Tested with: Gaia: 3ad4e525a43a44b3c6eb905304fc0648ca5fd9b8 Gecko: 74bff8c0506e6ff91666095b432fbc6aed9621fe build: 20121206 (Unagi) When testing with a Dual SIM card (two different subscriptions in one single SIM card), change of subscription through STK menu is currently not working. - Settings > STK > Dual > Subscription (to see which subscription is active), is working fine - Settings > STK > Dual > Change (to change from one subscription to the other), is not working right now. Behaviour is: 1. Enter "Settings > STK > Dual > Change" 2. In theory, order the change 3. No PIN is required 4. "Settings > STK" is presented 5. No change in subscription has been performed 6. Device continue working as usual (i.e. possible to browse through menus, open apps, etc) 7. After 70-150 (variable) seconds, a kind of "reset" appears a PIN code is requested again I'm not sure if step#7 is really a reset, or maybe the action ordered in step#2 that takes 80 seconds to complete. Apart from the delay, I'm not sure if it'd better to show to the user a kind of "work in progress..." message, just to get some feedback that something is ongoing after ordering the change. As Dual SIM card was addressed to V1, setting the bug as candidate for bb+. Thanks! David

Is this really required for certification? If so, it's a blocker so I'm marking it as such. If it's not required for certification, we can't block on it.

Hi Andrew, Yes, this is really required by OBs. Please, take into account that customers that have Dual SIM cards are generally the most valuable ones, and this is quite common for people with corporate SIM cards. If you need any further clarification, please, let us know. Thanks! David

Thanks, David. Yoshi, are you the best owner here? Please let us know if your plate is already full.

David, can you help to enable debug message for RIL ? in /system/b2g/defaults/pref/user.js pref("ril.debugging.enabled", false) -> pref("ril.debugging.enabled", true) and install the gaia again or adb remount adb pull /system/b2g/defaults/pref/user.js user.js [update user.js with 'pref("ril.debugging.enabled", true);'] adb push user.js /system/b2g/defaults/pref/user.js adb shell reboot

Hi, Sorry for the delay answering. I've retested this with today's build (20121217), Moz RIL. Behaviour remains the same. I've uploaded (comments #5 and #6) two logs with ril.debugging enabled. Steps followed: 1. Turn on the device 2. Introduce PIN code (for one of the subscription) 3. Check line through STK menu (Settings > carrier > Dual > Line) >> This is ok. 4. Order change of subscription via STK menu (Settings > carrier > Dual > Change) 5. Parent directory is shown (Settings > carrier > Dual) and nothing seems to happen Let us know if this helps. Thanks! David

(In reply to David Palomino [:dpv] from comment #2) > Yes, this is really required by OBs. Please, take into account that > customers that have Dual SIM cards are generally the most valuable ones, and > this is quite common for people with corporate SIM cards. > > If you need any further clarification, please, let us know. I think I'll take you up on that offer :) What are "OBs"? Sorry for not knowing what that means. Thanks!

basecamp- based on the following input from clee "This isn't a blocker for v1.0, but one of the highest priorities for our next release."

(In reply to Andrew Overholt [:overholt] from comment #8) > (In reply to David Palomino [:dpv] from comment #2) > > Yes, this is really required by OBs. Please, take into account that > > customers that have Dual SIM cards are generally the most valuable ones, and > > this is quite common for people with corporate SIM cards. > > > > If you need any further clarification, please, let us know. > > I think I'll take you up on that offer :) What are "OBs"? Sorry for not > knowing what that means. Thanks! OB stands for Operator Businesses

(In reply to Daniel Coloma:dcoloma from comment #10) > (In reply to Andrew Overholt [:overholt] from comment #8) > > (In reply to David Palomino [:dpv] from comment #2) > > > Yes, this is really required by OBs. Please, take into account that > > > customers that have Dual SIM cards are generally the most valuable ones, and > > > this is quite common for people with corporate SIM cards. > > > > > > If you need any further clarification, please, let us know. > > > > I think I'll take you up on that offer :) What are "OBs"? Sorry for not > > knowing what that means. Thanks! > > OB stands for Operator Businesses Thanks, Daniel.

(In reply to Lawrence Mandel [:lmandel] from comment #9) > basecamp- based on the following input from clee > "This isn't a blocker for v1.0, but one of the highest priorities for our > next release." This is not what is planned for +V1.0. For further release it is planned to support wwo multiple SIM Cards inserted in the device. What this bug is about is only one SIM Card inserted in the device with two subscriptions. As far as I know it does not require any extra support in the device but the standard STK Commands already in V1 scope. Chris, can we go back to the original situation (bb+) as this is blocking bug 819528?

Daniel, if this is truly blocking certification than it does sound like it is bb+. Chris - Can you clarify the certification requirement?

(In reply to Lawrence Mandel [:lmandel] from comment #13) > Daniel, if this is truly blocking certification than it does sound like it > is bb+. > From the top of the my head: yes, but double checking with dpv.

(In reply to Andrew Overholt [:overholt] from comment #8) > I think I'll take you up on that offer :) What are "OBs"? Sorry for not > knowing what that means. Thanks! Hi, Sorry, OBs = carriers.

Hi, Yes, this is a truly certification blocker, as there are a lot of customers with Dual SIM card (one single physical SIM card / two subscriptions). Besides, these customers used to have corporate subscriptions. Dual STK app is based, as Daniel has said, on commercial SIM Toolkit commands. Basically is based on 'Refresh - SIM Reset' proactive command. Cheers, David

Set up as bb candidate.

