You just have to configure all your Memcache severs (IP list) with every Memcache client, and it's done.
That being said, I would suggest that you take the first direction (I.e. deploying Memcache on a dedicated node), this way you decouple your Memcache scaling from Web scaling. Also u should take into account that in many cases Memcache generates a lot of network I/O which can slow down your web/app performance if deployed on the same node. Sent from my iPhone On 2 בפבר 2012, at 11:16, Robert Hanson <[email protected]> wrote: > Dear all, > > I have a plan to start a new project with this scenario : > > 1 server for database > 1 server for web > 1 server for memcache > > In the future, if the traffic increase, I will add more servers as web > server and also memcache. But many developers suggest me to use this > scenario : > > 1 server for database > 1 server for web + memcache > 1 server for web + memcache > > Of course with this second scenario, everytime I buy a new server will > be configured as web + memcache server. > > I don't mind at all, but the question is... > > If i'm using DNS round robin as load balancing method, then user A is > handled by server 1, his PHP session also stored on that server 1, but > on the second request he is handled by server 2, he won't get his > session back, right? > > How to make this memcache server can 'comunicate' each other? So that > user's session won't be broken. > > Thanks.
