1305478 - Use 0-1 min as the initial update delay for both V2 and V4

Status: RESOLVED FIXED

(Toolkit :: Safe Browsing, defect, P2)

Description

[Henry Chang [:hchang]]

Safe Browsing V2 and V4 uses different initial update delay: v2 is 0~5 and v4 is 0~1. We wonder if we can use the same initial delay for both v2 and v4.

Comment 1

[François Marier [:francois]]

I will check with Google to see if we can always do 0-1 minutes.

Comment 2

[François Marier [:francois]]

(In reply to François Marier [:francois] from comment #1)
> I will check with Google to see if we can always do 0-1 minutes.

Google is fine with our simplifying our code and using 0-1 minutes for both V2 and V4.

Comment 3

[Henry Chang [:hchang]]

Attachment: Bug 1305478 - Unify the initial update delay (for both v2 and v4) to 0~1 minute.

Review commit: https://reviewboard.mozilla.org/r/84766/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/84766/

Comment 4

[Henry Chang [:hchang]]

Comment on attachment 8799625 [details]
Bug 1305478 - Unify the initial update delay (for both v2 and v4) to 0~1 minute.

Review request updated; see interdiff: https://reviewboard.mozilla.org/r/84766/diff/1-2/

Comment 5

[Henry Chang [:hchang]]

There is a 3000ms minimum initial update delay since the very beginning [1]. No idea if we should remove this empirical bound.

[1] https://hg.mozilla.org/mozilla-central/annotate/0521f176e739/toolkit/components/url-classifier/content/listmanager.js#l219

Comment 6

[François Marier [:francois]]

(In reply to Henry Chang [:henry][:hchang] from comment #5)
> There is a 3000ms minimum initial update delay since the very beginning [1].
> No idea if we should remove this empirical bound.

I'm fine with this extra 3 seconds. It seems like a good idea to avoid doing anything in the first 3 seconds of startup. There is already a lot of I/O at this point.

Comment 7

[François Marier [:francois]]

Comment on attachment 8799625 [details]
Bug 1305478 - Unify the initial update delay (for both v2 and v4) to 0~1 minute.

https://reviewboard.mozilla.org/r/84766/#review83944

Comment 8

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/5dc333cf5a57
Unify the initial update delay (for both v2 and v4) to 0~1 minute. r=francois

Comment 9

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/5dc333cf5a57