Firefox/Channels/Meetings/2013-04-25
From MozillaWiki
Channel Meeting Details
- Tuesdays at 10am PDT, and Thursday at 2:00pm PDT
- Mountain View Offices: Warp Core Conference Room
- San Francisco Office: SFO-7I Independent (Tues), Noise Pop (Thurs)
- irc.mozilla.org #planning for backchannel
Video/Teleconference Details - NEW
- 650-903-0800 or 650-215-1282 x92 Conf# 99951 (US/INTL)
- 1-800-707-2533 (pin 369) Conf# 99951 (US)
- Vidyo Room: Release Coordination
- Vidyo Guest URL
REMEMBER
These notes are read by people who weren't able to attend the meeting. Please make sure to include links and context so they can be understood.
Contents
Previous Actions
Schedule Update
- FF21.0b4 for Desktop/Mobile shipped
Add-ons
- Nothing new to report.
Stability
Aurora
- https://bugzilla.mozilla.org/show_bug.cgi?id=856670 (js::RemapWrapper) remains the main concern, a possible bandaid fix is up now.
- KaiRo commented in the dev-planning "Using a pre-processing flag to auto-disable features in later Beta versions" thread how the low ADI esp. on Aurora but also on Beta lead us to make compromises and end up testing on higher channels than we'd like to.
Beta
- https://bugzilla.mozilla.org/show_bug.cgi?id=865701 is another Radeon crasher in b4.
- https://bugzilla.mozilla.org/show_bug.cgi?id=836263 (Crossrider) is decreasing as they switched their storages. We still should push for a fix on our side in 22 and/or 23.
- email :kyle
- https://bugzilla.mozilla.org/show_bug.cgi?id=851934 (Outlook.com) is #4 on b3, we have full dumps but need investigation on those, no fix in sight for now.
- https://bugzilla.mozilla.org/show_bug.cgi?id=852915 (mozilla::MediaDecoderReader::DecodeToFirstAudioData) should be fixed in b4.
- Overall crash rates for beta are good, we seem to look better than 20 Beta, probably due to the Flash fix.
Release
- https://bugzilla.mozilla.org/show_bug.cgi?id=838568 "the Brazilian Bank crash" is now rising in 20.
- email need-contact ?
- The previously reported (somewhat) lower level of D3D11 and empty crashes remains constant, as does the low overall crash rate.
Mobile
- A number of stagefright blocklist patches landed for 21.0b4, one or two additional ones might be needed to improve even further.
QA
QAwanted
- https://bugzilla.mozilla.org/show_bug.cgi?id=812573 - Fx 10-16 update hotfix
- Testing is complete and signed off, hotfix pending signing
- https://bugzilla.mozilla.org/show_bug.cgi?id=857199 - en-US Win32 beta stub installer
- Appears to be pending web-dev fixes, can we get a status update/timeline to test?
- akeybl will check up on the status of this bug
- https://bugzilla.mozilla.org/show_bug.cgi?id=863714 & https://bugzilla.mozilla.org/show_bug.cgi?id=865701 (Radeon 6310)
- Juan has a netbook with this card but has not been able to reproduce
Verification Needed
Aurora
Beta
Mobile
- 856445 – Startup crash on LG Optimus Black (LG P970)
RelEng
Nothing today
Marketing/Communications
Support
We are watching closely for feedback around Bug 865701 after Beta 4 releases, hopefully will get auser willing to work with us if needed
Thunderbird (Tue)
Metrics (Thu)
Roundtable
a) Since the data that we have is very early data with practically no users we push Fx 21 Beta 4 today as planned. b) Spin another beta with the same changeset and keep an eye on the crash-volume and do an early push if we are sure it helped by Monday ? c) Spin another beta and reconsider adding a patch with padding in this particular scenario to see if it helps ? d) Any other ideas that may be helpful here to mitigate the problem ?
- https://bugzilla.mozilla.org/show_bug.cgi?id=863714 & https://bugzilla.mozilla.org/show_bug.cgi?id=865701 (Radeon 6310)
- We should already build a beta for Desktop with the same changeset and put it on ftp
- Question here for :bsmedberg : making sure taking the same chageset is ok here in the hope of avoiding this crasher or would it matter if we took a newer changeset ?
- Any other speculative fixes(padding) that we could land for this particular top-crasher ?
- If the crash volume spikes unexpectedly in Beta 4 as early data suggests
- Check the early crash-data corresponding to the new beta and if it helped push it out early or verify it with STR if we were able to find one
- We should already build a beta for Desktop with the same changeset and put it on ftp
- For final Fx21RC build :
- Like we did in FX20, we should go-to-build right after our final beta & put it on ftp & rely on the crash-data to see any unusual spike this build may have for the Radeon-crasher
- If so, immediately spin a new RC with same changeset ...
- (next actions) :benjamin will have more data shortly which may help QA to assemble machines with those configuration and try reproduce
- try to identify particular OEM builds that could help QA
- device a plan for the RC build, mostly comparing pre crash data from ftp to the new build.
- this will involve comparing the data for every beta for the time its been on ftp vs post release to see if we can co-relate and increase confidence
Actions