980826 - [B2G] When export contacts to certain SIM card, system only stays in "export contacts" page without finishing it.

Status: VERIFIED FIXED

(Firefox OS Graveyard :: RIL, defect)

Description

[Enpei]

Attachment: time stamp: 17:22, 3/7

When export contacts, it only stays in "Export Contacts" page but no processing animation and won't finish. 

Take the SIM to other phone and check the contacts. In the test 4 contacts are selected but only one is exported.

* Build Number  
Fugu
Gaia      ef64aa12b65723a29b3d5dbe115820a557763fab
Gecko     f12cc94c28d3eadde2054dda6a19b072c6618ccd
BuildID   20140306053055
Version   30.0a1

Gaia      f692ee0e21ef4e25ec501d60226aa7da4b06ae60
Gecko     fcf711bbb0affff0925cb133c82073998bb46abf
BuildID   20140306041555
Version   28.0

* Reproduce Steps
1. Export 4 contacts to SIM 1.

* Expected Result
Process will be finished. 

* Actual Result
Keep in export contact page with no processing animation. Then take the SIM to other phone which can read SIM contacts, turns out 1 contact is exported to SIM card.

* Occurrence rate
100%

Worked build:(verified in bug 927893)
Fugu
Gaia      ee25b0e45649d2955f340ce4f71ad55712dd0fab
Gecko     913cf2b92845441c9578787362ddad6f2ac15df6
BuildID   20140121095108
Version   28.0a2

Comment 1

[Enpei]

After some experiments, I found that 3 SIM card I have (FET, TWN, Vibo) only Vibo can have contacts exported while it's in SIM 1 slot, not SIM 2. This is very odd. 

Due to Fugu is not built from pvt, the build records are not complete. The earliest build now I can get is this one: 
Gaia      26e8a950f3ac703310ee8efcb741ff2418e9d190
Gecko     0d2a915d0b66dba35ee9936bdeafef820a5c3357
BuildID   20140216061608
Version   28.0

Which also have this bug with FET and TWN SIM cards, and Vibo if it's in SIM 2 slot. Also I've updated base image from 2/27, it's hard to tell whether this bug is still reproducible with previous one since we didn't save early image(workd build in description is with old base image and that was the build used to verify export user story). 

So if developer can check the log first which can be a great help to see if this is vendor's bug or not.

Comment 2

[Enpei]

Attachment: fugu_980826.txt

This log covered actions below:
1. SIM 1 is Vibo, SIM 2 is TWN SIM. 
2. Export 3 contacts to SIM 1, contacts can be exported. 
3. Export 3 contacts to SIM 2, it stuck in export contact page. Time stamp:12:12, 3/10.

Comment 3

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

Hi Enpei, the log doesn't contain RIL log, could you help to capture again with RIL log enabled? Thank you.

Comment 4

[Enpei]

Attachment: fugu_980826_RIL.txt

Hi Edgar, can you help to check this log again? Time stamp: 15:03, 3/10. Thank you.

Comment 5

[Enpei]

(In reply to Enpei from comment #1)
> Which also have this bug with FET and TWN SIM cards, and Vibo if it's in SIM
> 2 slot. 
I would like to correct this comment. After double check again, I can export contacts to Vibo SIM card even it's in slot 1(for SIM 2).

Comment 6

[Enpei]

Hi Jason, 

Actually I can reproduce the bug on Buri 1.3 and 1.4, but I didn't use bug 957051 for tracking is because that one is focusing on QC's problem. Now in this bug, device can only export contacts to certain SIM(in my case Vibo carrier in Tawian), which is the same behavior on Buri. Should we find regression window on Buri again since it might not be QC ril bug at all?

Comment 7

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

(In reply to Enpei from comment #4)
> Created attachment 8388352 [details]
> fugu_980826_RIL.txt
> 
> Hi Edgar, can you help to check this log again? Time stamp: 15:03, 3/10.
> Thank you.

