822054 - USSD hangs on Sending

Status: RESOLVED FIXED

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

Description

[Alex Wafula]

Attachment: gaia hang sending screen during ussd action

It's still unclear whether the USSD hang happened because the mobile service provider's server took too long to respond or something was up with the phone...So steps to regenerate:

1. Enter a valid USSD code on the dialer for the SIM being used
2. Press call

Expected: USSD code is sent and result is returned by the mobile service provider
Actual: Sending screen appears and persists. Nothing else can be accessed on the phone except the power-off menu

Device/Software Details:
LG Optimus
Build Identifier: 20121207170359
Gaia: Nightly channel

Comment 1

[Fernando Jiménez Moreno [:ferjm]]

I am not able to reproduce this issue. I dont have an LG Optimus device to test.. It would be really helpful if you could provide the RIL debug output ( adb logcat -b radio ). 

In any case, I am afraid that the platform does nothing until the RIL notifies back about a successfull or failure delivery of a sent USSD message. It feels like we should have a timeout somewhere to cancel the spinner and let the user retry later.  IIRC there are no timeouts in the Gecko RIL code, so I would say that Gaia is probably the best place to host this timeout.  What do you think Philipp?

We could probably also add a cancelation option in the UX so the user can cancel an USSD request while it is being sent. Marco, do you think that this is something that we should add to the current UX?

Comment 2

[marcoc]

(In reply to Fernando Jiménez Moreno [:ferjm] from comment #1)
> 
> We could probably also add a cancelation option in the UX so the user can
> cancel an USSD request while it is being sent. Marco, do you think that this
> is something that we should add to the current UX?

Yes, I agree Fernando. I was able to reproduce this on the Unagi and there is no way of stopping the Sending cycle. 
Do you need a visual for this?

Comment 3

[Philipp von Weitershausen [:philikon]]

(In reply to Fernando Jiménez Moreno [:ferjm] from comment #1)
> In any case, I am afraid that the platform does nothing until the RIL
> notifies back about a successfull or failure delivery of a sent USSD
> message. It feels like we should have a timeout somewhere to cancel the
> spinner and let the user retry later.  IIRC there are no timeouts in the
> Gecko RIL code, so I would say that Gaia is probably the best place to host
> this timeout.  What do you think Philipp?

I agree. That or the cancel option that you sugest.

Comment 4

[Fernando Jiménez Moreno [:ferjm]]

Thanks Marco and Philipp! I'll ask for tracking-b2g18 since this issue makes the device unusable until reboot.

Comment 5

[Alex Keybl [:akeybl]]

Tracking due to the fact that the device is unusable until reboot, although this isn't a common user scenario.

Comment 6

[Fernando Jiménez Moreno [:ferjm]]

Attachment: Pointer to Github pull request: https://github.com/mozilla-b2g/gaia/pull/8243

Pointer to Github pull-request

Comment 7

[Fernando Jiménez Moreno [:ferjm]]

Comment on attachment 716636 [details]
Pointer to Github pull request: https://github.com/mozilla-b2g/gaia/pull/8243

Ismael could you take a look at the HTML and CSS changes, please?

Etienne could you check the JS and tests, please?

Thanks!

Comment 8

[Etienne Segonzac (:etienne)]

Comment on attachment 716636 [details]
Pointer to Github pull request: https://github.com/mozilla-b2g/gaia/pull/8243

Thanks this is great!

Comment 9

[Fernando Jiménez Moreno [:ferjm]]

Thanks!

https://github.com/mozilla-b2g/gaia/commit/62beab356df3283153b2e849822b820915792a72

Comment 10

[Fernando Jiménez Moreno [:ferjm]]

Comment on attachment 716636 [details]
Pointer to Github pull request: https://github.com/mozilla-b2g/gaia/pull/8243


User impact if declined: No way to cancel an ongoing USSD request with the potential issue of getting stuck in the "Sending" screen.
Testing completed: Manual testing and unit tests added.
Risk to taking this patch (and alternatives if risky): Low risk
String or UUID changes made by this patch: None

Comment 11

[John Ford [:jhford] CET/CEST Berlin Time]

v1-train: 14a6e68