Hi, On Fri, Jul 25, 2014 at 8:59 AM, anjan <[email protected]> wrote: > ...Do you see any > issue running multiple Sling instances concurrently in a Server..
Out of the box you'll need to run each Sling instance in its own JVM, you'll need tuning to find out how much RAM to allocate to each JVM. I think in theory you could run multiple instances of the OSGi framework, so multiple instances of Sling, in a single JVM, but I don't think we have the glue code to do that, and that would probably expose some bugs that we don't see when running Sling in its own JVM. You'll also need to configure each Sling instance to use its own port numbers, besides the HTTP port there's at least a control port which will also need to be configured, and Jackrabbit might open RMI ports, I'm not sure if that's the case in our setup. Using LXC containers might help, this sounds like a good use case for Docker. -Bertrand