From the RIL log, I did not see any log about exporting contact to sim2. :( 
Looks like gaia did not call the api to export contact to sim2.

Comment 8

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

(In reply to Edgar Chen [:edgar][:echen] from comment #7)
> (In reply to Enpei from comment #4)
> > Created attachment 8388352 [details]
> > fugu_980826_RIL.txt
> > 
> > Hi Edgar, can you help to check this log again? Time stamp: 15:03, 3/10.
> > Thank you.
> 
> From the RIL log, I did not see any log about exporting contact to sim2. :( 
> Looks like gaia did not call the api to export contact to sim2.

Oh, please ignore my previous comment. I did see log for sim2.
I think the problem is the exporting request doesn't receive either success or error callback.

===
03-10 15:03:47.588 I/Gecko   (  107): -*- RadioInterfaceLayer: Received 'RIL:UpdateIccContact' message from content process
03-10 15:03:47.588 I/Gecko   (  107): RIL Worker[1]: Received chrome message {"requestId":"id{d16be7bf-9087-43a4-8d71-8d933f39d51e}","contactType":"adn","contact":{"alphaId":"Agdgjuhgui","number":"258523369","contactId":"null"},"pin2":null,"rilMessageToken":8,"rilMessageType":"updateICCContact"}
===

Comment 9

[Francisco Jordano [:arcturus] [:francisco]]

Hei folks,

can someone confirm if it's a gecko or gaia issue, right now don't have a DSDS device to test, will have it in two days, so if someone could dig a bit, I will really appreciate it :)

Thanks,
F.

Comment 10

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

(In reply to Francisco Jordano [:arcturus] from comment #9)
> Hei folks,
> 
> can someone confirm if it's a gecko or gaia issue, right now don't have a
> DSDS device to test, will have it in two days, so if someone could dig a
> bit, I will really appreciate it :)
> 
> Thanks,
> F.

Hi Francisco, I think it is a gecko issue. Gecko did not dispatch either success or error event.
And this bug seems a sim-dependent issue, not related to DSDS. Because I can repro this issue on single sim device (unagi) with the same sim card.

I will take this bug. Thank you. :)

Comment 11

[Francisco Jordano [:arcturus] [:francisco]]

Thanks Edgar for the reply, I'm changing the component to reflect your comment.

Comment 12

[Jason Smith [:jsmith]]

(In reply to Enpei from comment #6)
> Hi Jason, 
> 
> Actually I can reproduce the bug on Buri 1.3 and 1.4, but I didn't use bug
> 957051 for tracking is because that one is focusing on QC's problem. Now in
> this bug, device can only export contacts to certain SIM(in my case Vibo
> carrier in Tawian), which is the same behavior on Buri. Should we find
> regression window on Buri again since it might not be QC ril bug at all?

Sure.

Comment 13

[Enpei]

Buri 1.3, 1.4 also have this bug, remove DSDS from summary.

Comment 14

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

Attachment: WIP, Patch, v1

I am working on test case now, and trying to find out other potential issue if any.

Comment 15

[Jason Smith [:jsmith]]

Going to hold off on the window if there's already a patch here in progress.

Comment 16

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

Attachment: Part 1: updateContactFieldType2() doesn't trigger either onsuccess or onerror callback in some case, v2

Hi Yoshi, could you help to review this patch? Test cases will be in part2. Thank you.

Comment 17

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

Attachment: Part 2: xpcshell test, v1

Add Tests for the case that recordId in IAP is 0xff.

Comment 18

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

Hi Yoshi,

Could you help to review this?
I added tests for the case that recordId in IAP is 0xff.

