852038 - [Buri][PLMN] 'automatic selection' mode is changed after reboot MS.

Status: VERIFIED FIXED

(Firefox OS Graveyard :: Gaia::Settings, defect, P2)

Description

[sync-1]

+++ This bug was initially created as a clone of Bug #413800 +++
 
 DEFECT DESCRIPTION:
 'automatic selection' mode is changed after reboot MS.
 
 REPRODUCING PROCEDURES:
 1.Select 'Setting'-->Celluar&Data-->Network operator;
 2.Switch off 'Automatic selection';
 3.Reboot MS, 'Automatic selection' is set as on;-->KO
 4.MS registers on VPLMN automatically;-->KO
 
 EXPECTED BEHAVIOUR:
 3.Select mode should not change after reboot;
 4.MS should not register on VPLMN except RPLMN;
 
 ASSOCIATE SPECIFICATION:
 
 TEST PLAN REFERENCE:
 
 TOOLS AND PLATFORMS USED:
 
 USER IMPACT:
 
 REPRODUCING RATE:
 5/5
 
 For FT PR, Please list reference mobile's behavior:
 
 ++++++++++ end of initial bug #413800 description ++++++++++
 
 
 
 CONTACT INFO (Name,Phone number):
 
 DEFECT DESCRIPTION:
 
 REPRODUCING PROCEDURES:
 
 EXPECTED BEHAVIOUR:
 
 ASSOCIATE SPECIFICATION:
 
 TEST PLAN REFERENCE:
 
 TOOLS AND PLATFORMS USED:
 
 USER IMPACT:
 
 REPRODUCING RATE:
 
 For FT PR, Please list reference mobile's behavior:
 Firefox os  v1.0.1
 Mozilla build ID: 20130310070203.

Comment 1

[Daniel Coloma:dcoloma]

I believe this would be fixed as soon as bug 834973

Comment 2

[Daniel Coloma:dcoloma]

We are not 100% sure if it is a duplicate of bug 834973 marking as tef+ to be on the safe side as it is a critical feature.

Comment 3

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

evelyn, can you take this and take a first look? Thanks

Comment 4

[Andrew Overholt [:overholt]]

(In reply to Daniel Coloma:dcoloma from comment #2)
> We are not 100% sure if it is a duplicate of bug 834973 marking as tef+ to
> be on the safe side as it is a critical feature.

Okay, let's make this depend on bug 834973 so we've got that linkage.

Comment 5

[Evelyn Hung]

After discussing with Hsinyi, I believe the issue may caused by bug 834973. We didn't really switch to the manually selected network, so the `networkSelectMode` in Gecko(modem) didn't change its state from 'automatic' to 'manual'. So the setting shows 'automatic' when the app is launched and check `networkSelectMode`.
I will test the patch of bug 834973 and see if the issue fixed.

Comment 6

[Daniel Coloma:dcoloma]

(In reply to Evelyn Hung [:evelyn] from comment #5)
> After discussing with Hsinyi, I believe the issue may caused by bug 834973.
> We didn't really switch to the manually selected network, so the
> `networkSelectMode` in Gecko(modem) didn't change its state from 'automatic'
> to 'manual'. So the setting shows 'automatic' when the app is launched and
> check `networkSelectMode`.
> I will test the patch of bug 834973 and see if the issue fixed.

If you manage to confirm that please mark this as duplicate

Comment 7

[Evelyn Hung]

(In reply to Daniel Coloma:dcoloma from comment #6)
> If you manage to confirm that please mark this as duplicate
After apply the patch, the problem is there. I found that when Settings app launched, the first query `mobileConnection.networkSelectionMode` always returns 'null', but it did changed to 'manual' if we successfully switch to another operator network.
I've loop Hsinyi to looking the problem in Gecko. It might be a Gecko issue.

Comment 8

[Evelyn Hung]

Okay, so I file bug 855643 for a Gecko patch, and I am fixing some UI display problem here.

Comment 9

[Evelyn Hung]

Attachment: point to https://github.com/mozilla-b2g/gaia/pull/8882

Comment 10

[Fabien Cazenave [:kaze]]

Comment on attachment 730986 [details]
point to https://github.com/mozilla-b2g/gaia/pull/8882

Looks good to me, merged on master:
https://github.com/mozilla-b2g/gaia/commit/b04dc95dc11a4c375875e56639f16c0be75bafed

Comment 11

[sarsenyev]

Environmental  Variables:
Unagi Build ID: 20130401070203
Kernel Date: Dec 5
Gecko: http://hg.mozilla.org/releases/mozilla-b2g18_v1_0_1/rev/b28463f2e718
Gaia: ddb38ac8a34f9e30e09d0ff3b5c1bfb9b664b7c3

The same issue reproduces with "Unagi" device

Comment 12

[Fabien Cazenave [:kaze]]

sarsenyev, your Gaia hash indicates a v1.0.1 branch, whereas I’ve merged this patch on master.

If you’re sure that this patch has landed on 1.0.1, please reopen this bug.

Comment 13

[John Ford [:jhford] CET/CEST Berlin Time]

v1-train: 57a1a64e1af4a5667db2e84fdd0942675034b5e6

There is a small merge conflict on v1.0.1.  You might be able to resolve it by doing:

  cd gaia
  git checkout v1.0.1
  git cherry-pick 57a1a64e1af4a5667db2e84fdd0942675034b5e6
  <resolve merge conflict>
  git add apps/settings/js/carrier.js
  git commit

Comment 14

[Daniel Coloma:dcoloma]

Evelyn, can you please help with the uplift here? This is important for partners and certification.

