OS version --------------------------------------------- FireFoxOS v1.4 Reproduce steps: --------------------------------------------- 1. set message outgoing to always ask 2. send a message with sim2 in no signal envirenment, then it will be sent failed. 3. click this message, it will be resent with sim1. Expected result: --------------------------------------------- It should be resent with sim2 which is last used sim. Probability: --------------------------------------------- Always Recurrence

Dear Steve I think we should preserve the sim card index last used in MozSmsMessage object. Could you help to have a look of it? Thanks a great.

Please see bug 983315 and https://github.com/mozilla-b2g/gaia/commit/760de18383f946a6cd65cb19429bf3a3b1fd636d You can fetch message iccid to know which sim slot to send the message and use it as the parameter to send the message again.

Dear Steve Thanks for your advice. That's perfectly right. So could the patch in bug 983315 be landed on v1.4? I think v1.4 needs it. Thanks.

I answered in bug 983315. If all you want is the resend behavior, it should be even safer to do a smaller patch from the patch in bug 983315.

Dear Julien Can this small patch be granted to landed? I really hope it can be landed on 1.4 thread. Thanks a great.

From what I see it should work, but I really don't have the time to test this currently. If you test if from your side (both in Dual and Single SIM conditions) then I think it's good. Please also try to keep passing unit tests.

(In reply to Julien Wajsberg [:julienw] (PTO 08/20 -> 09/15; contact schung instead) from comment #6) > From what I see it should work, but I really don't have the time to test > this currently. > > If you test if from your side (both in Dual and Single SIM conditions) then > I think it's good. > Please also try to keep passing unit tests. Yes, I have tested it locally(both in Dual and Single SIM conditions) and it can work normal. Thanks.

[Blocking Requested - why for this release]:

It would need a real review if it needs to land on our branch. In that case we'd need to update the unit tests and a review from schung@ (because I'm leaving for holidays soon).

Comment on attachment 8474935 [details] [diff] [review] use_the_last_used_sim_to_resend_sms.patch Dear Steve Could you help to review this small patch, as I think this issue should be resolved on 1.4. Thanks a great.

Comment on attachment 8474935 [details] [diff] [review] use_the_last_used_sim_to_resend_sms.patch Hi Wei, the solution is right but we will need unit test if you want to merge in 1.4 branch. Please ref unit tests in master[1] and ask for review again, thanks! [1] https://github.com/mozilla-b2g/gaia/blob/master/apps/sms/test/unit/message_manager_test.js#L493

Reopening this since it provides a different fix to bug 983315. Julien, Steve, If I understand this correctly, this bug doesn't cover bug 983315 entirely but only part of it, right?

(In reply to Wayne Chang [:wchang] from comment #12) > Reopening this since it provides a different fix to bug 983315. > > Julien, Steve, > > If I understand this correctly, this bug doesn't cover bug 983315 entirely > but only part of it, right? ya, this patch only focus on resend by correct sim

Setting this to 1.4+ since the described scenario in comment 0 may introduce bill shock or undesired user experiences. Wei, Please make sure the patch follows through the review and landing criteria before it gets landed. Thanks

Dear Steve This is a PR for 1.4, could you help to review it? Thanks a great.

Comment on attachment 8477250 [details] [v1.4] pull request r=me, thanks for the fixing on v1.4 branch

in v1.4: 1b46b46159604c8833f8fea173b87eaf350253f9

Are v2.0/master unaffected?

(In reply to Ryan VanderMeulen [:RyanVM UTC-4] from comment #18) > Are v2.0/master unaffected? No, it is a problem for 1.4 only. v2.0/master had been resolved in bug 983315 earlier. Thanks.