On 2/16/06, Greg Duffy <[EMAIL PROTECTED]> wrote: > > Would it be possible/prudent to allow a minimum thread pool size in > the TheadPoolFilter? After periods of inactivity I notice a small bit > of latency when new threads are being created, and it'd be nice to > have some control over it. My "clients" (actually automated processes) > are impatient and tend to retransmit if I don't send something back > quickly enough (and I need to do it in a separate connection, which > means more latency if I'm using different ThreadPoolFilter(s) for > that). For my purposes, retransmits are not fatal, but it would be > nice to avoid them nonetheless.
You can track this issue from now on here: http://issues.apache.org/jira/browse/DIRMINA-178 Also, I was wondering where the best place is to initialize the > IoFilterChain. Right now I'm doing it on the creation or opening of a > session, but the set of filters I use is static. Is there a better or > more performant place to put that? I've got two instances I'm > wondering about: one using a ServiceRegistry and one initializing a > connector directly (for client communication). Since 0.9.2, the configuration API has been improved. Here's the example: SocketAcceptorConfig cfg = new SocketAcceptorConfig(); cfg.getFilterChain().addLast( "...", ... ); ... cfg.setReuseAddress( true ); acceptor.bind( address, myHandler, cfg ); WDYT? On a side note, I'm just starting to work with MINA, so hello and > thanks for a cool project! Feel free to set me straight if I'm missing > a big concept in anything above. Thank you for your high-quality feedback. The Apache MINA team will always respect your criticism. Please feel free to keep feeding us back. :) 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
