Include: Nothing found for "^## meetings$"!

Include: Nothing found for "^## end-meetings$"!

This page is being re-factored. Most of the old content has been moved to LAVA


Agenda

  • Actions from last meeting - plars chasing
  • Engineer progress reports - All team members
  • Blueprints
  • New team member, meeting time changes
  • Lava server split
  • Any other business (AOB)

Past Action items

  • Last weeks action items to discuss.

Action Items

  • Everyone - update blueprint whiteboard with user stories sorted by relative priority/dependency order


Engineers Reports

Spring Zhang (springz)

Highlights

Plans

  • Simple error handler
  • New blueprint and work items.

Issues & Absence

  • Enduring a network issue in company, can't connect to Internet directly.


Zygmunt Krynicki <zyga>

Highlights

  • Slow week post UDS
  • Wrapping up some of the items from previous cycle:
    • data view sandbox documentation
    • sandbox configuration
    • testing new xml-rpc calls
  • Some token auth work for XML-RPC (got stuck on bug in django-testscenarios)
  • Fixed design bug in django-testscenarios + django-testproject that got exposed with django->fork->double import (the unfortunate way django works today). Now fixed in trunk.

  • Released launch-control-tool 0.5 with bug fixes for talking to production servers and version checks, new pull user interface and more.
  • Deployed redmine backlogs for tracking my activity during 2011-05 cycle: http://fracture.suxx.pl/rb/taskboards/11 (everyone is welcome to use it, sign-in with openid and ping me to get permissions)

  • Discussed weekly plans with Michael Hudson

Plans

  • Wrap up 04 cycle
  • XML-RPC auth support
  • Packaging cleanup after lava project revamp land
  • Check the backlog for details (http://fracture.suxx.pl/rb/taskboards/11)

  • Possible swap day Friday <=> Saturday (I'll be working on Saturday)

Issues

  • None


Mirsad Vojnikovic <mirsad>

Highlights

Plans

  • Discuss and finalize scheduler blueprints
  • Finish basic daemon blueprint

Issues

  • None


Dave Pigott <davepigott>

Highlights

  • Sourcing and ordering power cables for Snowballs and new Beagleboards
  • Sorting out details of how to get serial line to Snowballs
  • Building images for the Beagleboards
  • Trying to find out where to get an image for the Snowballs
  • Finding and fixing the IRC connection problem from the Linaro office (change made by james_w that opened port 80)
  • Upgrading Cyclades firmware and resolving problem with web access - zyga suggestion to spoof user agent string worked!
  • Updated Val Lab hardware Blueprint from UDS

Plans

  • Order and solder on HE10 serial headers to Snowboards
  • Deploy Snowballs and Beagleboards into Validation lab
  • Investigate why PIDs on Cyclades are continually increasing (regular reboot of Cyclades and conmux stop/start suggested)
  • Pick up some coding work!!!
  • Build Noritsuna's face recognition robot

Issues

  • Can't order HE10 headers without spec from ST-E - awaiting


Paul Larson <plars>

Highlights

  • Looking at all the blueprints we have, starting to think about priorities
  • Looking at tools that may be useful for helping us track our work this cycle
  • A few code reviews
  • Submitted a small change to setup for lava
  • Working to resolve hardware issues with some of the boards we have in the lab

Plans

  • Prepare for public plan review
  • Look at how the android support branch is shaping up, do some testing with it
  • merge cache support in lava
  • Get tests going on more image types, and using more tests for daily runs

Issues

  • Still some nagging issues with a few of the boards we have in the lab, including new ones


IRC logs

Meeting opened by plars at 13:05

https://wiki.linaro.org/Platform/Validation/Meetings/2011-05-19

  • <plars> [TOPIC] Actions from last meeting

Actions from last meeting

  • <plars> don't think we had any, especially since last week was LDS :)

    <davepigott> :)

    <plars> [TOPIC] Engineer progress reports

Engineer progress reports

  • <plars> I suspect most of our status for the last week is about the same, LDS, working on blueprints for the most part

    <plars> but does anyone have anything to highlight?