Thank you

Comment 19

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

Comment on attachment 8389655 [details] [diff] [review]
Part 2: xpcshell test, v1

Review of attachment 8389655 [details] [diff] [review]:
-----------------------------------------------------------------

::: dom/system/gonk/tests/test_ril_worker_icc.js
@@ +1729,5 @@
>      recordHelper.readIAP = function(fileId, recordNumber, onsuccess, onerror) {
>        do_check_eq(fileId, IAP_FILE_ID);
>        do_check_eq(recordNumber, ADN_RECORD_ID);
> +      onsuccess([(aHaveIapIndex) ? EMAIL_RECORD_ID : 0xff,
> +                 (aHaveIapIndex) ? ANR0_RECORD_ID : 0xff]);

aHaveIapIndex ? [...] : [0xff, 0xff]

Comment 20

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

Attachment: Part 2: xpcshell test, v2, r=allstars.chh

Address review comment #19.

Comment 21

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

Try result: https://tbpl.mozilla.org/?tree=Try&rev=a4d42f43d7ef

Comment 22

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

Attachment: [v1.3] Part 2: xpcshell test, v2, r=allstars.chh, a=1.3

Patch for v1.3 branch.

Comment 23

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

https://hg.mozilla.org/integration/b2g-inbound/rev/c55ac673a750
https://hg.mozilla.org/integration/b2g-inbound/rev/b2498cfb7e36

Comment 24

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

1.3 try result: https://tbpl.mozilla.org/?tree=Try&rev=f2ea31e4110c

Comment 25

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/c55ac673a750
https://hg.mozilla.org/mozilla-central/rev/b2498cfb7e36

Comment 26

[Enpei]

Verified on Fugu v1.4 first, result is passed. 

Gaia      29a990ac0347c057ef6998cb42e4548768a18559
Gecko     20396557b9a738da44b33f58e6109efe1d5914af
BuildID   20140314053055
Version   30.0a1

Hi Ryan, can this be uplifted to 1.3 since it's 1.3+?

Comment 27

[Ryan VanderMeulen [:RyanVM]]

As has been the policy for the past few weeks now, all patches need explicit approval to land on 1.3 You need to request approval-mozilla-b2g28 on the patches for them to be considered by Fabrice & RelMan.

Comment 28

[Enpei]

Thank you, Ryan.

Hi Fabrice,

Can you check if this bug can be uplifted to 1.3? We can't forecast what kind of SIM card users will use, but it will impact basic contact export function which should be fixed from my opinion.

Comment 29

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

Comment on attachment 8389590 [details] [diff] [review]
Part 1: updateContactFieldType2() doesn't trigger either onsuccess or onerror callback in some case, v2

NOTE: This flag is now for security issues only. Please see https://wiki.mozilla.org/Release_Management/B2G_Landing to better understand the B2G approval process and landings.

[Approval Request Comment]
Bug caused by (feature/regressing bug #): Bug 951177.
User impact if declined: Contact could not be exported into sim card if the sim card has a specific contact field type.
Testing completed: Yes.
Risk to taking this patch (and alternatives if risky): No.
String or UUID changes made by this patch: No.

Comment 30

[Edgar Chen [:edgar] (OOO untils 09/01, slow response)]

Comment on attachment 8390342 [details] [diff] [review]
[v1.3] Part 2: xpcshell test, v2, r=allstars.chh, a=1.3

NOTE: This flag is now for security issues only. Please see https://wiki.mozilla.org/Release_Management/B2G_Landing to better understand the B2G approval process and landings.

[Approval Request Comment]
Bug caused by (feature/regressing bug #): Bug 951177.
User impact if declined: Contact could not be exported into sim card if the sim card has a specific contact field type.
Testing completed: Yes.
Risk to taking this patch (and alternatives if risky): No.
String or UUID changes made by this patch: No.

Comment 31

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-b2g28_v1_3/rev/d3812d9d872c
https://hg.mozilla.org/releases/mozilla-b2g28_v1_3/rev/e25e38b7cda4

Comment 32

[Enpei]

Verified on Buri.

Gaia      2ea2aab306bd1c941719160cdcb49ee9d755dc17
Gecko     https://hg.mozilla.org/releases/mozilla-b2g28_v1_3/rev/cf2042938526
BuildID   20140317164002
Version   28.0