|
On Wed, 1 Nov 2000, Trevor Marshall wrote:
> Thanks,
> I had already read Ted's mail but it didn't mention how the source port
> comes into play.
> Am I to assume then that even if the client's source port changes then
> the cookie will ensure that the director forwards to the same RIP ?
> regards,
> -Trevor
you've just exhausted my knowlege on the matter. There are people here who
know lots about cookies.
Here's my understanding: if you use LVS persistance (which is really port
affinity) or its equivalent by using fwmark, then you can make a second
connection request from the same clientIP (or client network, if you set
the netmask to say /24 for the persistance) to a different port on the
VIP, go to one real-server. Are you talking about the client making a new
connect request (from a different port) to the same port on the VIP?
The director doesn't look inside packets, so it doesn't look inside the
cookie for content and doesn't know whether a packet contains a cookie.
All cookie handling is done between the client and the real-server that
gets the forwarded connection.
Joe
--
Joseph Mack mack@xxxxxxxxxxx
|
