Firefox/Channels/Meetings/2020-09-08
From MozillaWiki
Channel Meeting Details
- Tuesdays at 10am and Thursdays at 8am Pacific Time
- Riot #release-drivers for backchannel
- Past meeting notes: https://wiki.mozilla.org/Firefox/Channels/Meetings
Video/Teleconference Details
- Zoom Meeting ID: 536 097 768
- https://mozilla.zoom.us/j/536097768
- +17207072699,,536097768# US
- Find your local number: https://zoom.us/u/agItTmqIu
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
Attendees
TaniaM, jcristau, RyanVM, philipp, Aryx, kbrosnan, JR
Schedule Update
- Last week of Beta 81 before RC week, early Beta ended with b7
- Fenix 80.1.2 shipped @ 99% last week, bumped to 100% today
- Fenix 81.1.0-beta.2 live @ 100%, aiming to get at least one more Beta shipped this cycle
Desktop Stability
Nightly
mc2 crash rate: 1.73, crash incidence: 2.07%, mc1 summary score: 5.44 last week mc2 crash rate: 1.50, crash incidence: 2.71%, mc1 summary score: 6.04
- Crash in [@ memcpy | nsJARInputStream::Read] https://bugzilla.mozilla.org/show_bug.cgi?id=1551562
- 3.92% of crashes with Nightlies from Sunday or newer
- volume down since https://bugzilla.mozilla.org/show_bug.cgi?id=1656261 got backed out for uniknown reason
- Crash in [@ mozilla::dom::ScriptLoader::EvaluateScript]
- https://bugzilla.mozilla.org/show_bug.cgi?id=1663440
- 1.36% of crashes
- should get fixed by removal of assertion
- Crash in [@ mozilla::RangeBoundaryBase<T>::GetNextSiblingOfChildAtOffset]
- 0.60% of crashes
- https://bugzilla.mozilla.org/show_bug.cgi?id=1663601
- recent regression, fix work in progress
- Crash in [@ gfxPlatform::Init | nsWindow::IsMainWindowTransparent ]
- https://bugzilla.mozilla.org/show_bug.cgi?id=1663576
- 1.04% of crashes, but only affected Linux
- fixed now
Beta
mc2 crash rate: 1.73, crash incidence: 1.66%, mc1 summary score: 5.60 last week mc2 crash rate: 1.93, crash incidence: 1.95%, mc1 summary score: 6.48
- symbols were lacking for 81.0b6, so no proper crash signatures on socorro yet
- work ongoing on reprocessing
[81.0b7]
- OOM crashes on google web properties: https://bugzilla.mozilla.org/show_bug.cgi?id=1663718 - 3% of browser crashes
- Crash in [@ memcpy | nsJARInputStream::Read]: https://bugzilla.mozilla.org/show_bug.cgi?id=1551562 - 2.3% of browser crashes
- Crash in [@ mozalloc_abort | moz_xmalloc | IPC::Message::Message]: https://bugzilla.mozilla.org/show_bug.cgi?id=1660063 - 0.9% of content crashes
- Crash in [@ mozilla::dom::CloneOrUndefined]: https://bugzilla.mozilla.org/show_bug.cgi?id=1660539 - 0.5% of content crashes
- Crash in [@ mozalloc_abort | moz_xmalloc | mozilla::TransportLayerDtls::GetDecryptedPackets]: https://bugzilla.mozilla.org/show_bug.cgi?id=1659823 - 0.5% of content crashes
Release
mc2 crash rate: 1.39, crash incidence: 0.82%, mc1 summary score: 2.08 last week mc2 crash rate: 1.43, crash incidence: 0.86%, mc1 summary score: 2.71
- Crash in [@ nsJARChannel::Open] during shutdown: https://bugzilla.mozilla.org/show_bug.cgi?id=1663315
- seems to be a one-time crash during the update process
Mobile Stability
Fenix
- Beta #1 top crash https://github.com/mozilla-mobile/fenix/issues/14852
- https://github.com/mozilla-mobile/android-components/issues/8356
- Patch landed that sharded out the Display already aquired https://github.com/mozilla-mobile/android-components/issues/7960
QA
Nightly
- feature tech doc due, trello creation: tomorrow Sep 9th
- Regression Testing in Nightly 82.0a1 - Round 1 is completed (GREEN)
- 4 new bugs were reported: https://mzl.la/32a7vvN
Aurora / Dev Edition
Beta
- Pre-Release Prelim reports were published
- PDF Form Fill - Preliminary report sent with status Yellow due to https://github.com/mozilla/pdf.js/issues/12319
- Modernized Print UI - Preliminary report sent with status Red due to large number of open bugs https://mzl.la/3haLqBx
- Credit Card Autofill - Preliminary Report was sent with Yellow color, more info at: https://docs.google.com/document/d/1-jcC7a_MF-3K7cYE7-Ejby9Z1hqH4oaGf7VolKruzPM/edit
- Support Langpacks across updates - Preliminary status report sent - Yellow because we couldn't test the feature on actual Beta 81 builds, Nightly 81 was used instead
- Tania to reach out to Sebastian about beta builds
- try build could help but if they are unbranded Betas then it would not really help (won't update to another version)
- Dynamic FPI - Bugs behind the yellow preliminary status turned out to be expected behaviour, so the feature has been changed back to on-track
- Haven't received a WNP testing request for 81, deadline is tomorrow
- ryan to ping in https://bugzilla.mozilla.org/show_bug.cgi?id=1661585
- Tania to check with Justin Crawford
Mobile
Action items
Feature Status Review
SUMO Report
Roundtable & Special Topics
- QA is in the process of creating an updated test plan for Nightly Regression Testing
- Merging two test rounds into one
- Rotation based OS coverage (macOS, Linux and Win7, ARM64)
- Removing tests that never failed/ tests that are covered in other test cases