[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kolab-devel
Subject:    Re: [Kolab-devel] Completing ActiveSync integration in Kolab,
From:       Alain Abbas <alain.abbas () libertech ! fr>
Date:       2010-03-22 18:03:22
Message-ID: A9F5B3F4-74D2-4B76-AB13-7A784B30FA5A () libertech ! fr
[Download RAW message or body]





Le 22 mars 2010 Ã  18:21, "Georg C. F. Greve" <greve@kolabsys.com> a  
écrit :

> On Monday 22 March 2010 17:37:42 Bernhard Reiter wrote:
>> Given that serial number just is a unique number, this looks  
>> universal to
>> me.
>
> Ah, but this is an ActiveSync based serial number, from what I  
> understood,
> generated on the client through some weird magic.
>
this is the serial number of the phone ( hardwired) it is inique
> So in the end the number will be ActiveSync specific, and we cannot  
> safely
> exclude namespace conflicts with other synchronisation schemes, e.g.  
> one
> ActiveSync phone having the same serial id as another phone with  
> Cool2012Sync
> while the user would want them to behave separately.
>
i think its not possible it is hard like the mac address
there are another unique id that is generated to the activesync state
but it change at each reset of the synchronisation
> As we are unlikely to end up with more than 10 supported  
> synchronisation
> schemes (which would mean 10 annotations), the overhead seems  
> manageable, and
> keeping it focused seems sensible.
>
>
>> Just out of curiosity: Where is the status of the sync saved in the  
>> z-push
>> backend?
>
> You mean the state & mapping of the objects? In database files in a  
> specific
> directory on the hard disk right now, one file per user, IIRC.
>
>
>> I am asking because I think that should be a full cache, so if thrown
>> completely away, the system should be self healing.
>
> I am not sure what you mean by "system" and what is your scenario  
> for "thrown
> completely away" in the above.
>
> For synchronisation the answers to the above would be largely  
> dependent upon
> the ActiveSync standard and the way it is implemented on the device.
>
> Ultimately my experience is that attempts to self-heal are easy  
> paths for
> eternally replicating information, potentially tearing down your  
> Kontact
> client with thousands and thousands of duplicate entries.
>
> Yes. I had this with the Funmabol connector once. It is not fun.
>
> Sometimes dumber is smarter. ;)
>
> Best regards,
> Georg
>
> -- 
> Georg C. F. Greve
> Chief Executive Officer
>
> Kolab Systems AG
> Zürich, Switzerland
>
> e: greve@kolabsys.com
> t: +41 78 904 43 33
> w: http://kolabsys.com
>
> pgp: 86574ACA Georg C. F. Greve
> _______________________________________________
> Kolab-devel mailing list
> Kolab-devel@kolab.org
> https://kolab.org/mailman/listinfo/kolab-devel

_______________________________________________
Kolab-devel mailing list
Kolab-devel@kolab.org
https://kolab.org/mailman/listinfo/kolab-devel

[prev in list] [next in list] [prev in thread] [next in thread]