[Aerogear-users] Cannot receives pushes via iOS Production variant

Thomas Schweitzer thomers at me.com
Tue Dec 9 09:11:48 EST 2014


Follow up question - could the “INVALID_TOKEN” problem also result from a production token from another app?

The reason I’m asking is that I deleted all the development tokens from the DB, but still couldnt receive a push message on a device.

But I know that I’m sending push registrations to the production variant not just from the “Free” version of my app (using the correct “Free” variant ID and key), but also from the “Paid” version (incorrectly using the “Free" variant ID and key). Since I do not know the tokens sent from the “Paid” version, I cannot disable them. Could this be the reason why I’m not receiving pushes?

Since I’m using the OpenShift cartridge - what’s the best way how to proceed? Can I change the “withNoErrorDetection” of ApnsService somehow?

Thanks,
Thomas




> On 18.11.2014, at 17:04, Matthias Wessendorf <matzew at apache.org> wrote:
> 
>> Hi thomas,
>> 
>> here is an update on the actual issue:
>> https://github.com/notnoop/java-apns/issues/124#issuecomment-63493965
>> 
>> I will file a PR for that
>> 
>> On Tue, Nov 18, 2014 at 5:02 PM, Thomas Schweitzer <thomers at me.com> wrote:
>> Hi Matthias,
>> 
>> java version "1.7.0_71"
>> OpenJDK Runtime Environment (rhel-2.5.3.1.el6-i386 u71-b14)
>> OpenJDK Server VM (build 24.65-b04, mixed mode)
>> 
>> (I assume this is the default version installed by OpenShift, since I didn’t configure anything specific).
>> 
>> Cheers,
>> T
>> 
>> 
>> On 18.11.2014, at 14:57, Matthias Wessendorf <matzew at apache.org> wrote:
>> 
>>> Thomas,
>>> 
>>> I can not reproduce the issue (running Java 1.7.0_65). Corinne is able to reproduce it, but she is currently running on an older Java version (1.7.0_15). She is updating to a newer (the latest) of Java7 and will rerun her tests.
>>> 
>>> In the meantime, what is your version?
>>> 
>>> Greetings,
>>> Matthias
>>> 
>>> On Tue, Nov 18, 2014 at 12:27 PM, Matthias Wessendorf <matzew at apache.org> wrote:
>>> Corinne shared her tokens, and using them, I get the same output:
>>> 
>>> :::::::::::::::::::::::::::::::::::::::::::::::::>>>messageSent
>>> :::::::::::::::::::::::::::::::::::::::::::::::::>>>messageSent
>>> :::::::::::::::::::::::::::::::::::::::::::::::::>>>Closed: INVALID_TOKEN
>>> :::::::::::::::::::::::::::::::::::::::::::::::::>>>messageSendFailed[B at 143f032d
>>> :::::::::::::::::::::::::::::::::::::::::::::::::>>>Retry: 0
>>> 
>>> 
>>> 
>>> On Tue, Nov 18, 2014 at 12:02 PM, Matthias Wessendorf <matzew at apache.org> wrote:
>>> what JDK version do you have ? 
>>> 
>>> I am running on jdk1.7.0_65.jdk
>>> 
>>> I am unable to reproduce this behavior.... 
>>> 
>>> I have a connection to prod-stage from APS:
>>> 
>>> 
>>> 
>>> .withCert("/Users/matzew/Desktop/PushCerts/prod.p12", "XXXX").withProductionDestination()
>>> 
>>> 
>>> and I give it too tokens:
>>> iPodToken (production version of the "net.wessendorf.prod" app-id)
>>> iPhoneToken (development version of the "net.wessendorf.prod" app-id) 
>>> 
>>> Since connecting to the PROD, only the iPod is delivered. Also the ordering of the tokens does not matter .. 
>>> 
>>> 
>>> 
>>> 
>>> 
>>> 
>>> 
>>>  
>>> 
>>> On Mon, Nov 17, 2014 at 3:45 PM, Corinne Krych <corinnekrych at gmail.com> wrote:
>>> Adding a test as you suggested Matt
>>> see: https://github.com/corinnekrych/java-apns/blob/adding.test.with.mixed.variant.similation/src/test/java/com/notnoop/apns/APNSTest.java#L31
>>> Where line 67 content valid prod certificate for simulated “prod mixed variant list” and line 68 content invalid dev certificate.
>>> 
>>> the output is:
>>> :::::::::::::::::::::::::::::::::::::::::::::::::>>>messageSent
>>> :::::::::::::::::::::::::::::::::::::::::::::::::>>>messageSent
>>> :::::::::::::::::::::::::::::::::::::::::::::::::>>>Closed: INVALID_TOKEN
>>> :::::::::::::::::::::::::::::::::::::::::::::::::>>>messageSendFailed[B at 143f032d
>>> :::::::::::::::::::::::::::::::::::::::::::::::::>>>Retry: 0
>>> 
>>> I case of error for the variant line68 we go on all callback: success, failure, retry etc..
>>> 
>>> however when testing it with pan I do received the message for the valid token.
>>> 
>>> @matthias does it help?
>>> 
>>> ++
>>> Corinne
>>> 
>>> 
>>> > On 17 Nov 2014, at 13:27, Matthias Wessendorf <matzew at apache.org> wrote:
>>> >
>>> >
>>> >
>>> > On Mon, Nov 17, 2014 at 12:55 PM, Matthias Wessendorf <matzew at apache.org> wrote:
>>> > Hi Corrine,
>>> >
>>> > thanks for the detailed mail. To me this looks like a bug on java-apns...
>>> > Can you try that, with the standalone code (and your scenarios, like mixing tokens).
>>> >
>>> > here is some code snippets:
>>> > https://github.com/notnoop/java-apns#sample-code
>>> >
>>> >
>>> > perhaps like:
>>> > https://gist.github.com/matzew/35b47c74f8eb6a8353de
>>> >
>>> >
>>> >
>>> > I am currently on a train with a bad connection
