I have a tibrv app that I'm potentially looking to replace with zeromq. Could you guys help me figure out how I could accomplish this use case?
Zeromq will be used to route logging information to a central repository. Here's how I use tibrv: - a rendezvous daemon runs on each application server - applications on the server publish to the local rendezvous daemon - the local daemon then pushes those messages to a set of fault tolerant subscribers on machines dedicated to persisting the data to a database (each message is received only once) Here's what I'm looking for: - each application on an app server that publishes message needs to be unaffected by network conditions (zmq_queue on each server?) - the subscriber application needs to be spread redundantly across multiple machines -- messages to be received once in a load balanced fashion across subscribers I'd like some specifics on how I would actually do this (xreq, noblock, zmq_queue, etc?). Once I can get up to speed I can hopefully contribute to the project in some way. Thanks in advanced for your help... Jarred
_______________________________________________ zeromq-dev mailing list [email protected] http://lists.zeromq.org/mailman/listinfo/zeromq-dev
