Your response is what I was sort of expecting but wasn't sure if there might've been something behind the scenes mina provided that I could use.
Thanks though for the quick response. S.D. --- Trustin Lee <[EMAIL PROTECTED]> wrote: > On 3/4/06, Samuel Doyle <[EMAIL PROTECTED]> wrote: > > > > Nobody has any suggestions on using some form of > mina > > provided approach for this? > > > It takes some time for you to get a response; > because it's an open source > software. It can even take a week. :) > > > Hi, I'm basically attempting to use mina to act as > a > > > proxy server something like discussed in the > > > previous > > > example: > > > > > > > > http://www.nabble.com/-MINA-Using-MINA-to-write-a-proxy-server-like-application-t656623.html > > > However, the case I have is not 1:1 on the > client > > > and > > > server session basis. > > > > > > The situation I have involves a client which > spawns > > > multiple processes, each opening its own > connection > > > to > > > my proposed mina server. The mina server itself > has > > > a > > > persistent fixed amount of connections to the > actual > > > server to distribute the client requests with. > The > > > client connections are short lived are only open > to > > > process a single request while the mina->server > > > should > > > always be open. > > > > > > Client 1<---------| > > > | Persistent Conn 1 > > > Client 2<--------|| |--------------> > > > Mina Server Actual > Server > > > Client 3<--------|| |--------------> > > > | Persistent Conn 2 > > > Client 4<---------| > > > > > > The amount of client connections is variable but > the > > > connections from mina to the actual server is > always > > > fixed. My question is, what is the best way to > demux > > > the responses from the actual server to the > > > appropriate client? One of the fixed > mina->server > > > sessions may have several client->mina sessions > > > attached to it at one time. Is there something > Mina > > > provides to do this or do I need to come up with > > > some > > > home grown based version? > > > There's no special way to implement it. To do so, > there has to be an > identifier for each message so you can determins > which message belongs to > which client. Plus, the proxy should be able to > interpret the message so it > can extract the ID field. That's all. > > HTH, > Trustin > -- > what we call human nature is actually human habit > -- > http://gleamynode.net/ > -- > PGP key fingerprints: > * E167 E6AF E73A CBCE EE41 4A29 544D DE48 FE95 4E7E > * B693 628E 6047 4F8F CFA4 455E 1C62 A7DC 0255 ECA6 >