    <davepigott> Updated the wiki report.

    <springz> Serial log attaching mp should be ok with new commits when I change attachment format to v1.2

    <davepigott> Mostly it's been trying to get Snowball issues, IRC issues and Cyclades issues resolved

    <zyga> plars, I think it's worth pointing out that lava revamp will happen soon

    <plars> zyga: it's on the agenda

    <zyga> plars, and that mwhudson will touch all our projects to make it happen

    <zyga> ok

    <springz> plars, the rename

    <plars> see link above ^ :)

    <fabo> plars: your report is missing. please, could you update the wiki too.

    <plars> so let's get to that then

    <zyga> davepigott, did the user agent trick "fix" cyclades?

    <davepigott> zyga: Well, it means I can now use ff 4.0, so yes. :)

    <plars> zyga: we have a ticket open with them

    <davepigott> It didn't fix cyclades, just allowed access after the upgrade to the firmware

    <davepigott> plars: I got feedback from them. Have upgraded the firmware

    <plars> the browser thing is just a minor thing though

    <plars> oh, do tell

    <davepigott> plars: They know less than we do. :)

    <plars> lovely

    <plars> that's not encouraging :(

    <davepigott> Sent back instructions to them on how to do an ftp upgrade. No it's not.

    <zyga> davepigott, ??! are they just re-selling the thing?

    <plars> davepigott: so their recommendation was just "upgrade the firmware and hopefully it will work"?

    <davepigott> They opened the ticket and closed it without contacting me. I jumped all over them

    <plars> !

    <davepigott> Their instructions on how to upgrade were all but useless. Had to reverse engineer it myself.

    <davepigott> But it's all done

    <davepigott> And up and running.

    <plars> wow, considering how long these guys have been around and being used for exactly this purpose, I'm a bit surprised by this

    <davepigott> Then zyga came up with the spoofing for firefox to fix the browser access problem.

  • davepigott hands bottle of virtual champagne to zyga. :)

  • zyga takes a virtual sip :-)

    <plars> *sigh* ok, we will certainly re-examine whether there are other good alternatives to cyclades next time around as well

  • zyga needs to check if 'sip' is a word...

    <zyga> anyway

    <davepigott> sip is a word. :)

    <plars> ok, let's move on then

    <plars> [TOPIC] Blueprints

Blueprints

  • <plars> I think I've talked to most of you about this already

    <plars> but just in case

    <plars> If you ran a session last week and/or are listed as a drafter on a blueprint, you need to take a look at it this week and update it based on LDS discussion

    <plars> For now, update the whiteboard on the blueprint with user stories, and actions that came out of the discussion

    <davepigott> Done

    <plars> as much as possible, try to arrange them by what you feel the priority should be, obviously putting dependencies higher on the list

    <plars> davepigott: I saw the update, will take a look later, thanks!

    <plars> there is some ongoing discussion about what the next step is, and any tools we might be adopting to help this cycle

    <plars> so I'll get back to you on that, but this is an important first step towards understanding what work we have for the blueprints

    <plars> and helping us break it up

    <springz> plars, you mean within team or linaro

    <plars> springz: within linaro

    <plars> [TOPIC] New team member, meeting time changes

New team member, meeting time changes

  • <plars> not new so much...

    <plars> but as most of you know, mwhudson has been helping us out for a few weeks on Lava

    <plars> He's going to be staying on longer and continuing to help with the project

    <plars> the problem for that, now becomes time zones

    <plars> because he is in NZ, it's some ridiculous hour right now for him, 1 or 2 AM

    <zyga> plars, we're all moving to NZ now, right? ;-)

    <plars> zyga: you want to move again already?

    <plars> :)

    <plars> no

    <zyga> haha :-)

    <davepigott> Yay!

    <plars> So my proposal is that we alternate weekly meeting times

    <plars> we'll keep this time every other week, for those of us who have time zones that align better with this time

    <plars> 13:00 UTC

    <plars> and every other week, have one at 01:00 UTC

    <plars> so michael, and maybe springz might find that one more useful

    <zyga> so next week we meet at 1:AM UTC?

