975793 - [Sora][STK]It will exit system menu screen after display text cleared.

Status: RESOLVED WONTFIX

(Firefox OS Graveyard :: Vendcom, defect, P1)

Description

[sync-1]

Created an attachment (id=633150)
 adb log
 
 DEFECT DESCRIPTION:
 It will exit system menu screen after display text cleared.
 
  REPRODUCING PROCEDURES:
 1.MS load simulated SIM card and SATK application which can send "display text" procative SIM command.
 2.Run test case STM01004->System menu,enter to a system menu (eg: Device information) wait 'display text' coming.
 3.When 'display text' screen displays,press 'OK' to clear it,will find it will exit system menu and enter to SAT menu->KO. 
 
 Note:v1.1 buri has no this issue.
 
  EXPECTED BEHAVIOUR:
 After clear display text, it should also stay at system menu.
 
  ASSOCIATE SPECIFICATION:
 
  TEST PLAN REFERENCE:
 
  TOOLS AND PLATFORMS USED:
 V108.
 
  USER IMPACT:
 Middle.
 
  REPRODUCING RATE:
 2/2.
 
  For FT PR, Please list reference mobile's behavior:

Comment 1

[sync-1]

Attachment: adb log

Comment 2

[Fernando R. Sela (no CC, needinfo please) [:frsela]]

Hi,

Can you attach traces with GAIA DEBUG TRACES enabled please?

BTW, STK_CMD_DISPLAY_TEXT should sent STK_RESULT_OK if user press OK, if timeout STK_RESULT_NO_RESPONSE_FROM_USER and if back is pressed, STK_RESULT_BACKWARD_MOVE_BY_USER and if close is pressed STK_RESULT_UICC_SESSION_TERM_BY_USER.

Comment 3

[xiaokang.chen]

Hi Fernando,

I can't reproduce this bug with real card. I will upload the log if I got it from qa.

Thanks a lot!

Comment 4

[Jason Smith [:jsmith]]

Waiting on additional input to understand impact before making a blocking decision.

Comment 5

[bhavana bajaj [:bajaj]]

clearing the nomination for now, feel free to renom(1.3 ?) with additional input as requested as we are still not convinced that this could be a ship blocker issue.

Comment 6

[xiaokang.chen]

Attachment: log.rar

new log from our lab

Comment 7

[Fernando R. Sela (no CC, needinfo please) [:frsela]]

(In reply to xiaokang.chen from comment #6)
> Created attachment 8385021 [details]
> log.rar
> 
> new log from our lab

Thank you.

In a first review, I saw that all STK response sent are with resultCode=0 which means STK_RESULT_OK.

I also saw this

I/GeckoDump( 1071): [settings] sendStkResponse to command: {"commandNumber":1,"typeOfCommand":36,"commandQualifier":0,"options":{"title":"STM01004","items":[{"identifier":1,"text":"System menu"},{"identifier":2,"text":"SAT menu"}],"presentationType":0}}
I/GeckoDump( 1071): [settings] sendStkResponse -- # response = {"resultCode":0,"itemIdentifier":"1"}
I/setupmenu--------------------( 1071): Icc::SendStkResponse
I/Gecko   ( 1071): -----------------response.command = 36
I/Gecko   (  296): -*- QCMessageManager_QC_B2G: receiveMessage: RIL:SendStkResponse arrived from content process
I/Gecko   (  296): -*- QCMessageManager_QC_B2G: Forwarding message to clientId = 0
D/use-Rlog/RLOG-RIL_QC_B2G(  296): [SUB0] [0108]> STK_SEND_TERMINAL_RESPONSE 810301240002028281830100900101
I/Gecko   (  296): -*- [SUB0] QCContentHelper_QC_B2G: receiveMessage: 'RIL:SendStkResponse'
D/use-Rlog/RLOG-RIL_QC_B2G(  296): [SUB0] [0108]< STK_SEND_TERMINAL_RESPONSE
E/RILQ    (  268): (0/268): RIL[0][event] qcril_gstk_qmi_copy_and_save_proactive_cmd: Invalid ril_unsol_resp parameters, cannot save
D/use-Rlog/RLOG-RIL_QC_B2G(  296): [SUB0] [UNSL]< UNSOL_STK_SESSION_END
I/Gecko   (  296): -*- [SUB0] QCContentHelper_QC_B2G: sendMessage to content process: RIL:StkSessionEnd{  }
I/Gecko   (  296): -*- QCMessageManager_QC_B2G: Received sendTargetMessage with topic='icc', message='RIL:StkSessionEnd'

