Zhihong,
Our consumers are our applications (binaries) running on mobile
devices. HTTP header suggestion seems to be a good idea.

Manish,
True, your concern gave me an idea to enforce TS+Nonce checking for
CRUD operation calls only and not for simple fetch calls.

Thanks folks,
Monis

On Jul 6, 10:55 am, Manish Pandit <[email protected]> wrote:
> On Jul 4, 3:10 pm, Monis <[email protected]> wrote:
>
> > Any suggestions?
>
> As a provider implementor, I found the timestamp-nonce verification a
> little heavy on the database side (using multiple VMs with no sticky
> sessions, so in-memory caching is not an option). From a database
> standpoint, it would mean storing the TS+nonce for the +/- time window
> in the db, regular cleanup which gets more and more high-maintenance
> with heavy transactional volumes. I'd think setting up the window to
> +/- 3 minutes should help mitigate reply attacks, however this will
> most certainly guarantee protection. The only thing that can protect
> is persisting the ts-nonce tuple either in-memory on in-database.
> You'd have to consider if adding this extra baggage is worth it based
> on the application and its audience IMO.
>
> -cheers,
> Manish
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"OAuth" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [email protected]
For more options, visit this group at http://groups.google.com/group/oauth?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to