    <plars> you are, of course, welcome to attend both if it is convenient for you

    <plars> if not, catch up on the meeting minutes afterward

    <plars> I'll be attending both of course

    <plars> zyga: yes

    <davepigott> I have matchsticks to prop my eyes open. :)

    <springz> I may remain on 13:00 UTC in home, and 1:00UTC I may be on my way to company

    <plars> does anyone object to that? or have a better idea?

    <davepigott> Not really. Makes sense

    <davepigott> Alternately, we set a universal time for the world with no time zones and just live that way. :)

    <plars> ok, so let's try that for a month and see if it works

    <plars> davepigott: I've long been in favor of that

    <plars> or something similar

    <davepigott> Would help with jet lag. :)

    <plars> [ACTION] plars to update meeting schedule

plars to update meeting schedule

  • <plars> [TOPIC] Lava server split

Lava server split

  • <zyga> [ACTIÓN] plars to abolish time zones in favor of Linaro Time

    <zyga> ;-)

    <plars> not a whole lot to discuss here, Michael's note was pretty clear I think

    <plars> zyga: if only I had that kind of power

    <plars> he's holding off slightly to wait for an update in launchpad that will help a lot

    <plars> so that stacked branches are tied to the branch id, not the branch name

    <zyga> oh, right

    <zyga> that will also resolve a lot of grief we had when changing branch owners

    <plars> but he says it should be a really quick update, though I suspect we need to do a little work after the fact to document how to set up a development/test environment, and things like that

    <plars> zyga: yep

    <zyga> right, I have an action to repackage everything as soo as he's done

    <zyga> it will take some more time as it's also code-facing changes

    <zyga> launch-control -> lava-server + dashboard split

    <zyga> ~> energy ;-)

    <plars> zyga: are you planning to do a rename of launch-control after this also?

    <zyga> yes

    <zyga> l-c will be no more

    <zyga> actually the lp.net thing is only about renaming

    <zyga> the hard work comes afterwards

    <plars> yes

    <zyga> I'm sure I'll be busy with this next week

    <zyga> we may also investigate namespace packages

    <zyga> (as in lava.dashboard, lava.scheduler and so on)

    <zyga> (at python level)

    <plars> zyga: I expected that would be the result of all this, right?

    <zyga> plars, not directly

    <zyga> plars, it's a secondary optional step

    <zyga> plars, the direct result would be that launchpad projects would be named lava-$foo

    <plars> it seems logical

    <zyga> plars, the indirect result is that we can clean up debian packages and python packages (stuff with init.py)

    <davepigott> easier to grok

    <plars> right

    <zyga> plars, and while we'll do the former the latter is still optional

    <zyga> plars, I'd rather start with lava_dashboard and then do namespacing

    <zyga> as namespace packages _are_ tricky

    <plars> zyga: is it more tricky because of the conversion in the case of the dashboard, and you expect something new like the scheduler will be easier?

    <zyga> no, it's tricky in general because it's rare and few people know how it actually works

    <zyga> then you need to package that at python level

    <zyga> and at debian level

    <zyga> it's not like java org.linaro.lava.foo

    <zyga> (unfortunately on the complexity side)

    <springz> zyga, do we need to do it in our existing code both in l-c and lava?

    <zyga> AFAIR only zope does it

    <zyga> springz, in everything

    <zyga> springz, each piece of code will be changed

    <zyga> springz, both at packaging and code level, virtually every import of in each file

    <plars> updating the imports is trivial though

    <zyga> springz, it will be a flag day

    <zyga> plars, yeah

    <zyga> plars, next week I'll tell you what I did not know and what we missed ;-)

    <plars> the setup/packaging is a bit more

    <plars> zyga: heh, sounds good

    <plars> hmm, forgot to put this in earlier

    <plars> [ACTION] everyone to update blueprints with user stories and actions

everyone to update blueprints with user stories and actions

  • <plars> ok, anything else to discuss on the lava server split?

