On Wed, 2012-06-20 at 19:19 -0400, David Kerber wrote: > On 6/19/2012 8:07 PM, André Warnier wrote: > > James Lampert wrote: > >> . . . and when I looked back at the box I was testing, Tomcat *had* > >> finally shut down. And when I ran both the start and stop scripts > >> this time, the stop script worked perfectly (and promptly). > >> > >> Weird. Why would the shutdown take so long as to give the impression > >> it had failed entirely, then eventually work, then later work promptly? > >> > > > > 1) Gremlins. You need to exorcise your datacenter. > > or > > 2) because the first time, Tomcat had been running for a while, so it > > had a lot of things to shut down nicely and cleanup; while the second > > time, it had only be running for a much shorter time, and had less to > > clean up ? > > It's not gremlins, it's magic. I see this all the time on windows: it > shuts down in the time I'm willing to wait for it ~50% of the time. The > other 50% I end up killing the task. > Any time I've seen this it's because some app or library has carelessly started non-daemon threads that are still spinning. Everything Tomcat can shutdown has shutdown. You could always try 'catalina.sh stop -force'.
> > --------------------------------------------------------------------- > To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org > For additional commands, e-mail: users-h...@tomcat.apache.org >
signature.asc
Description: This is a digitally signed message part
