|Submitter||Diff||Changes||Open Issues||Last Updated|
|Error loading review requests:|
59 bytes, text/x-review-board-request
|Details | Review|
As tested on https://bugzilla.mozilla.org/show_bug.cgi?id=1392204#c18, the following prefs can now be set on Fennec: urlclassifier.malwareTable = goog-harmful-proto,goog-unwanted-proto,test-malware-simple,test-unwanted-simple,test-harmful-simple urlclassifier.phishTable = goog-phish-proto,test-phish-simple These prefs however will need to stay on V2 for Fennec for now because these lists are not available on the V4 server when the platform is set to Android: urlclassifier.downloadAllowTable urlclassifier.downloadBlockTable
We are choosing to do this after 57 in order to have a full Nightly cycle of testing for this feature.
Comment on attachment 8912052 [details] Bug 1394017 - Default to Safe Browsing V4 on Fennec. https://reviewboard.mozilla.org/r/183434/#review188580 Looks good to me, thanks!
Tested locally on Desktop and Fennec.
Pushed by email@example.com: https://hg.mozilla.org/integration/autoland/rev/bca021fc3073 Default to Safe Browsing V4 on Fennec.r=dimi
This bug may have contributed to a sudden change in the Telemetry probe URLCLASSIFIER_VLPS_FALLOCATE_TIME which seems to have occurred in Nightly 20170927. There was a slight increase in file allocation time which might mean a heavier use of variable-length prefix sets... or something? Is this an improvement? A regression? Is this intentional? Is this expected? Is this probe (still) measuring something useful? : http://alerts.telemetry.mozilla.org/index.html#/detectors/1/metrics/2026/alerts/?from=2017-09-27&to=2017-09-27 : https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=7d15bc419c6cd7e9f3b4d41370c3b0e5990c8d1b&tochange=35fbf14b96a633c3f66ea13c1a163a3f3a4219b9
(In reply to Chris H-C :chutten from comment #7) > This bug may have contributed to a sudden change in the Telemetry probe > URLCLASSIFIER_VLPS_FALLOCATE_TIME which seems to have occurred in Nightly > 20170927. > There was a slight increase in file allocation time which might mean a > heavier use of variable-length prefix sets... or something? That makes sense. Android devices are most likely slower than desktops at allocating variable prefixes (or anything, really). Prior to this change, the variable prefixes were only on Desktop. > Is this an improvement? A regression? > > Is this intentional? Is this expected? Given the above, I'd say it's expected. > Is this probe (still) measuring something useful? Small changes are not very meaningful, but this probe will alert us if we land a bad regression in this part of the code.