>>> >
>>> >
>>> > On Mon, Nov 17, 2014 at 12:45 PM, Corinne Krych <corinnekrych at gmail.com> wrote:
>>> > Hello all
>>> >
>>> > Indeed I can reproduce the problem as described in AGPUSH-1087
>>> >
>>> > I have one variant: PROD with one installation on iPhone5S with Distribution certificate
>>> >
>>> > I have another variant: PRODMIX with 2 installations:
>>> > - on iPad mimi with distribution certificate
>>> > - on iPad2 with dev certificate
>>> >
>>> > When sending a notification to all variants, only the iPhone5S on the PROD variant received the notification.
>>> >
>>> > If I go to admin console and desactivate the iPad2 installation and sent again a Push message to all variants, now iPhone 5S and IPad mini received the notification
>>> > => Therefore, it is indeed when one installtion has non valid certificate thta all installations of this corrupted variants will not received the notification. I also confirm that the work around does work.
>>> >
>>> > Looking on FINE grained logging, I noticed that this succeed callback is called:
>>> > https://github.com/aerogear/aerogear-unifiedpush-server/blob/master/push/src/main/java/org/jboss/aerogear/unifiedpush/message/SenderServiceImpl.java#L140
>>> >
>>> > The problem seem to be in deepsest level of code. I will install a UPS server locally to test further.
>>> >
>>> > ++
>>> > Corinne
>>> >
>>> >
>>> >
>>> > > On 17 Nov 2014, at 11:36, Sebastien Blanc <scm.blanc at gmail.com> wrote:
>>> > >
>>> > > Hi Thomas,
>>> > > You can browse your installations by clicking the blue number of "x installations" :
>>> > >
>>> > > <installation.png>
>>> > > ​
>>> > > From there, you can toggle installations on on/off.
>>> > >
>>> > > Sebi
>>> > >
>>> > >
>>> > > On Mon, Nov 17, 2014 at 10:46 AM, Thomas Schweitzer <thomers at me.com> wrote:
>>> > >
>>> > > > another option is: toggle them as disabled on the UI of the Admin UI ;-)
>>> > >
>>> > > Hmm, where do I see the device tokens and can disable them in the admin UI?
>>> > >
>>> > >
>>> > >
>>> > > _______________________________________________
>>> > > Aerogear-users mailing list
>>> > > Aerogear-users at lists.jboss.org
>>> > > https://lists.jboss.org/mailman/listinfo/aerogear-users
>>> > >
>>> > >
>>> > > _______________________________________________
>>> > > Aerogear-users mailing list
>>> > > Aerogear-users at lists.jboss.org
>>> > > https://lists.jboss.org/mailman/listinfo/aerogear-users
>>> >
>>> >
>>> > _______________________________________________
>>> > Aerogear-users mailing list
>>> > Aerogear-users at lists.jboss.org
>>> > https://lists.jboss.org/mailman/listinfo/aerogear-users
>>> >
>>> >
>>> >
>>> > --
>>> > Matthias Wessendorf
>>> >
>>> > blog: http://matthiaswessendorf.wordpress.com/
>>> > sessions: http://www.slideshare.net/mwessendorf
>>> > twitter: http://twitter.com/mwessendorf
>>> >
>>> >
>>> >
>>> > --
>>> > Matthias Wessendorf
>>> >
>>> > blog: http://matthiaswessendorf.wordpress.com/
>>> > sessions: http://www.slideshare.net/mwessendorf
>>> > twitter: http://twitter.com/mwessendorf
>>> > _______________________________________________
>>> > Aerogear-users mailing list
>>> > Aerogear-users at lists.jboss.org
>>> > https://lists.jboss.org/mailman/listinfo/aerogear-users
>>> 
>>> 
>>> _______________________________________________
>>> Aerogear-users mailing list
>>> Aerogear-users at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/aerogear-users
>>> 
>>> 
>>> 
>>> -- 
>>> Matthias Wessendorf 
>>> 
>>> blog: http://matthiaswessendorf.wordpress.com/
>>> sessions: http://www.slideshare.net/mwessendorf
>>> twitter: http://twitter.com/mwessendorf
>>> 
>>> 
>>> 
>>> -- 
>>> Matthias Wessendorf 
>>> 
>>> blog: http://matthiaswessendorf.wordpress.com/
>>> sessions: http://www.slideshare.net/mwessendorf
>>> twitter: http://twitter.com/mwessendorf
>>> 
>>> 
>>> 
>>> -- 
>>> Matthias Wessendorf 
>>> 
>>> blog: http://matthiaswessendorf.wordpress.com/
>>> sessions: http://www.slideshare.net/mwessendorf
>>> twitter: http://twitter.com/mwessendorf
>>> _______________________________________________
>>> Aerogear-users mailing list
>>> Aerogear-users at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/aerogear-users
>> 
>> 
>> _______________________________________________
>> Aerogear-users mailing list
>> Aerogear-users at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/aerogear-users
>> 
>> 
>> 
>> 
>> -- 
>> Matthias Wessendorf 
>> 
>> blog: http://matthiaswessendorf.wordpress.com/
>> sessions: http://www.slideshare.net/mwessendorf
>> twitter: http://twitter.com/mwessendorf
>> _______________________________________________
>> Aerogear-users mailing list
>> Aerogear-users at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/aerogear-users
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/aerogear-users/attachments/20141209/a0585f31/attachment-0001.html 


More information about the Aerogear-users mailing list