989727 - [tarako] message setting does not open when invoke from SMS for the second time

Status: RESOLVED DUPLICATE of bug 991023

(Firefox OS Graveyard :: General, defect)

Description

[Peipei Cheng (needinfo if you need my action)]

Steps to reproduce
----------------------------------------------------------------------------
1. Launch SMS
2. Create a new message
3. on New message page, tap on the top right coner and select Settings
4. Message settings opens
5. Tap on Home button, kill SMS and Settings
6. Repeat 1~4 for 1 or two times
   --> You will see message settings cannot be launched from SMS.

Please see attached adblogcat. Test starts at: 03-30 19:10:13.070

Build info
--------------------------------------------------------------------
Gaia 14ef4fcdf9199f04f7678755c917dc77f51e13ba
Gecko https://hg.mozilla.org/releases/mozilla-b2g28_v1_3t/rev/b574a7967338
BuildID 20140329004002
Version 28.1
ro.build.version.incremental=70
ro.build.date=Fri Mar 28 06:17:40 CST 2014

Comment 1

[Julien Wajsberg [:julienw]]

I see no attached logcat

Comment 2

[Peipei Cheng (needinfo if you need my action)]

Attachment: adb logcat

Sorry I forgot. Here is the adb log. Test start: 03-30 19:10:13.070

Comment 3

[Tim Guan-tin Chien [:timdream] (please needinfo)]

We need to look at if mozChromeEvent/system message does get filed again.

Comment 4

[John Hammink]

Can we get an about:memory log on this one?  Also I'll try to repro on 1.3

Comment 5

[John Hammink]

I checked this on yesterday's 1.3:  

Gaia      c5cd3a11e91339163b351d50769eaca0067da860
Gecko     https://hg.mozilla.org/releases/mozilla-b2g28_v1_3/rev/5045a67b47ed
BuildID   20140401164001
Version   28.0
ro.build.version.incremental=324
ro.build.date=Thu Dec 19 14:04:55 CST 2013

and I'm able to do this up to 10 times on the buri, so as expected, it's related to 1.3T.

Comment 6

[John Hammink]

Reproduced. Attached about memory logs.

Gaia      b4ed71d94c495d690ceaa43033c74d4249d5b426
Gecko     8c06de2739ed27e84b70100a6a0c33e8278eba1d
BuildID   20140402145142
Version   28.1

Comment 7

[John Hammink]

Attachment: about-memory-0.zip

Comment 8

[John Hammink]

Clarification:  I could NOT reproduce this on 1.3 (buri).  I could reproduce this on 1.3T; memory logs attached.

Comment 9

[Tim Guan-tin Chien [:timdream] (please needinfo)]

Ian, would you be able to add some console.log() to the activity handling code and do some testing to see if this is a Gecko or Gaia issue?

Cervantes, can you tell if this is a system message issue from reading the logcat?

Comment 10

[Cervantes Yu [:cyu] [:cervantes]]

I can't see why it fails to launch the settings app. Do you have dmesg for it?

Comment 11

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

Cervantes,

I add some console.log() and enabled debug flag in activity handling code. If settings app is launched successfully via activity request, the log is as following.
======= success =======
I/Gecko   ( 1855): [Parent 1855] WARNING: waitpid failed pid:1984 errno:10: file /var/lib/jenkins/jobs/B2G.v1.3.0.Tarako.v1.3t/workspace/gecko/ipc/chromium/src/base/process_util_posix.cc, line 254
E/GeckoConsole( 1855): Content JS LOG at app://system.gaiamobile.org/js/activities.js:12 in act_handleEvent: --> evt.type = mozChromeEvent
E/GeckoConsole( 1855): Content JS LOG at app://system.gaiamobile.org/js/activities.js:16 in act_handleEvent: --> detail.type = activity-choice
E/GeckoConsole( 1855): Content JS LOG at app://system.gaiamobile.org/js/activity_window_factory.js:16 in awm_debug: [ActivityWindowFactory][1325532923.04]handling launchapp
E/GeckoConsole( 1855): Content JS LOG at app://system.gaiamobile.org/js/activity_window_factory.js:16 in awm_debug: [ActivityWindowFactory][1325532923.105]handling mozChromeEvent

======= unsuccess =======
E/GeckoConsole( 1855): [JavaScript Error: "[Exception... "Component returned failure code: 0xc1f30001 (NS_ERROR_NOT_INITIALIZED) [nsIMessageSender.sendAsyncMessage]"  nsresult: "0xc1f30001 (NS_ERROR_NOT_INITIALIZED)"  location: "JS frame :: jar:file:///system/b2g/omni.ja!/components/SystemMessageInternal.js :: _sendMessageCommon :: line 620"  data: no]" {file: "jar:file:///system/b2g/omni.ja!/components/SystemMessageInternal.js" line: 620}]
E/GeckoConsole( 1855): [JavaScript Error: "[Exception... "Component returned failure code: 0xc1f30001 (NS_ERROR_NOT_INITIALIZED) [nsIMessageSender.sendAsyncMessage]"  nsresult: "0xc1f30001 (NS_ERROR_NOT_INITIALIZED)"  location: "JS frame :: jar:file:///system/b2g/omni.ja!/components/SystemMessageInternal.js :: _sendMessageCommon :: line 620"  data: no]" {file: "jar:file:///system/b2g/omni.ja!/components/SystemMessageInternal.js" line: 620}]
E/GeckoConsole( 1855): Content JS LOG at app://system.gaiamobile.org/js/activities.js:12 in act_handleEvent: --> evt.type = mozChromeEvent
E/GeckoConsole( 1855): Content JS LOG at app://system.gaiamobile.org/js/activities.js:16 in act_handleEvent: --> detail.type = activity-choice
E/audio_hw_primary(   89): Error: s_vbpipe_fd(-1) open failed, No such file or directory 
E/GeckoConsole( 1855): Content JS LOG at app://system.gaiamobile.org/js/activity_window_factory.js:16 in awm_debug: [ActivityWindowFactory][1325532977.878]handling mozChromeEvent

Looks like activity_window_factory.js does not receive "launchapp" event in this failed case. Any other help in detail, please let me know.

Comment 12

[Tim Guan-tin Chien [:timdream] (please needinfo)]

Ian, thank you very much!

This is now a system message bug then. Gene, I vaguely remember you were working on this API. Who I should needinfo too now for Tarako project?

Comment 13

[Gene Lian (I already quit Mozilla)]

This symptom is pretty much the same as the Bug 991023.

Comment 14

[Gene Lian (I already quit Mozilla)]

Hi Fabrice, I wonder this is caused by Bug 982491. System Message doesn't have any changes on 1.3T recently.

Comment 15

[Gene Lian (I already quit Mozilla)]

Could TAs please support to verify if this regression is caused by Bug 982491 (or not)? Thanks a lot!

Comment 16

[Jason Smith [:jsmith]]

(In reply to Gene Lian [:gene] (needinfo? encouraged) from comment #15)
> Could TAs please support to verify if this regression is caused by Bug
> 982491 (or not)? Thanks a lot!

We don't have enough builds to do windows on tarako right now.

Comment 17

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

let's track Bug 991023