991577 - [Tarako][MMS][CardView] A white screen shows up after tapping on SMS card in cardview instead of resuming SMS app.

Status: RESOLVED DUPLICATE of bug 935750

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

Description

[Bingqing Li]

Environment:
---------------------------------------------------
Gaia      c418ec10d1e1d53c6757ad12b2320c204808d251
Gecko     https://hg.mozilla.org/releases/mozilla-b2g28_v1_3t/rev/36b1279ef6df
BuildID   20140402164000
Version   28.1
ro.build.version.incremental=eng.cltbld.20140402.202437
ro.build.date=Wed Apr  2 20:24:44 EDT 2014


Steps to reproduce:
---------------------------------------------------
1. Send multiple mms to the test device using the same number.
   (e.g. Send 3 mms to the test device - one is attached with a jpg file, one is attached with an amr file, and the last one is attached with a 3gp file.)

2. Launch SMS, and open the message thread.

3. Save the attachments

4. View the attachments(e.g view pictures, and play the music and videos)

5. Repeat step 3 and 4 for several times

6. Long press 'Home' button
   --> Cardview shows up

7. Tap on the SMS card in cardview


Actual result:
---------------------------------------
A white screen shows up. Cannot resume SMS app.


Expected result:
---------------------------------------
Resume SMS app.


Additional info:
---------------------------------------
This cannot be reproduced 100%. 
Pls try more times.

Comment 1

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

from the steps, it seems like you have to intentionally trying to break it, not blocking tarako release.

Comment 2

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

This is a dup bug.  Bug 983772, I did it with the Music app.  You can do it with any app.

Comment 3

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

Oh sorry.  Actually it's not quite a dup.

Comment 4

[Jason Smith [:jsmith]]

Bingqing - Can you try reducing the STR here?

Comment 5

[Jason Smith [:jsmith]]

(In reply to Jason Smith [:jsmith] from comment #4)
> Bingqing - Can you try reducing the STR here?

i.e. try to find the minimal STR to use here to reproduce the bug

Comment 6

[[e-mail], [e-mail]]

Attachment: dmesg_white_screen_blocking_while_launching_SMS.txt

This issue is hard to reproduce. I try to minimal reproduced steps. My steps are as following.

1. Launch SMS via app icon in homescreen
2. Press home button go back to homescreen
3. Repeat steps 1~2 several time(might be more than 10?)

Sometimes, you will see a white screen while launching SMS app. I attach dmesg log for reference. But I don't attach console log. Because I don't see some useful message in console log.

Comment 7

[[e-mail], [e-mail]]

(In reply to Ian Liu [:ianliu] from comment #6)
> Created attachment 8405160 [details]
> dmesg_white_screen_blocking_while_launching_SMS.txt
> 
> This issue is hard to reproduce. I try to minimal reproduced steps. My steps
> are as following.
> 
> 1. Launch SMS via app icon in homescreen
> 2. Press home button go back to homescreen
> 3. Repeat steps 1~2 several time(might be more than 10?)
> 
> Sometimes, you will see a white screen while launching SMS app. I attach
> dmesg log for reference. But I don't attach console log. Because I don't see
> some useful message in console log.

If you can reproduce it as my steps, try to long press home button for viewing recent apps. Tap on the SMS card in card view. Then, you can see the actual result.

Actual result:
---------------------------------------
A white screen shows up. Cannot resume SMS app.

Build version:
Gaia      27a0e773e01eed74e20709bdcab6894469f42a72
Gecko     f6893b30657b8b8208a5c9de8b6b96b9a09a3fb1
BuildID   20140410042058
Version   28.1
ro.build.version.incremental=250
ro.build.date=Thu Apr 10 04:16:35 CST 2014

Build: sp6821a_gonk-userdebug 4.0.4.0.4.0.4 OPENMASTER 250 test-keys

Comment 8

[Julien Wajsberg [:julienw]]

FTR I think I saw this on a Peak 1.3 too, from time to time, but no absolute STR.

Comment 9

[[e-mail], [e-mail]]

I try to use the same steps as comment 6 with Contact app. And I got a black screen while I'm able to reproduce. Shall I suspect that an app is launching/re-launch while it just be killed in the same time? 

