879387 - Is not possible to display the buttons of " install or cancel " when the payment is completed

Status: RESOLVED FIXED

(Core :: DOM: Device Interfaces, defect)

Description

[eder]

Attachment: Is not possible to display the buttons of install or cancel

steps to reproduce:

1. open the MP_dev (App)
2. Select the any pricetier 
3. Click on the purchase button
4. Enter PIN 
5. Select the option "Buy", Carrier Billing
6. in next screen remains in "Payment completed" but the install or cancel buttons does not appear
 

expected behavior:


The user should be able to install or cancel of the application after complete the payment


observed behavior:


Is not possible to display the buttons of " install or cancel  " when the payment is completed


Note: the payment of the application is not found  in the user's carrier bill

Comment 1

[eder]

Attachment: when the payment is completed Is not possible to display the buttons of " install or cancel "

Comment 2

[Kumar McMillan [:kumar]]

For this one I'd actually like to look at the output of `adb logcat` not the http log. Can you attach that?

Comment 3

[eder]

Attachment: Is not possible to display the buttons of " install or cancel "_Poland

Comment 4

[eder]

done Kumar

Comment 5

[Kumar McMillan [:kumar]]

Thanks. I don't see any JS error or any hint about what's going on here. I also can't reproduce it with a US payment.

Comment 6

[Kumar McMillan [:kumar]]

Does it happen every time?

Comment 7

[eder]

yes , this happens all the time.

Comment 8

[eder]

Attachment: payment completed screen is taking too long_wifi_poland

Comment 9

[eder]

Attachment: screenshot _payment completed screen is taking too long_wifi_poland

Comment 10

[Vikas Nanda]

Note that this issue also occurs when purchasing a premium package.

Comment 11

[Vikas Nanda]

Attachment: Log for package paid app

Comment 13

[Kumar McMillan [:kumar]]

this is happening in the UK too (not just Poland). I still don't see it in the US.

Comment 14

[krupa raj[:krupa]]

Looks like this is a 1.1 platform bug.

Comment 15

[Kumar McMillan [:kumar]]

ferjm, are you set up to test the pay flow end to end? Perhaps a reduced trusted UI test will also shed light on this. Here's what I see:

* our code calls paymentSuccess() here https://github.com/mozilla/webpay/blob/0a8cbac2122d6749127edee3897912eb8ae5a53d/media/js/pay/pay.js#L102
* the log message one line above is printed
* there are no JavaScript errors
* however, the window does not close after paymentSuccess() is called
* on 1.01 this bug is not happening
* on 1.1 this bug is happening

I just reproduced it with this build: https://pvtbuilds.mozilla.org/pub/mozilla.org/b2g/nightly/mozilla-b2g18-unagi/2013/06/2013-06-05-07-02-07/

Zac linked to the gecko revision and build ID he was on when he reproduced it: https://bugzilla.mozilla.org/show_bug.cgi?id=879761#c2

Comment 16

[Jason Smith [:jsmith]]

I'm pretty convinced this is a regression from bug 872987.

Comment 17

[Jason Smith [:jsmith]]

Interesting logcat points from dupe:

E/GeckoConsole( 5622): [JavaScript Error: "rp is undefined" {file: "resource://gre/modules/identity/MinimalIdentity.jsm" line: 152}]

E/GeckoConsole( 5622): [JavaScript Error: "Error: Permission denied to access object"]
E/GeckoConsole( 5622): [JavaScript Error: "Error: "]

Comment 18

[Jason Smith [:jsmith]]

Note to triagers - yes, this needs to block. This is blocking country-specific testing for payments right now.

Comment 19

[Fernando Jiménez Moreno [:ferjm]]

Attachment: v1

I've managed to reproduce this issue with Kumar's in-app payment test app.

The reason of the failure seems to be that 'this' is not what I expected at [1] and [2] when the payment flow is mixed up with the identity one. It worked with a simple payment flow though.

[1] https://mxr.mozilla.org/mozilla-central/source/b2g/chrome/content/payment.js#84
[2] https://mxr.mozilla.org/mozilla-central/source/b2g/chrome/content/payment.js#94

Comment 20

[Lukas Blakk [:lsblakk] use ?needinfo]

Partner triage - see comment 18.

Comment 22

[Fernando Jiménez Moreno [:ferjm]]

http://hg.mozilla.org/projects/birch/rev/f7489cbd7385

Comment 23

[Ed Morley [:emorley]]

https://hg.mozilla.org/mozilla-central/rev/f7489cbd7385

Comment 24

[Vikas Nanda]

Attachment: The payment fails to go through when purchasing an app using FW 1.1 over the moviestar ES network. Note that MP Stage was used for this log.

