913738 - [B2G] [Buri] [1.2] [Dialer] Caller is unable to hang up after a second call is made while connected to the first call

Status: RESOLVED DUPLICATE of bug 905941

(Firefox OS Graveyard :: Gaia::Dialer, defect)

Description

[Chris Kreinbring [:CKreinbring]]

Attachment: Log of calling second device and hang up attempt

Description:
After making a call while already connected to a call, the caller will be stuck on the call screen of the first caller when both called devices hang up.  A possible suspect is "E/AudioPolicyManager(144): setPhoneState() state 0"

Repro Steps:
1) Update Buri to Build ID: 20130906040204
2) Make a call to another device and have the device answer it.
3) While in the call, tap the New Call button (call symbol with a + next to it).
4) Call another device and have that device answer.
5) Attempt to hang up on either call.

Actual:
The caller is unable to hang up on either call.  The hang up button does not respond, and attempts to close the dialer by long-tapping the Home button fail.

Expected:
The caller can hang up on both calls.

Environmental Variables
Occurs on Buri 1.2 mozilla RIL
Build ID: 20130906040204
Gecko: http://hg.mozilla.org/mozilla-central/rev/ab5f29823236
Gaia: 94e5f269874b02ac0ea796b64ab995fce9efa4b3
Platform Version: 26.0a1
RIL Version: 01.01.00.019.206 

Notes:
Repro frequency: 100%
See attached logcat logs
The only way to become unstuck is to restart the device

Comment 1

[Sarah Parsons]

This issue started to occur on the 9/06 build. It seems that this "New Call" or Conference Call feature was recently added because it does not appear on 9/05 1.2 build or earlier. 

Environmental Variables
Occurs on Buri 1.2 mozilla RIL
Build ID: 20130906040204
Gecko: http://hg.mozilla.org/mozilla-central/rev/ab5f29823236
Gaia: 94e5f269874b02ac0ea796b64ab995fce9efa4b3
Platform Version: 26.0a1
RIL Version: 01.01.00.019.206

Comment 2

[Gene Lian (I already quit Mozilla)]

Looping some Gaia folks.

Comment 3

[Etienne Segonzac (:etienne)]

It doesn't happen on a unagi with the same gecko, so we don't think it's gaia related.

Comment 4

[Hsin-Yi Tsai (she/her) [:hsinyi]]

Attachment: logburi

It looks like a device-dependent problem. There are ril parcel errors there while I don't see them on unagi.

Comment 5

[Joe Cheng [:jcheng] (please needinfo)]

this seems to be on commercial ril. ni? Anshul/mvines

Comment 6

[Michael Vines [:m1] [:evilmachines]]

There's no 1.2 QA testing happening with commercial RIL at the moment.  But I hope we can fix that soon!  It's certainly available to Moz.

Comment 7

[Jason Smith [:jsmith]]

(In reply to Joe Cheng [:jcheng] from comment #5)
> this seems to be on commercial ril. ni? Anshul/mvines

No it's not - this bug was filed involving the Mozilla RIL build on Buri.

Comment 8

[Hsin-Yi Tsai (she/her) [:hsinyi]]

The root cause is the same as bug 905941. Marked as duplicate.

Comment 9

[Hsin-Yi Tsai (she/her) [:hsinyi]]

Not regression. This issue is about Buri device having its own ril parcels.