860836 - Error importing contacts from SIM card

Status: VERIFIED FIXED

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

Description

[David Palomino [:dpalomino]]

Attachment: log file described before

Buildid "20130321070205", device: ikura
gecko commit: b5183c99228bdc5be33340e359efd1b4f0859e92 
gaia commit: 577d13088ebdbd353d13910d3317e713a140415b

Reported in certification as a blocker (for both LatAm and Spain). Adding dep and nominating. 

Importing from a SIM card does not succeed, and it the process is stopped. 

In log attached, the steps followed are: 
1. User a SIM card with big amount of contacts (148 in the log)
2. Import contacts from SIM card
3. Process start ok, progressing bar goes on fine
4. At 34/148 imported contacts, the process is frozen and does not continue (it's importing 34th contact for ever)


1. Enter a SIM card with many contacts (e.g. 148 entries filled)
2.- Import contacts from SIM to phone
3.- It starts copying but progress bar is frozen before finishing (e.g. 34/148)
4.- Wait 5 minutes in this screen to assure that the importing is stopped
5.- Press the home button to exit from the contacts app and close it
6.- Enter again in contacts app and check the contacts copied (33 contacts)

It's possible to exit from the app. 33 contacts has been imported.

Comment 1

[David Palomino [:dpalomino]]

Attachment: screenshot

Comment 2

[David Palomino [:dpalomino]]

Attachment: contacts from the sim card

Comment 3

[Gregor Wagner [:gwagner]]

David, are you in Madrid next week? I can't see anything obvious in the logcat but it would be good to use your SIM card with our debug flag enabled.

Comment 4

[Anshul]

Nivi, this was run using commercial RIL so wondering if you could take a look?

Comment 5

[Anshul]

David, which AU was used for the test?

Comment 6

[David Palomino [:dpalomino]]

Hi Anshul, 

According to ZTE info: 
AU_LINUX_GECKO_ICS_STRAWBERRY_V1.01.00.01.19.046

Comment 7

[Nivi]

Hi David/Anshul,
I will take a look and see what's going on here. Can they reproduce the issue on the lastest AU as well?

Nivi.

Comment 8

[Nivi]

Hi,

I did try to import contacts from a card with empty contacts in between and was able to do it but I had only 11 contacts.(tested with latest code.)
Could someone send me the radio logs along with the main log? Please use - |adb logcat -v threadtime -b main -b radio| to collect all logs.

Also, the last column (Column D) from the excel sheet having the contacts data doesn't have a header? Does anyone know what that means? I see that for the 34th contact, the value for column D is Si` (yes) whereas all other records have the value no. That's one difference I notice. Don't know if that's an issue? Could someone please confirm.

Nivi.

Comment 9

[Vivien Nicolas (:vingtetun) (:21) - (NOT reading bugmails, needinfo? please)]

Does not seems like a Gaia bug. Moving to 'general' to reflect that.

Comment 10

[David Palomino [:dpalomino]]

(In reply to Gregor Wagner [:gwagner] from comment #3)
> David, are you in Madrid next week? I can't see anything obvious in the
> logcat but it would be good to use your SIM card with our debug flag enabled.

Hi Gregor!

Yes, I'm in Madrid now. The SIM card used is from testing house, it's not our SIM. However testing house office is also in Madrid. 

Thks!
David

Comment 11

[alberto.pastor]

Can we ask at least what's the content (in plain text) of that contact? What does the last column in the excel mean?

Comment 12

[Nivi]

Could someone send me the adb logs (main and radio)? Since the export contacts screen already shows the total number of contacts (148) and gets stuck importing at record no 34, I would say that RIL (commercial) has already sent the complete list of contacts to gaia. It might be a gaia bug. Vivien, how did you rule this out as not being a gaia bug? I am curious.(In reply to Vivien Nicolas (:vingtetun) (:21) from comment #9)
> Does not seems like a Gaia bug. Moving to 'general' to reflect that.

Comment 13

[Daniel Coloma:dcoloma]

Dietrich, I think comment 12 raise some concerns about whether this is a commercial bug or not. I'd suggest we remove NPOTB until it is confirmed.

David, can we get that SIM Card?

Comment 14

[Anshul]

I will be in Madrid for two more days so would be happy to help debug the issue.

Comment 15

[Nivi]

Anshul, Did you get a chance to get hold of the SIM card?

Comment 16

[Anshul]

No I didn't. David, can you please help me find the SIM card?

Comment 17

[David Palomino [:dpalomino]]

Hi, 

As commented f2f, SIM card is not available (it was not tested in Madrid). We're requesting a dump of this SIM card, I hope to have it soon.

Comment 18

[Andrew Overholt [:overholt]]

ni?(dpv) while we await the dump.