Comment 25

[Vikas Nanda]

Attachment: Payment and installation of application successful. Application which was installed was pricetier 1 using MP stage over the Moviestar ES network (roaming in Poland). Firmware used for this log is 1.01

Comment 26

[Jason Smith [:jsmith]]

(In reply to Vikas Nanda from comment #24)
> Created attachment 760387 [details]
> The payment fails to go through when purchasing an app using FW 1.1 over the
> moviestar ES network. Note that MP Stage was used for this log.

This hasn't landed on b2g18 yet, so this won't work on 1.1. The patch has only landed on central.

Comment 27

[Steve Ruston]

(In reply to Jason Smith [:jsmith] from comment #26)
> (In reply to Vikas Nanda from comment #24)
> > Created attachment 760387 [details]
> > The payment fails to go through when purchasing an app using FW 1.1 over the
> > moviestar ES network. Note that MP Stage was used for this log.
> 
> This hasn't landed on b2g18 yet, so this won't work on 1.1. The patch has
> only landed on central.

Jason...what does this mean? When will it land? 

We are now in Certification in Columbia where Payments are being tested by the Movistar Columbia team to get sign off for launch

Comment 28

[Jason Smith [:jsmith]]

(In reply to Steve Ruston from comment #27)
> (In reply to Jason Smith [:jsmith] from comment #26)
> > (In reply to Vikas Nanda from comment #24)
> > > Created attachment 760387 [details]
> > > The payment fails to go through when purchasing an app using FW 1.1 over the
> > > moviestar ES network. Note that MP Stage was used for this log.
> > 
> > This hasn't landed on b2g18 yet, so this won't work on 1.1. The patch has
> > only landed on central.
> 
> Jason...what does this mean? When will it land? 

This bug needs a leo+ to land. So we're waiting on leo to triage this bug.

> 
> We are now in Certification in Columbia where Payments are being tested by
> the Movistar Columbia team to get sign off for launch

This bug should not be affecting 1.01 testing, so I don't think you should be blocked on certification testing for 1.01. This would block certification testing on 1.1, however.

Comment 29

[Ryan VanderMeulen [:RyanVM]]

John, can you please assist with the uplift? :)

Comment 30

[Jason Smith [:jsmith]]

Please provide information on how this was tested when verifying the bug. This hasn't landed on b2g18, so I don't know how you verified this working.

Comment 31

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

(In reply to Ryan VanderMeulen [:RyanVM][UTC-4] from comment #29)
> John, can you please assist with the uplift? :)

Is there a Gaia patch for this bug?

Comment 32

[Jason Smith [:jsmith]]

(In reply to John Ford [:jhford] -- If you expect a reply from me, use needsinfo? instead of CC from comment #31)
> (In reply to Ryan VanderMeulen [:RyanVM][UTC-4] from comment #29)
> > John, can you please assist with the uplift? :)
> 
> Is there a Gaia patch for this bug?

Nope. It's a gecko patch.

Comment 33

[Ryan VanderMeulen [:RyanVM]]

Argh, I'm blind. Sorry about that, I'll get it in the morning.

Comment 34

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-b2g18/rev/4ca769f0f1f2

Comment 35

[Jason Smith [:jsmith]]

Please see https://bugzilla.mozilla.org/show_bug.cgi?id=879387#c30. You still haven't provided any of the correct information to indicate this is verified.

Comment 36

[Jason Smith [:jsmith]]

eder - please stop setting the flag until you include the information I have stated in comment 35.

Comment 37

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-b2g18_v1_1_0_hd/rev/4ca769f0f1f2

Comment 38

[Kumar McMillan [:kumar]]

Eder, starting on June 14th you should be able to use the nightly 1.1 build (right, jsmith?) to verify this change. Ask Krupa if you need help to get the right build.

Comment 40

[Jason Smith [:jsmith]]

Once again, please see comment 36.

Comment 41

[Vikas Nanda]

Hi Jason, 

I installed the latest build for 1.01 (19/06/13). I received this build from Steve Ruston. I have tried the steps again and this issue no longer occurs for me. Can I mark this as verified? 

Thanks

Vikas

Comment 42

[Jason Smith [:jsmith]]

(In reply to Vikas Nanda from comment #41)
> Hi Jason, 
> 
> I installed the latest build for 1.01 (19/06/13). I received this build from
> Steve Ruston. I have tried the steps again and this issue no longer occurs
> for me. Can I mark this as verified? 
> 
> Thanks
> 
> Vikas

This bug only reproduces on 1.1 or later, so that test coverage doesn't ensure that the fix provided here works. You need to test this on a 1.1 build.