The same symptom is that:
1. The app is not able to launch in recent app. It should be killed I think.
2. Launch the app from home screen. It looks like re-launch.

Comment 10

[Bingqing Li]

This bug can be easily reproduced on build with the steps that Ian mentioned in comment #6.

> 1. Launch SMS via app icon in homescreen
> 2. Press home button go back to homescreen
> 3. Repeat steps 1~2 several time(might be more than 10?)

Build Info:
---------------------------------------------------
Gaia      bbba09c732c35d9434ecb04d5e2e41d05511f4d9
Gecko     https://hg.mozilla.org/releases/mozilla-b2g28_v1_3t/rev/65b4c58a7da9
BuildID   20140410164004
Version   28.1
ro.build.version.incremental=eng.cltbld.20140410.210129
ro.build.date=Thu Apr 10 21:01:38 EDT 2014

Comment 11

[Bingqing Li]

(In reply to Bingqing Li from comment #10)
> This bug can be easily reproduced on build with the steps that Ian mentioned
This bug can be easily reproduced on build 20140410164004.

Comment 12

[Jason Smith [:jsmith]]

Can we include a reproduction rate using comment 10's STR?

Comment 13

[Jason Smith [:jsmith]]

(In reply to Jason Smith [:jsmith] from comment #12)
> Can we include a reproduction rate using comment 10's STR?

Along with knowledge of how many launches of the SMS app it took to reproduce the bug in each case.

Comment 14

[Brogan Zumwalt [Inactive]]

Able to reproduce using STR in comment 10 on todays 1.3 Tarako PVT build. It is, however, a very low reproduction rate. 

Repro Rate: 1/30, 3%

I got a slightly higher Repro rate using these STR:

1) Launch SMS app via icon on homescreen
2) Open existing MMS message thread
3) Press home button to return to homescreen
4) Select SMS icon to reopen SMS app
[Repeat Steps 3-4]

Repro rate: 1/20, 5%


Attempted reproducing both STR's with 2 existing MMS threads.

1.3T Environmental Variables:
Device: Tarako 1.3T
BuildID: 20140414004001
Gaia: 0e7f21e61625b75a9149480cd5a259211549f020
Gecko: 3b02811c314a
Version: 28.1
Firmware Version: sp6821a

Comment 15

[Steve Chung [:steveck]]

If you consume too much memory in message app(e.g.  loading the huge thread), app is possible to be killed while switching to background(launching cardview will also push the app to backgroung) because of oom issue. Maybe this white screen issue happens when we are trying to launch an app which should be killed but action is not completed yet. ni? Alive for the possibility of this guessing and how we could avoid this in 1.3 window manager.

Comment 16

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

Bug 935750 fixed this issue in master branch.
For v1.3 I think we could:
1. remove the card when cardsview.js got appterminated event.
2. Close cardview or shift to next card.

Comment 17

[James Zhang (Spreadtrum)]

(In reply to Alive Kuo [:alive][NEEDINFO!][God bless Taiwan.] from comment #16)
> Bug 935750 fixed this issue in master branch.
> For v1.3 I think we could:
> 1. remove the card when cardsview.js got appterminated event.
> 2. Close cardview or shift to next card.

Can we cherry-pick this patch from v1.4?

Comment 18

[Steven Yang [:styang]]

(In reply to James Zhang from comment #17)
> (In reply to Alive Kuo [:alive][NEEDINFO!][God bless Taiwan.] from comment
> #16)
> > Bug 935750 fixed this issue in master branch.
> > For v1.3 I think we could:
> > 1. remove the card when cardsview.js got appterminated event.
> > 2. Close cardview or shift to next card.
> 
> Can we cherry-pick this patch from v1.4?

Ｉdon't think the work is trivial to pick the patch back to 1.3T. Hi Alive, could you confirm it. Thanks!

Comment 19

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

Yap, this is a non-trivial feature request to me. I recommend not blocking on this bug.

Comment 20

[bhavana bajaj [:bajaj]]

given its not trivial here and since a decision was made earlier on this not being a blocker, minusing it.

Comment 22

[Julien Wajsberg [:julienw]]

If this is fixed on master, like alive said in comment 16, shouldn't we close this bug or dupe it to bug 935750 ?