0

This is going to sound very vague and I'm sure it will be voted down for not giving enough information however I don't really have any to give as you will see.

We have an AWS instance running Amazon Linux, Apache, Tomcat and Railo and from time to time the Tomcat/Railo simply stops responding to requests and there are no errors output in the catalina.out file or any of the other log files in the Tomcat logs directory.

When I issue the command to restart Tomcat/Railo the restart scripts sits there for a while then says that Tomcat has not responded so it has killed it off and then it starts up again and everything is fine until it happens again, anything from a couple of minutes to a couple of days later. I have done my best to check other logs on the server but have found no messages at all to indicate why Tomcat/Railo has given up and stopped responding.

Can anyone suggest any reason why it might be doing this and / or any other log file(s) that we could check to see what is happening.

Thanks.

Andrew.

andrewdixon
  • 228
  • 2
  • 9
  • Don't know anything about Railo, but I'd guess that some resource is running out. All I could suggest is to set up some logging of your own to see if you can start narrowing down what it is trying to do when it stops. See Railo log files http://www.getrailo.org/index.cfm/documentation/configuration/additional-files/ – Jaydee Jun 04 '14 at 14:01

1 Answers1

1

Try these to get additional diagnostics when your application server hangs:

  • Enabling garbage collection logging. Refer to your install of java for flags and directions.
  • Examine threads, typically kill -3 <java process id> is enough, and the data is sent to your application server log file.
  • Heap dumps (refer to your install of java to exact command and flags). Then analyze the heap dump in Memory Analysis Tool (MAT), from Eclipse.
KM.
  • 1,746
  • 2
  • 18
  • 31