Hi all, I was reading this thread (http://community.jboss.org/message/543377#543377) and was wondering whether idle timeout for Hot Rod should by default be -1. As I explained in the past, this parameter is used to handle any misbehaving clients that might be sending incomplete requests. Without an idle timeout, we couldn't catch those rogue clients. However, this shouldn't really be the norm. If there's any suspicion of such things going on, users could restart the server with a positive idle timeout and iron out any issues. Besides, it's pretty clear that the majority of users will be using one of the existing Hot Rod clients and hence the likelihood of seeing such issues is low. Thoughts? -- Galder Zamarreño Sr. Software Engineer Infinispan, JBoss Cache _______________________________________________ infinispan-dev mailing list infinispan-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/infinispan-dev

+1. Isn't 0 (vs -1) going to disable the timeout? I'm also going to add recommendation to client to set the idle connection reaper( timeBetweenEvictionRunsMillis) to a value smaller than server's idle setting. Thanks, Mircea On 26 May 2010, at 10:58, Galder Zamarreno wrote: > Hi all, > > I was reading this thread (http://community.jboss.org/message/543377#543377) and was wondering whether idle timeout for Hot Rod should by default be -1. As I explained in the past, this parameter is used to handle any misbehaving clients that might be sending incomplete requests. Without an idle timeout, we couldn't catch those rogue clients. > > However, this shouldn't really be the norm. If there's any suspicion of such things going on, users could restart the server with a positive idle timeout and iron out any issues. > > Besides, it's pretty clear that the majority of users will be using one of the existing Hot Rod clients and hence the likelihood of seeing such issues is low. > > Thoughts? > -- > Galder Zamarreño > Sr. Software Engineer > Infinispan, JBoss Cache > > _______________________________________________ > infinispan-dev mailing list > infinispan-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/infinispan-dev _______________________________________________ infinispan-dev mailing list infinispan-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/infinispan-dev