Sounds to me like a permissions/environment problem. For example, starting Tomcat as root manually, but then the script tries to start tomcat as some non-root user. If Tomcat starts as root, log files are owned by root and non-root users cannot write to them. It should throw an error somewhere, though maybe its getting swallowed by a redirect to /dev/null or something.
I would verify the script and doublecheck that the script environment (variables, user account, etc) matches exactly with the way you start Tomcat manually.
John
Noel Rappin wrote:
I have a set up where tomcat is supposed to start on boot via an rc shell script on a Linux system (tomcat 4.1.18). I have this intermittent failure mode where tomcat fails to start up on reboot.
I can _always_ trigger this failure by deleting all tomcat log files, and then rebooting. After the reboot, the only thing in the log directory is a zero-length catalina.out file, and the java process is not running. Invoking the tomcat startup again at this point causes it to work normally.
I can find no evidence that the tomcat java process exits, and no matter how much I turn up debugging, I get no logged output. Is there any possible mechanism that could be causing this failure, or at least some place or some way I could get better debugging information?
Noel
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
