This content has been marked as final. Show 2 replies
797805 wrote:I suspect something is handling on to this .... try something like this to detect what is:
Hi Solaris Experts,
I am wondering whether it is possible to clear a network socket port 17005 left by Apache Tomcat/6.0.20 after having to terminate it forcefully, since it has run away due to remote JDBC resource contention on another server. A normal Tomcat stop / shutdown proves ineffective from then on. This listening port below would hang around for quite some time before disappearing:
This port would prevent Tomcat from starting up successfully again until it is clear, either by waiting for a minimum of 1/2 - 1hr or restarting the server altogether which affect other applications / users on a production server.
localhost.17005 *.* 0 0 49152 0 LISTEN localhost.17005 *.* 0 0 49152 0 TIME_WAIT
I am running jdk1.6.0_11, on SunOS ramble 5.10 Generic_139556-08 i86pc i386 i86pc.
Thanks a lot,
( I also recall a magic 4 min wait number on similar situations, which is now down to 1 min, the timeout wait had something to do with 4 RFC's). But that is a 4 min max; not 1/2 hour.