806965 - B2G Network Manager: navigator.onLine starts true even if false

Status: RESOLVED FIXED

(Firefox OS Graveyard :: General, defect, P3)

Description

[Ran Ben Aharon [:ranbena] EverythingMe]

User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_2) AppleWebKit/537.4 (KHTML, like Gecko) Chrome/22.0.1229.94 Safari/537.4

Steps to reproduce:

Once the device boots and connection is offline by default


Actual results:

navigator.onLine returns true


Expected results:

navigator.onLine returns false.

Only after connecting and disconnecting does the property return false as expected.

Comment 1

[Chris Lee [:clee]]

Does this impact WiFi connections for users or is it just the flag that is off?

Comment 2

[Ran Ben Aharon [:ranbena] EverythingMe]

Just a flag

Comment 3

[Ran Ben Aharon [:ranbena] EverythingMe]

It's critical though to Evme.
When Evme thinks it's connected but really isn't - neither results nor an appropriate message are displayed.

Comment 4

[Jose Manuel Cantera]

Fixing this is also needed for FB integration. UX has advised to disable FB options is the user is not onLine.

Comment 5

[(inactive after 6/18) Alive Kuo [:alive]]

(In reply to Ran Ben Aharon (everything.me) from comment #2)
> Just a flag

This is exactly a flag but also a HTML5 feature. With navigator.onLine the web gains the status of network connection and then could do something specially if they knows they are at offline mode now.

Comment 6

[Blake Kaplan (:mrbkap) (inactive)]

Many thanks to Andrea for helping me look at this yesterday.

Comment 7

[Blake Kaplan (:mrbkap) (inactive)]

Attachment: Proposed fix

The actual fix here is pretty straightforward: nsIOService assumes that we're online until told otherwise. Bug 552829 added notifications to tell child processes that when the online/offline state changed, however didn't do anything to initialize the state on process startup.

There are two questions of taste to address though: is calling a Recv method directly poor form? I named the new state getter pretty generically. I could rename it to something more specific until we're sure we have to add other stuff to it.

Comment 8

[Justin Lebar (not reading bugmail)]

Comment on attachment 679803 [details] [diff] [review]
Proposed fix

Written just for you: https://wiki.mozilla.org/B2G/Hacking#Git

Anyway, this seems fine to me.  Maybe ask cjones if he's OK delegating the review to me?  I don't have explicit review powers here.

Comment 9

[Blake Kaplan (:mrbkap) (inactive)]

(I spoke to cjones and he was fine with jlebar reviewing here).

https://hg.mozilla.org/integration/mozilla-inbound/rev/efb2dca44965

Comment 10

[(no longer active)]

https://hg.mozilla.org/mozilla-central/rev/efb2dca44965

Comment 11

[Blake Kaplan (:mrbkap) (inactive)]

https://hg.mozilla.org/releases/mozilla-aurora/rev/b05679d0dc77