A resultCode 0 is sent by Gaia (STK_RESULT_OK) but at gecko level a STK_SEND_TERMINAL_RESPONSE is managed. Not sure if it's normal.

Adding Yoshi for double check ;)

Comment 8

[Yoshi Cheng-Hao Huang [:allstars.chh][:allstarschh][:yoshi]]

From the log it's form partner's RIL, not Moz RIL.

Also to answer frsela, TERMINAL_RESPONSE is a normal RIL requests, it replies the result of the proactive command, that's why the term 'response' is used.

BTW, I spotted some error

*E/RILQ    (  268): (0/268): RIL[0][event] qcril_gstk_qmi_copy_and_save_proactive_cmd: Invalid ril_unsol_resp parameters, cannot save*

I guess that's why of SessionEnd is happened. 
But it's from partner's RIL, I don't know what exactly went wrong here.

Comment 9

[Noemí Freire (:noemi)]

(In reply to Yoshi Huang[:allstars.chh] from comment #8)
> From the log it's form partner's RIL, not Moz RIL.
> 
> Also to answer frsela, TERMINAL_RESPONSE is a normal RIL requests, it
> replies the result of the proactive command, that's why the term 'response'
> is used.
> 
> BTW, I spotted some error
> 
> *E/RILQ    (  268): (0/268): RIL[0][event]
> qcril_gstk_qmi_copy_and_save_proactive_cmd: Invalid ril_unsol_resp
> parameters, cannot save*
> 
> I guess that's why of SessionEnd is happened. 
> But it's from partner's RIL, I don't know what exactly went wrong here.

Carol, could you please take a look at it?. Thanks!

Comment 10

[Carol Yang [:cyang]]

Please open an SR so we can provide support there. Please also make sure to attach the QXDM logs for this issue.

Comment 11

[xiaokang.chen]

Dear Carol,
The SR is 01470955. And I have attach qxdm logs there.

Dear Yoshi,
Did you mean that the "SessionEnd" caused "exit system menu and enter to SAT menu after confirm the display text"?

Comment 12

[Yoshi Cheng-Hao Huang [:allstars.chh][:allstarschh][:yoshi]]

(In reply to xiaokang.chen from comment #11)
> Dear Yoshi,
> Did you mean that the "SessionEnd" caused "exit system menu and enter to SAT
> menu after confirm the display text"?

No, it's caused by the error from partner's RIL.

Comment 13

[xiaokang.chen]

(In reply to Yoshi Huang[:allstars.chh] from comment #12)
> (In reply to xiaokang.chen from comment #11)
> > Dear Yoshi,
> > Did you mean that the "SessionEnd" caused "exit system menu and enter to SAT
> > menu after confirm the display text"?
> 
> No, it's caused by the error from partner's RIL.

I can't figure out why "exit system menu and enter to SAT menu after confirm the display text". Did gaia got some wrong message from lower layer? Which message caused the enter to SAT menu again? 

Then I will check did the wrong message relate to the errors from ril. Thanks a lot!

Comment 14

[xiaokang.chen]

Hi Fernando,
After check the vendor code, I don't think 

"qcril_gstk_qmi_copy_and_save_proactive_cmd: Invalid ril_unsol_resp parameters, cannot save"

is the reason for this bug. It's not a fatal error and didn't affect the behavior of gaia. The key point is why "enter to SAT menu after user confirm display text". Did gaia get some unappropriate message ?

What do you think?

