stuff?
[13:29:05] <davepigott> One slight problem with the Mac/Ubuntu is that I have problems forwarding the flash drive. Working on that
[13:29:16] <JamieBennett> forwarding?
[13:29:41] <plars> JamieBennett: ought to be able to help you there, I think he has a similar setup
[13:29:45] <davepigott> Yes. Need to talk to the flash drive from Ubuntu as a raw device. Mac doesn't want to let go of it
[13:30:02] <davepigott> Or VBox doesn't want to grab it.
[13:30:12] <JamieBennett> ah, I use vmware
[13:30:13] <plars> JamieBennett: or are you using parallels?
[13:30:14] <plars> oh
[13:30:17] <plars> vmware even :)
[13:30:35] <davepigott> No biggy. I can dd it from the Mac. Just would be nice to do it all in one place so I can script it
[13:30:51] <JamieBennett> right, vmware is the way to go then, no problems here
[13:31:00] <davepigott> Noted
[13:31:28] <davepigott> Also started looking at Launch Control. Will need a little guidance (offline).
[13:31:57] <plars> davepigott: you'll want to discuss with zyga, I suspect he'll be quite interested in helping you get started :)
[13:32:07] <davepigott> Excellent. :)
[13:32:20] <plars> he's been really hoping for someone else to help out there, so this will be welcome news to him
[13:32:43] <plars> great
[13:32:45] <plars> anything else?
[13:33:01] <davepigott> One last one: Putting plan in place and started executing for Validation lab. Includes the network LAN separation here
[13:33:07] <plars> ah yes
[13:33:13] <plars> how's that coming?
[13:33:28] <davepigott> Will mean some Val Lab downtime. Planning to do it late next week
[13:33:37] <davepigott> Will give pleanty of warning. Promise
[13:33:37] <plars> ok
[13:33:44] <davepigott> At least 5 minutes
[13:33:46] <davepigott> ;)
[13:33:50] <plars> that's plenty
[13:33:57] <plars> are you going to do the router swap at the same time?
[13:34:03] <davepigott> Yep. All in one
[13:34:04] <plars> would be nice if we could do that ASAP
[13:34:09] <JamieBennett> \o/
[13:34:17] <plars> the router swap will help everyone, including the office staff
[13:34:23] <davepigott> Yep
[13:34:31] <plars> and I'm concerned that it already seems to be lagging on handing out ip addresses, dns, the usual
[13:34:33] <davepigott> That BT hub is a pile of ----
[13:35:13] <plars> davepigott: would it make sense to go ahead and do the router swap now?
[13:35:18] <plars> do you have the new router already?
[13:35:42] <plars> (now as in: tomorrow, or early next week... not right this second :)
[13:35:47] <davepigott> Nope. DaveR and I agreed on one and will be here next week I hope
[13:35:51] <plars> ok
[13:36:33] <plars> thanks davepigott
[13:36:47] <plars> [TOPIC] Work Item progress and team issues discussion
[13:36:50] <plars> http://status.linaro.org/linaro-validation.html
[13:36:55] <plars> hmm
[13:37:01] <plars> is it me or is that not updating?
[13:37:09] <JamieBennett> umm
[13:37:15] <JamieBennett> I'll investigate
[13:37:17] <plars> looks a few days behind or something
[13:37:35] <plars> my gut feeling is that I haven't seen a whole lot of +DONE in email
[13:37:40] <JamieBennett> Its still shows us behind though
[13:37:51] <plars> so please take a look at work items and see if we can cross some more off
[13:38:11] <plars> in fact, I see one I can :)
[13:38:19] <JamieBennett> :)
[13:38:34] <plars> but yes, we are beind, particularly on the scheduler
[13:38:48] <plars> which is why ChiThu has been directed towards that
[13:39:10] <plars> and we may get some additional support from another team in the next few weeks on that as well
[13:39:22] <plars> that's the current game plan, and what I hope will help us bootstrap it quickly
[13:39:34] * JamieBennett nods
[13:39:36] <mirsad> plars, great news!
[13:39:49] <ChiThu> I was pushed there :-)
[13:40:18] <plars> ChiThu: it seems a good fit for you too, as you have had experience doing this before on similar systems
[13:41:04] <plars> final question for this topic though...
[13:41:08] <plars> is anyone blocked on anything?
[13:42:16] <plars> ok, guess that's a no
[13:42:29] <davepigott> plars: Can't deploy more boards until cables come in, but that's it
[13:42:36] <plars> but just because I ask here, don't think that means you have to hold it for the meeting if you have something blocking
[13:42:42] <plars> please raise it immediately
[13:42:51] <plars> [TOPIC] Any other business (AOB)
[13:43:15] <JamieBennett> I have a question for mirsad
[13:43:22] <mirsad> JamieBennett, yes?
[13:43:28] <JamieBennett> mirsad: did you get any feedback on the proxy issue you are having?
[13:43:44] <JamieBennett> with bzr
[13:44:04] <mirsad> JamieBennett, the ticket is raised and some minor progress is that the laptop I'm using cannot be identified in the system !?
[13:44:30] <mirsad> so they will identify why it is like that and import it
[13:44:34] <mirsad> as first step
[13:44:40] <JamieBennett> OK, let me know how this goes (cc me if appropriate)
[13:44:40] <mirsad> then they will look at bzr issue
[13:44:46] <mirsad> yes, I will
[13:44:49] <JamieBennett> thanks
[13:44:51] <plars> this isn't limited to just you though
[13:44:58] <plars> this is still a problem for everyone in STE right?
[13:45:03] <mirsad> exactly
[13:45:06] <plars> is there some kind of global ticket open on it?
[13:45:17] <ChiThu> what bzr issue ?
[13:45:26] <mirsad> we were advised to raise a ticket per person
[13:46:08] <mirsad> ChiThu, bzr issue is that you cannot use it within STE proxy
[13:46:14] <plars> mirsad: has anyone else been able to make progress? are they allowing it just for people who open a ticket?
[13:46:40] <plars> either way, ChiThu should probably go ahead and open one now as well, so that he can go ahead and get the process started for him
[13:46:52] <mirsad> plars, no input on that, the people seem not to bother filing tickets, they use workarounds
[13:47:19] <plars> mirsad: do we have a good workaround, other than working from home?
[13:47:41] <ChiThu> The work around is to connect to a mobile network when work with Linaro.
[13:47:57] <mirsad> yes, some ppl are using 3G NW
[13:48:24] <plars> heh
[13:48:31] <plars> ok
[13:48:49] <plars> working COMPLETELY around it
[13:48:58] <plars> ok, anything else?
[13:49:43] <JamieBennett> not from me
[13:49:48] <davepigott> nor me
[13:49:50] <plars> alright have a good morning/day/evening everyone
[13:49:53] <plars> #endmeeting
Meeting ended.

internal/archive/Platform/LAVA/Meetings/2011-03-31 (last modified 2013-08-23 13:19:52)