Comment 19

[David Palomino [:dpalomino]]

Hi, 

We've requested it again, but dump not received yet. 

Sorry about that. 
David

Comment 20

[Kai-Chih Hu [:khu]]

Reviewed on April 23th: 
    Asking for new logs, but no reply yet. 
    Can't reproduce this issue in house. AT4 can reproduce this issue. 
    Will provide the logs by the end of tomorrow. Also suggest to check the latest build. 
    Agreed to have this fixed before the 3rd certification.

Comment 21

[David Palomino [:dpalomino]]

Attachment: logs crash importing contacts

Hi, 

Please, find attached the new log about the issue reported. 

Thanks!
David

Comment 22

[Nivi]

Hi David,

Is this log from commercial RIL? Seems like it's not. Could you please confirm?
Also, if you are using reference ril, do you see the previous behavior (import contacts screen stuck for contact 34) or a crash?

Nivi.

Comment 23

[Kai-Chih Hu [:khu]]

Hi, Nivi, may I know if there has any specific reason that you removed the target milestone? Thanks.

Comment 24

[Anshul]

Khu, that might have been a mistake. I added it back.

Comment 25

[David Palomino [:dpalomino]]

Hi, 

Yes, sorry about that, we forgot to comment this. A new log was requested with Moz RIL just to check wether the issue was related to RIL or gaia/gecko. As it is reproduced with both RIL, I guess it's not caused by RIL. 

Do we need any other specific logs to investigate further this issue?

Thanks, 
David

Comment 26

[Daniel Coloma:dcoloma]

Alberto, Francisco, can you have a look at the logs? Based on comment 25 it seems not to be a RIL issue.

Comment 27

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

Hi folks, after checking the last log provided and there is no a single line regarding gaia, everything is related to the RIL.

Will provide a version of gaia with verbose comments when SIM import, at least to see if we are able to detect the error in the frontend.

Thanks!
F.

Comment 28

[David Palomino [:dpalomino]]

Hi Francisco, 

In the first log attached it should be included, aren't there?

Please, let us know if we need more logs here. We don't have the SIM card where they are reproducing the bug, so we've to ask to testhouse to collect new logs if needed. 

Thanks!
David

Comment 29

[Kai-Chih Hu [:khu]]

May I know is there anyone knows where the issue locates? It looks like it's in RIL. Anshul, could we have your comments here? Thanks!

Comment 30

[Anshul]

Khu, at this point it doesn't seem like the issue is in RIL as the issue is seen both with Moz ril and commercial RIL. We haven't been able to replicate the exact scenario and so not able to debug any further. 

I am hoping Francisco can shed some more light on the issue.

Comment 31

[Kyle Machulis [:qdot] [:kmachulis] (INACTIVE)]

Would it be a good idea to just get a USB SIM reader/writer and write a ton of contacts to a SIM ourselves to test this?

Comment 32

[Kai-Chih Hu [:khu]]

(In reply to Anshul from comment #30)
> Khu, at this point it doesn't seem like the issue is in RIL as the issue is
> seen both with Moz ril and commercial RIL. We haven't been able to replicate
> the exact scenario and so not able to debug any further. 
> 
Thanks! Ken, do you have any comments on this bug? I think the 1st step would be: try to reproduce this issue in house or check the attached logs.

Comment 33

[Kai-Chih Hu [:khu]]

Al, can you help to reproduce this issue? Thanks.

Comment 34

[Ken Chang[:kenkai]]

D/RILC    (  117): qcril_uim_read_record_resp: token=125 qmi_err_code=0 
D/RILC    (  117): RIL_SIM_IO_Response: sw1=0x90 sw2=0x0 data=434f4e545241544f2035ffffffff068136887468f5ffffffffffffff
....
D/RILC    (  117): qcril_uim_read_record_resp: token=374 qmi_err_code=0 
D/RILC    (  117): RIL_SIM_IO_Response: sw1=0x90 sw2=0x0 data=ffffffffffffffffffffffffffffffffffffffffffffffffffffffff
I saw RILC had tried to read all contacts(374-125+1=250) from modem, which means that RIL had received the ICC Contacts without error.

Need David's help to provide what Gregor requested in comment 3 to know if contact manager receives those contacts from RIL.
Gregor, did you get any clue in Madrid work week?

Comment 35

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

Hi,

just pushed a branch, from master, that adds lots of logging to the sim import.

Here you can download it:

https://github.com/arcturus/gaia/tree/sim_import_log

Please, can someone with access to the SIM flash a phone with this branch and send the logcat generated?

Thanks,
F.

Comment 36

[Juan Perez-Bedmar [:juanpbf]]

(In reply to Francisco Jordano [:arcturus] from comment #35)
> Hi,
> just pushed a branch, from master, that adds lots of logging to the sim
> import.
> Here you can download it:
> https://github.com/arcturus/gaia/tree/sim_import_log
> 
> Please, can someone with access to the SIM flash a phone with this branch
> and send the logcat generated?
I will request my colleagues to create a specific build with this branch
  --> [e-mail], could you please do this?