Comment 15

[Fernando R. Sela (no CC, needinfo please) [:frsela]]

(In reply to xiaokang.chen from comment #14)
> Hi Fernando,
> After check the vendor code, I don't think 
> 
> "qcril_gstk_qmi_copy_and_save_proactive_cmd: Invalid ril_unsol_resp
> parameters, cannot save"
> 
> is the reason for this bug. It's not a fatal error and didn't affect the
> behavior of gaia. The key point is why "enter to SAT menu after user confirm
> display text". Did gaia get some unappropriate message ?
> 
> What do you think?

It's possible, but I didn't see any strange in the log traces ... In order to have a clearer view of the issue, can you upload a little video showing the error? or steps to reproduce it ...

Comment 16

[xiaokang.chen]

Dear Fernando,
I can't reproduce with real card. I will attach the video If I got it from the qa. Thanks a lot!

Comment 17

[xiaokang.chen]

Attachment: VID_20140311_103431.part1.rar

separate the video because of the attachment limitation.
part 1

Comment 18

[xiaokang.chen]

Attachment: VID_20140311_103431.part2.rar

Comment 19

[xiaokang.chen]

Attachment: VID_20140311_103431.part3.rar

Comment 20

[xiaokang.chen]

Attachment: VID_20140311_103431.part4.rar

Comment 21

[xiaokang.chen]

Attachment: VID_20140311_103431.part5.rar

Comment 22

[xiaokang.chen]

Attachment: VID_20140311_103431.part6.rar

Dear Fernando,
This is the video from QA. You can download all the 6 parts into one directory and then extract there.

What do you think ?

Comment 23

[xiaokang.chen]

Dear Fernando,
I found this in Settings/js/icc.js
    icc.onstksessionend = function handleSTKSessionEnd(event) {
      updateMenu();
      Settings.currentPanel = '#icc'; //this is the reason why this bug occur
    };

And I found there's no "Settings.currentPanel = '#icc';" in v1.1. So I just delete this code in our side.

Comment 24

[Fernando R. Sela (no CC, needinfo please) [:frsela]]

(In reply to xiaokang.chen from comment #23)
> Dear Fernando,
> I found this in Settings/js/icc.js
>     icc.onstksessionend = function handleSTKSessionEnd(event) {
>       updateMenu();
>       Settings.currentPanel = '#icc'; //this is the reason why this bug occur
>     };
> 
> And I found there's no "Settings.currentPanel = '#icc';" in v1.1. So I just
> delete this code in our side.

I saw the video.

The line:

Settings.currentPanel = '#icc';

Is used to change the Settings view to the ICC panel which should be updated with the MainMenu (called before this -> updateMenu() )

So, this is the correct way of working by the decision taken at bug

https://bugzilla.mozilla.org/show_bug.cgi?id=920983

Should we change this behavior?

Comment 25

[xiaokang.chen]

So 975793 and 920983 affected each other. Shall we just find another better solution??

Comment 26

[Fernando R. Sela (no CC, needinfo please) [:frsela]]

(In reply to xiaokang.chen from comment #25)
> So 975793 and 920983 affected each other. Shall we just find another better
> solution??

The bug 920983 was a TCL requirement to show the STK menu when a session end message was received.

Now within this bug the requirement is the opposite, so we would need a final decision about the desirable behavior from your side.

Please, let us know the final decision and we'll proceed accordingly.

Comment 27

[xiaokang.chen]

(In reply to Fernando R. Sela (no CC, needinfo please) [:frsela] from comment #26)
> (In reply to xiaokang.chen from comment #25)
> > So 975793 and 920983 affected each other. Shall we just find another better
> > solution??
> 
> The bug 920983 was a TCL requirement to show the STK menu when a session end
> message was received.
> 
> Now within this bug the requirement is the opposite, so we would need a
> final decision about the desirable behavior from your side.
> 
> Please, let us know the final decision and we'll proceed accordingly.

I'll talked to the former developer of STK about this and give you the result here. Thanks a lot!

Comment 28

[Wesly Huang (EPM)]

link all Fire C (codename: Sora) bugs to a meta one.

Comment 29

[BMO Automation]

Firefox OS is not being worked on