Firefox/Channels/Meetings/2019-04-02
Channel Meeting Details
- Tuesdays 10am and Thursdays at 8am Pacific Time
- 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
Contents
Attendees
marcia, Aryx, lizzard, RyanVM, philipp, jcristau, pascalc, overholt, jlund, wlach, Thomas, TomGrab, Tania
Schedule Update
Current releases dashboard: https://mozilla.github.io/delivery-dashboard 67.beta 7 just shipped
Off-train releases
Current Shield studies: https://strategy-and-insights.mozilla.com/shield-studies/index.html
Stability
- Some work on Socorro done yesterday. See the email from Will-KG to @stability for more info
- Fallout from CheckedInt assertions restricted to fennec so far
Nightly
firefox nightly summary score: 7.99, Thursday 10.17
- Earlier today the score was missing, jcristau filed https://bugzilla.mozilla.org/show_bug.cgi?id=1541033
- Jan filed quite a few DOM | Web Storage bugs https://mzl.la/2U9tSPv
- note that this will not ride in 67 (we were hoping for an uplift but the hard-to-reproduce crashes coming from Socorro are proving hard to fix (well, all of them))
- Filed https://bugzilla.mozilla.org/show_bug.cgi?id=1541114 to account for a crash happening with a cloud based videoconferencing site
- Core | A/V Playback https://bugzilla.mozilla.org/show_bug.cgi?id=1535335 is rising. One fix attempted.
- Latest nightly crash triage: https://wiki.mozilla.org/NightlyCrashTriage/2019Q1#2019-04-02_-_gsvelto
Beta
firefox beta summary score: 4.38, Thursday 5.01
[67.0b6] regressing crashes
- Content crash in [@ mozilla::dom::ClientSource::SetController]: https://bugzilla.mozilla.org/show_bug.cgi?id=1535699 - 5.8% of tab crashes
- under active investigation; regressed by https://bugzilla.mozilla.org/show_bug.cgi?id=1525125
- IndexedDB crash: https://bugzilla.mozilla.org/show_bug.cgi?id=1528690 - 3.5% of browser crashes
- regressing in volume in the 67 cycle
- Tom Tung is on this
- shutdown crash (sec): https://bugzilla.mozilla.org/show_bug.cgi?id=1535301 - 1.5% of browser crashes --@Marsha - Please provide bug access to Wennie Leung; done
- Crash in mozilla::dom::quota::QuotaObject::LockedMaybeUpdateSize: https://bugzilla.mozilla.org/show_bug.cgi?id=1516333 - 1.5% of browser crashes
- related to LSNG work, so later beta builds probably won't be affected as this is not not riding to release yet (correct)
- Crash in walCleanupHash (sqlite): https://bugzilla.mozilla.org/show_bug.cgi?id=1539640 - 0.5% of browser crashes
Release
firefox release summary score: 2.26, Thursday 2.64
- [@ OOM | small ] signature rising in 66 by ~1300 daily reports - no apparent correlations
- filed https://bugzilla.mozilla.org/show_bug.cgi?id=1541092 (only covers a fraction of the increase though)
Mobile
- Crash rate is calculated based on usage hours (the time Firefox is actively used).
- Usage hours will be significantly lower on Fennec than on desktop Firefox, so we can't compare mobile rates to desktop rates.
- Release
fennec release summary score: 17.42, Thursday 18.79
- https://bugzilla.mozilla.org/show_bug.cgi?id=1526396 continues to be high, but not much we can do about it. Filed
https://bugzilla.mozilla.org/show_bug.cgi?id=1541090 to add that signature to the skip list.
- https://bugzilla.mozilla.org/show_bug.cgi?id=1508740 is #6 overall, and unowned at the moment
- Beta
fennec beta summary score: 144.82, Thursday 323.18
- Top crash https://bugzilla.mozilla.org/show_bug.cgi?id=1472526 is correlated to Adreno 530
- #3 overall crash in DOM - Web Storage - https://bugzilla.mozilla.org/show_bug.cgi?id=1534882
- #4 overall crash - https://bugzilla.mozilla.org/show_bug.cgi?id=1539465 - [@ mdb_env_cthr_toggle]. Still trying to isolate the cause.
- Nightly
fennec nightly summary score: 70.57, Thursday 75.35
- https://bugzilla.mozilla.org/show_bug.cgi?id=1521158 is still the top crash is there is work in progress
- #2 top crash is a series of crashes including https://bugzilla.mozilla.org/show_bug.cgi?id=1540748 which are because of integer overflow checks which were made into release asserts
- #3 top crash is https://bugzilla.mozilla.org/show_bug.cgi?id=1318810. Theory is it is because of PGO builds
- We should be on the lookout for other signatures such as this one that may have resurfaced
- #4 top crash https://bugzilla.mozilla.org/show_bug.cgi?id=1523185, Crash in SnowWhiteKiller::Visit