    <zyga> nope

    <zyga> just don't expect merge proposals on that one

    <plars> zyga: heh, yeah those would be ugly

    <plars> [TOPIC] Any other business (AOB)

Any other business (AOB)

  • zyga can now advertise redmine again

    <zyga> I installed an "agile workflow" webapp to keep track of my TODO list

    <plars> zyga: yep, it has some interesting features due to recent changes, and we have more freedom to explore such things now

    <zyga> I encourage anyone to use it, AFAIR linaro in general will have something like that (not based on redmine, some commercial SAAS)

    <davepigott> Just heard from ST-E that we may not be able to just add a serial header to the Snowballs. Awaiting further news.

    <fabo> what about the BPs for 11.05?

    <zyga> from my discussion wiy joey it's about +1 +2 months away

    <zyga> if anyone is interested in my deployment have a look at http://fracture.suxx.pl/rb/taskboards/11 and ping me after the meeting

    <springz> davepigott, serial header? what's the relation with serial port

    <plars> fabo: a few of those are in the process of getting wrapped up, the remaining ones will be postponed

    <davepigott> springz: Same thing. Just the component

    <zyga> plars, we should also mark the blueprints as implemented

    <fabo> plars: ok

    <springz> davepigott, but there is serial port in Snowball, isn't it

    <plars> zyga: yes, for the ones that are satisfied

    <davepigott> springz: Only through USB on the ones we got

    <plars> springz: no

    <springz> zyga, great

    <fabo> wrt validation for Android platform, what's plan?

    <springz> davepigott, oh, unfortunately

    <plars> fabo: android team is working on that. They have some code they have sent us for it, but talking to them last week they have some updates to do to it before they consider it ready. Waiting on a resubmit right now

    <davepigott> plars: I have to leave for a hospital appointment - can't be late

    <zyga> plars, I would like to volunteer on panda + lava + android hacking

    <plars> davepigott: go, see you later!

    <fabo> plars: so you're waiting for them, right?

    <davepigott> Bye all!

    <plars> fabo: right

    <plars> fabo: was going to ping jeremy to see where he is at with it today

    <fabo> plars: last question :) afair, you mentionned some issues with ST-E hardware. it it resolved?

    <plars> zyga: oh right you have a panda now right?

    <plars> zyga: unfortunately panda still has problems with android I think

    <plars> fabo: nope

    <zyga> plars, bugger... what about beagle (non XM)

    <plars> fabo: that's what davepigott was talking about earlier

    <plars> fabo: also with samsung

    <plars> freescale...

    <fabo> plars: any help needed to move forward?

    <plars> as an AOB item, I did get a proposed solution from Angus about the samsung boards

    <zyga> plars, samsung may do something helpful, I talked with the liaison for the landing team during Friday party and he was extremely interested in getting LAVA working

    <zyga> plars, we should talk about that later

    <plars> the email started with something like "This has not been approved by samsung, and might damage the component..."

    <plars> zyga: if you have a name different from the ones I've been talking to, that might help

    <plars> fabo: I can start cc'ing you on the emails if you want

    <zyga> plars, bpark

    <fabo> plars: yes, then I can take actions if you're stuck

    <plars> fabo: mostly it's just been pinging them every week to see if they have anything yet

    <plars> all of them have an action to work internally on a solution and get back to us

    <plars> fabo: if you'd like, you and I can talk later and I can catch you up on where they are all at

    <fabo> ok. thanks.

    <plars> great, a barrage of AOB items!

    <fabo> :)

    <plars> any more?

    <zyga> :-)

    <zyga> let's get back to hacking

    <plars> ok, thanks everyone. I'll send out updated meeting invites shortly

    <plars> #endmeeting


Meeting closed at 13:44

People Present

  • plars
  • davepigott
  • springz
  • zyga
  • fabo

Actions Recorded

  • plars to update meeting schedule
  • everyone to update blueprints with user stories and actions

internal/archive/Platform/LAVA/Meetings/2011-05-19 (last modified 2013-08-23 13:18:52)