Comment 15

[Evelyn Hung]

(In reply to Daniel Coloma:dcoloma from comment #14)
> Evelyn, can you please help with the uplift here? This is important for
> partners and certification.
Sorry I wasn't noticed the request here. Conflict resolved branch:
https://github.com/evelynhung/gaia/tree/uplift-852038-v1.0.1

Comment 16

[Daniel Coloma:dcoloma]

John, can you try the uplift now?

Comment 17

[John Ford [:jhford] CET/CEST Berlin Time]

Evelyn, I'm seeing a small conflict still:

++<<<<<<< HEAD
 +        messageElement.dataset.l10nId = 'operator-status-connected';
++=======
+         updateSelectionMode(false);
++>>>>>>> 4facf15... Merge pull request #8882 from evelynhung/issue-852038

Could you update your conflict resolution?

Comment 18

[Evelyn Hung]

Not sure what happened but I did the `git cherry-pick 57a1a64e1af4a5667db2e84fdd0942675034b5e6` again based on the latest v1.0.1 branch, no conflict at all. I also update my branch in comment 15.
Thank you so much.

Comment 19

[John Ford [:jhford] CET/CEST Berlin Time]

(In reply to Evelyn Hung [:evelyn] from comment #18)
> Not sure what happened but I did the `git cherry-pick
> 57a1a64e1af4a5667db2e84fdd0942675034b5e6` again based on the latest v1.0.1
> branch, no conflict at all. I also update my branch in comment 15.
> Thank you so much.

Interesting.  It is working now, I wonder what was going wrong.

v1.0.1: e16ebac1e3f02ce98dd0e4138188fc9ab5f14d39

Comment 21

[Amelie Kong]

According to our verification result, based on version:
AU_LINUX_GECKO_ICS_STRAWBERRY_V1.01.00.01.019.080
Firefox os  v1.0.1
Mozilla build ID:20130418070206

The problem is still there.

So I reopen it.

Comment 22

[Evelyn Hung]

(In reply to Amelie Kong from comment #21)
> According to our verification result, based on version:
> AU_LINUX_GECKO_ICS_STRAWBERRY_V1.01.00.01.019.080
> Firefox os  v1.0.1
> Mozilla build ID:20130418070206
> 
> The problem is still there.
> 
> So I reopen it.
Could you describe your reproduce steps?
I think the bug is fixed but we have a confusing UX.

If you switch the toggle off but didn't manually select an operator,
it's still in automatic mode, the mode will be set by RIL interface (not a setting value) only when we successfully switch to another operator network, and RIL API `mobileConnection.networkSelectionMode` returns 'manual' at that time, and then UI updates.

Try this STR to verify the bug fixed:
1. Select 'Setting'-->Celluar&Data-->Network operator;
2. Switch off 'Automatic selection';
3. wait network operator list shows and select a network to connect
4. see the network is connected and the subtitle under 'Automatic selection' is changed to 'manual'.

Expected:
5. Reboot phone, 'Automatic selection' is set as 'manual', and the network is connected to the manually selected one.

@Josh, Larissa: could you give some feedback on the UX?

Comment 23

[Evelyn Hung]

ni Casey because Josh is PTO.

Comment 24

[Beatriz Rodríguez [:brg]]

I have tested it(buri with AU 19.088)  and I do not see any wrong behaviour. Maybe UI is a bit confusing but I think behaviour is right.

Comment 25

[David Palomino [:dpalomino]]

Hi, 

I think this would be the expected behaviour: 

- If the user manually select a network, and it's registered, after reboot the device, manual network selection is maintained

- If the user enables manual network selection, but it's not finally registered, then after reboot the device, automatic network selection is enabled (if not doing so, device would be not registered)

We think both behaviour are correct. 

Thanks!
David

Comment 26

[Casey Yee]

deferring to Francis on this one.

Comment 27

[Evelyn Hung]

(In reply to Beatriz Rodríguez [:brg] PTO till 6thMay from comment #24)
> I have tested it(buri with AU 19.088)  and I do not see any wrong behaviour.
> Maybe UI is a bit confusing but I think behaviour is right.

(In reply to David Palomino [:dpv] from comment #25)
> Hi, 
> 
> I think this would be the expected behaviour: 
> 
> - If the user manually select a network, and it's registered, after reboot
> the device, manual network selection is maintained
> 
> - If the user enables manual network selection, but it's not finally
> registered, then after reboot the device, automatic network selection is
> enabled (if not doing so, device would be not registered)
> 
> We think both behaviour are correct. 
> 
> Thanks!
> David
That's what it behaves right now.
I would like to close this issue first and wait for someone's verifying.

Comment 28

[Jeni]

Verified fixed on 
Inari Build ID: 20130429070204
Kernel Date: Feb 21
Gecko: http://hg.mozilla.org/releases/mozilla-b2g18_v1_0_1/rev/45aa5ba0ed53
Gaia: cf2d4136f0ebc66039637fdbeb72ed184dfbc0f2

and on 

Unagi Build ID: 20130429070204
Kernel Date: Dec 5
Gecko: http://hg.mozilla.org/releases/mozilla-b2g18/rev/49a908f052ec
Gaia: e26eadbc598bd09c80192016c1024a4a999a5361

User is now able to change status to Manual under Network Operator and have device stay manual after power cycling.

Comment 29

[Al Tsai [:atsai]]

Verified w/ Unagi
on v1-train
Gaia:  0ddb515f15cbc6b74fc2742b7599d6ae74c6413f
Gecko: 5d2cf5184ac6f99758668da9e6a5b1a34f9fefc8