-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 10/4/09 3:03 PM, Winfried Tilanus wrote:
> On 09/30/2009 08:39 PM, Peter Saint-Andre wrote:
> 
> Hi,
> 
> I needed to think about this one for a while.
> 
>> Because of how HTTP is deployed, there might be intermediaries between a
>> BOSH client and a BOSH connection manager (CM). Such intermediaries
>> might include proxies, caching servers, and load balancers. Some of
>> these entities, especially load balancers, might redirect an HTTP
>> request from the client to a different CM.
> 
> First of all the *only* entity I can think of that redirects requests to
> different CM's is (by definition) a load-balancer.
> 
> (One other scenario I can think of that would result in the sesion of
> one client getting connected to different CM's, is the use of a
> round-robin DNS. That would bring up a whole other range of problems,
> just like automatic fail over of CM's would do.)
> 
> So apparently we are speaking only about load balancers here. AFAIK are
> load balancers always deployed by the owner of the server park. So both
> the CM's and the load balancers are under the same control.

Not always. It depends on the deployment.

>> Discussions with implementers have led me to conclude that the best way
>> to do this is for the CM to set a temporary cookie on the client
>> (containing a session ID) in its response to the first BOSH request, and
>> for the client to return that cookie value to the CM when it sends
>> future requests during the life of the BOSH session.
> 
> Sending cookies is fine with me, as long as sending them is optional. If
> the operator of a site thinks it is the best for their load balancer to
> use the cookies, they can configure their CM's to do so. For our project
> I would prefer to have the option *not* to store cookies. In some cases
> the idea that a cookie is stored (how unintrusive it might be in
> reality) is just too much. In some other cases even a cookie that
> expires at the end of the session really might be too much.

Agreed. Even if the client supports cookies, it needs to provide a way
for the user to disable that support.

Peter

- --
Peter Saint-Andre
https://stpeter.im/


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkrLfYAACgkQNL8k5A2w/vz0uACfcihj2ojpJslokBVpd1PK4a5d
qIwAn3ZGiCj9wQV1u5yE0AT/rjHEuQYl
=En71
-----END PGP SIGNATURE-----

Reply via email to