Once I get the build I will request the lab to repeat it and take logs.

thanks

Comment 37

[David Palomino [:dpalomino]]

> Need David's help to provide what Gregor requested in comment 3 to know if
> contact manager receives those contacts from RIL.
> Gregor, did you get any clue in Madrid work week?

Hi Ken, 

This issue was tested by the testhouse, in the offices out of Madrid. I'll try to request them to send us the sim card to Madrid, so we could test this easier. 

Thanks!
David

Comment 38

[Jason Smith [:jsmith]]

I'm not entirely sure what anyone on QA can do here - there's not enough information to move forward here with any level of QA analysis. I would need the following information from David:

- What type of SIM reproduced this bug?
- Does TEF QA (i.e. in Madrid) have the ability to get access to a SIM of this type?

Given that this was tested in Madrid with a particular SIM, my first guess is that we should ask someone from TEF QA to look into this when we know the above information.

Comment 39

[Carlos Martínez Toral [:carlosmartinez]]

Adding the logcat generated while importing the contacts on a inari device with:

ZTE build id: 20130506003326
gaia commit: eeb713b657
QC RIL: V1.01.00.01.019.094

Comment 40

[Carlos Martínez Toral [:carlosmartinez]]

(In reply to Carlos Martínez Toral [:carlosmartinez] from comment #39)
> Adding the logcat generated while importing the contacts on a inari device
> with:
> 
> ZTE build id: 20130506003326
> gaia commit: eeb713b657
> QC RIL: V1.01.00.01.019.094

I deny my own words, testing was finally done on a inari device, using a ZTE Eng build (P752D04_TME_ENG_FOR_PARTENER_20130426) and gaia coming from https://github.com/arcturus/gaia/tree/sim_import_log.

ZTE build id: 20130426093145
gaia commit: 28a84f1
QC RIL: V1.01.00.01.019.085

Logcat added.

Comment 41

[Carlos Martínez Toral [:carlosmartinez]]

Attachment: SIM import log with extra debug info from arcturus branch

Comment 42

[Jason Smith [:jsmith]]

I'm guessing your comment implies that you've reproduced this on a 4/26 build, right? If so, then the qawanted request has been met to confirm a reproduction.

Comment 43

[Gregor Wagner [:gwagner]]

That might be it:
["PRIVE70"],"honorificPrefix":null,"givenName":null,"additionalName":null,"familyName":null,"honorificSuffix":null,"nickname":null,"email":null,"photo":null,"category":null,"adr":null,"tel":[{"type":null,"value":"+0988782456","carrier":null}],"org":null,"jobTitle":null,"bday":null,"note":null,"impp":null,"url":null,"anniversary":null}
E/GeckoConsole(  878): [JavaScript Error: "formats is undefined" {file: "resource://gre/modules/PhoneNumber.jsm" line: 80}]

Comment 44

[Gregor Wagner [:gwagner]]

Attachment: patch

Comment 45

[Ben Turner (not reading bugmail, use the needinfo flag!)]

Comment on attachment 745941 [details] [diff] [review]
patch

Review of attachment 745941 [details] [diff] [review]:
-----------------------------------------------------------------

Ok.

Comment 46

[Gregor Wagner [:gwagner]]

(In reply to Carlos Martínez Toral [:carlosmartinez] from comment #41)
> Created attachment 745847 [details]
> SIM import log with extra debug info from arcturus branch

Could you try with my fix?

Comment 47

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/projects/birch/rev/0af678fcaeef

Note that the various uplift queries ignore bugs that have NPOTB in the whiteboard, so you may want to remove that if you expect this to end up on b2g18.

Comment 48

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/0af678fcaeef

Comment 49

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-b2g18/rev/db94ce1cb17b
https://hg.mozilla.org/releases/mozilla-b2g18_v1_0_1/rev/7e9e10889942

Comment 50

[Carlos Martínez Toral [:carlosmartinez]]

Finally I´ve been able to try Gregor´s patch and seems to be working fine, although I don´t have the original SIM anymore, I´ve created a contact with same phone numbers.

Comment 51

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

@Carlos, the same phone number in a version without Gergor's patch has this problem?

Comment 52

[Gregor Wagner [:gwagner]]

I got the problematic number from the excel sheet and added a testcase for it.
I can't remember the exact number but it was one from:
+TestProperties("+0988782456");
+TestProperties("[phone]");
+TestProperties("[phone]");
+TestProperties("[phone]");
+TestProperties("[phone]");

Comment 53

[John Hammink]

Verified fixed on 5/13 partner build