[yocto] Minutes: Yocto Project Technical Team Meeting - Tuesday, March 26, 2013 8:00 AM-9:00 AM (UTC-08:00) Pacific Time (US & Canada).

[Liu, Song]

Attendees:
MichaelH, Sean, KevinS, Corneliu, TomZ, BruceA, DaveST, JeffP, MarkH, Belen, Paul, Nitin, ScottR, Björn Stenberg, RP, CristianI, Ioana, AlexG, FahadU, SongL  
 
 
Agenda:
 
* Opens collection - 5 min (Song)
* 1.3.1:
  . We had a build and passed QA full pass testing. Got approval from CCB to release. But the build was automatically removed before we reached the decision to release. Beth recovered part of the build and is trying to regenerate the missing part. Expecting it to be released soon.
  . MichaelH: 14 days, how long we keep the nightly. We changed this and will move release build out of the auto removal directory. This won't happen again.
  . ScottR: Will the release be in April? If so I need to change the doc. Song: hope we can make March. But we may need more QA and it may slip into April.
* Yocto 1.4 status - 10 min (Song/team)
  https://wiki.yoctoproject.org/wiki/Yocto_Project_v1.4_Status
    . 3/17 feature complete. Two medium+ features left.  Small BSP to support as many hardware as possible:  (RP)Still formulating plans, will not make 1.4.
    . Bug count is up 298 as last Sunday. Expected for various reasons including more QA testing, etc. The team should focus on bug fixing in the remaining weeks of 1.4.
    . 1.4 M5 RC1 is incomplete, weekly test is done. regenerating RC2 this week. QA will run full pass testing.
    . Schedule: https://wiki.yoctoproject.org/wiki/Yocto_1.4_Schedule#M6_Stabilization_Milestone_.283.2F18.2F2013_-_4.2F26.2F2013.29
    . RP/1.4 AB infrastructure change: there has been changes upstream, we worked on the new AR. We switched to the new infrastructure. Over the past 2 weeks, we are trying to get the build succeed. Has been problematic. Lots of issues resolved. There are still some random sanity failures. Occasional segfault for some images. Using SMART command against the database. Looks like database corruption. If you download and do it locally, it succeeds. It reproduced locally once. It might be the new version of qemu. If anyone can help or has seen this, would like to have a discussion. We have been doing a lot work to make the AB switch successful.
 
* Yocto 1.5 Planning - 5 min (RP)
  - Mid way through 1.5 planning. If you would like to have 1.5 features, please put them in Bugzilla as soon as possible
  - It would be great if you can get them in by the end of this month for 1.5 planning to pick them up.
 
* SWAT team rotation: Cristiana Voicu -> Radu Moisan (Song to send an email to Radu)
 
* Opens - 10 min 
  - MichaelH: infrastructure downtime.
    Working on getting the power upgraded. Finally have a date for that. This Friday (3/29). An hour down time. It's probably around noon to 1pm PST. Already coordinated with several people in YP team, they are ok. Once we are done we will have additional room for more AB machines.
* Team Sharing - 20 min
  - AlexD: still trying to debug problem (3828?). Vmware driver problem? Why the patch change does not go down from driver to qemu.
  - Corneliu: RO QA. A new report we made to replace the old one: https://wiki.yoctoproject.org/wiki/1.4_QA_Status
  - Paul: working on bug fixing, reducing bug count. Build history. Report content of the SDK. Sent a patch to change the way we recording source rev information. You can actually print out the actual revision, etc. 
  - RP: separated source and build directory, now enabled by default in poky. 
  - Michael: working on getting the layer index app out. Finally got something for Paul last night. Changing the backup script to keep data as long as the space is available. Package reporting system has a new version, ready to be deployed in production.