Firefox/Channels/Meetings/2017-08-01
From MozillaWiki
Channel Meeting Details
- Tuesdays and Thursdays at 10am 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
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
marcia, jcristau, mtabara, nicole, dveditz, ritu, elan, philipp
Previous Actions
Schedule Update
- shipping 55.0b14 (Fennec/DevEdition) and 55.0 rc1 (beta) today
- Fix for gmail issue that might drive a RC2
Add-ons
Stability
Nightly
awsy rate: 5.25 (browser 2.85, content 2.39) telemetry (m+c-s) from saturday: 15.53 (last week: 7.66, this time last cycle: 6.34)
[marcia]
- Nightly rate is in the high range and slightly higher than what was reported at last Thursday's Channel meeting. Over the weekend the rate increased as well.
- Top crashers
- TFO crash https://bugzilla.mozilla.org/show_bug.cgi?id=1381362. Should go away when it is turned off. Same with https://bugzilla.mozilla.org/show_bug.cgi?id=1362978 (turned off in today's build)
- Networking crash https://bugzilla.mozilla.org/show_bug.cgi?id=1363372 - patch checked into central
- Async shutdown crashes https://bugzilla.mozilla.org/show_bug.cgi?id=1380267 - has a patch
- Stylo crashes https://bugzilla.mozilla.org/show_bug.cgi?id=1382568 and https://bugzilla.mozilla.org/show_bug.cgi?id=1384824
- Core|DOM bug https://bugzilla.mozilla.org/show_bug.cgi?id=1364138 is active again and has moved up to #14
- Top crashers
- Shutdown Hangs on Mac and Linux https://bugzilla.mozilla.org/show_bug.cgi?id=1383742 seems to be fixed
- Nightly issue with Kaspersky - marcia reported it to them and they added comments to the bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1384327
Beta
awsy rate: 0.85 (browser 0.62, content 0.22) telemetry (m+c-s) from saturday: 4.43 (last week: 4.65, this time last cycle: 3.98)
[philipp] remaining top regressions in 55.0b13
- https://bugzilla.mozilla.org/show_bug.cgi?id=1373934 - 1.3% of browser crashes
- Crashes in js::Scope::clone: https://bugzilla.mozilla.org/show_bug.cgi?id=1367896 - 1.1% of browser crashes
- Crashes in LongestPhaseSelfTime: https://bugzilla.mozilla.org/show_bug.cgi?id=1366217 - 1% of browser crashes
- Crashes with McAfee Webadvisor: https://bugzilla.mozilla.org/show_bug.cgi?id=1376514 - 1% of browser crashes
- https://bugzilla.mozilla.org/show_bug.cgi?id=1367727 - 0.9% of browser crashes
- Crashes with Webroot SecureAnywhere in BaseThreadInitThunk https://bugzilla.mozilla.org/show_bug.cgi?id=1361410 - 0.6% of browser crashes
- igd10iumd32.dll | CContext::EmptyOutAllDDIBindPoints: https://bugzilla.mozilla.org/show_bug.cgi?id=1351349 - 0.5% of all browser crashes
Release
awsy rate: 0.80 (browser 0.57, content 0.23) telemetry (m+c-s) from saturday: 2.77 (last week: 2.62)
[marcia]
- Explosiveness report is quiet: https://crash-analysis.mozilla.com/release-mgmt/2017-07-31/2017-07-31.firefox.release.explosiveness.html
- Not much actionable in the top 10. Some shutdown hangs and JS crashes, Websense, etc
- https://bugzilla.mozilla.org/show_bug.cgi?id=1337105 is the crash where Websense was called out. All of the crashes seem to be GC
Mobile
awsy rate: release 0.57, beta 1.08, nightly 2.24
[marcia]
- Mobile rate looks good across all three channels
- Nightly #2 browser crash https://bugzilla.mozilla.org/show_bug.cgi?id=1384828 has a patch
Performance (Thu)
Aurora / Dev Edition
Beta
Release
QE
Nightly
Aurora / Dev Edition
Beta
Action items
Mobile
Performance
RelEng
Marketing/Communications
User Advocacy
Roundtable
[elan] multi for 55 is a 'go'
- We will deal with any regressions as they come. Risk areas continue to be stability, re:TCP Fast open pref on (occasionally), 20% of Nightly has stylo on
[marcia]
- Tracking Talos Regressions?
- [ritu] Makes perfect sense to do it, we should triage and track them and mark them as blockers if need be
- [kevin] Makes sense to mark the branches affected when the bug is filed