I didn't see any REFRESH proactive command from log, CC to M. Vines Hi Michael, does the modem will handle the REFRESH proactive command by itself? If so, when should we do in gecko layer? Thanks

We received a Dual SIM card just today and will be taking a look at it soon. We'll be sure to let you know if we discover any requirements on Gecko as we analyse the issue.

(In reply to Michael Vines [:m1] from comment #19) > We received a Dual SIM card just today and will be taking a look at it soon. > We'll be sure to let you know if we discover any requirements on Gecko as we > analyse the issue. Let's await info from Michael et al before re-triaging this.

(cutting out the middleman)

Andrew, most of the team is on vacation till 2nd Jan. Carol or I will get back to you by 1st week of Jan.

David, in the meanwhile could you please let us know what is the expected behavior. The original description only talks about the observed behavior.

I'm 99% sure this needs to be a blocker since it's not the dual SIM slot that people thought when we minused. To avoid it sitting in the nom queue and to stop bothering our friends with needinfo requests, I'm going to + it.

How are things going here, Yoshi?

(In reply to Andrew Overholt [:overholt] from comment #25) > How are things going here, Yoshi? See comment 22. QC team will check this.

(In reply to Andrew Overholt [:overholt] from comment #25) > How are things going here, Yoshi? This works with commercial ril. After the "Cambio" the SIM requires that we enter the PIN associated with the new card application. However there is no listener for this event so I manually enter the lock screen to get to PIN-entry. Because PIN-entry can technically happen at any time, the most complete solution would be to notify PIN-entry using a system message; otherwise we can just enable this one case and add a listener in the STK code.

(In reply to Michael Schwartz from comment #27) > (In reply to Andrew Overholt [:overholt] from comment #25) > > How are things going here, Yoshi? > > This works with commercial ril. After the "Cambio" the SIM requires that we > enter the PIN associated with the new card application. However there is no > listener for this event so I manually enter the lock screen to get to > PIN-entry. Because PIN-entry can technically happen at any time, the most > complete solution would be to notify PIN-entry using a system message; > otherwise we can just enable this one case and add a listener in the STK > code. Thanks for the info. This sounds to me like the complete solution can be done post-v1 (ie. mark this as a "soft blocker"). Does anyone disagree?

(In reply to Andrew Overholt [:overholt] from comment #28) > (In reply to Michael Schwartz from comment #27) > > (In reply to Andrew Overholt [:overholt] from comment #25) > > > How are things going here, Yoshi? > > > > This works with commercial ril. After the "Cambio" the SIM requires that we > > enter the PIN associated with the new card application. However there is no > > listener for this event so I manually enter the lock screen to get to > > PIN-entry. Because PIN-entry can technically happen at any time, the most > > complete solution would be to notify PIN-entry using a system message; > > otherwise we can just enable this one case and add a listener in the STK > > code. > > Thanks for the info. This sounds to me like the complete solution can be > done post-v1 (ie. mark this as a "soft blocker"). Does anyone disagree? I would like to see it on a device to check how bad the UX is before agreeing with this approach.

IMO, we should block on the Gaia change to fix the PIN entry notification. This would be pretty bad UX IMO. We don't need to block on any changes needed to the reference ril

(In reply to Michael Vines [:m1] from comment #30) > IMO, we should block on the Gaia change to fix the PIN entry notification. > This would be pretty bad UX IMO. We don't need to block on any changes > needed to the reference ril Yeah, I tend to agree, but I would like to see it first. I think the option to enable this one case by adding a listener in the STK code (as suggested by mschwarz) might be feasible. frsela, can you have a look?

Seems it is not related to mozilla RIL, assign owner to default. If you think it's still related to mozilla RIL, please reassign to me.

(In reply to Michael Vines [:m1] from comment #30) > IMO, we should block on the Gaia change to fix the PIN entry notification. > This would be pretty bad UX IMO. We don't need to block on any changes > needed to the reference ril I filed bug 827246 on adding this UI, and bug 827244 to add the platform support needed to enable that.

Given that v1 won't ship with this code we won't block on this.

(In reply to Johnny Stenback (:jst, jst@mozilla.com) from comment #34) > Given that v1 won't ship with this code we won't block on this. Why it has been un-nominated? Certification requires this to be supported regardless it is done through Commercial or Reference RIL.

I've nominated the related gecko and gaia bugs: bug 827244 and bug 827246

The patch https://github.com/mozilla-b2g/gaia/pull/7434 fixes the Dual SIM change and its merged (https://github.com/mozilla-b2g/gaia/commit/53fc61e659fb900295ef1e292bda1267041d43b1)

Video showing the change: http://www.youtube.com/watch?v=qk7E6Jxx72Y&feature=youtu.be

Fernando, test in this video is done with commercial RIL or Moz RIL?

(In reply to Anshul from comment #39) > Fernando, test in this video is done with commercial RIL or Moz RIL? Moz RIL. I'd some issues installing commercial one, I want to test with the commercial RIL ASAP. I'll keep you informed.

(In reply to Anshul from comment #39) > Fernando, test in this video is done with commercial RIL or Moz RIL? Just verified with Commercial RIL with success. In some minutes I'll share the video.

Using the commercial RIL Video: http://www.youtube.com/watch?v=db0_LpGMTCs&feature=youtu.be