[Wrapper-user] Odd intermittent problem on Solaris

[Robert D. <rdi...@tr...>]

Let me start off by stating that the Wrapper is really great.
Unfortunately, we have an intermittent problem under Solaris. Sometimes,
a few days or weeks after continuous running, the JVM seems to just exit
and the wrapper does not restart the JVM. In fact, the wrapper thinks it
is not getting any CPU (this is even after the JVM exists).

So, there are two odd things here: (a) the JVM mysteriously exits and
(b) the wrapper does not restart it. We have not seen this under
Windows. We are doing a DEBUG run to see if we can get more information,
but just thought I'd send this out to see if anyone had any initial
thoughts.

Here's an example of the log:

INFO   | jvm 1    | 2003/12/10 14:08:22 | [connection
Thread-2,id=3D5,uses=3D232]: Connection has exceeded idle lease!=20

The above is a normal log message from our system. Then it just silently
stopped running. The wrapper was still running and we heard nothing
until we shut it down to restart here:

STATUS | wrapper  | 2003/12/11 13:59:11 | Shutting down.
INFO   | jvm 1    | 2003/12/11 13:59:11 | Wrapper Manager: The Wrapper
code did not ping the JVM for 30 seconds.  Quit and let the Wrapper
resynch.
INFO   | jvm 1    | 2003/12/11 13:59:11 | Stopping service
Tomcat-Standalone
INFO   | wrapper  | 2003/12/11 13:59:11 | Wrapper Process has not
received any CPU time for 85640 seconds.  Extending timeouts.
STATUS | wrapper  | 2003/12/11 13:59:11 | <-- Wrapper Stopped DEBUG=20

Odd that the wrapper said it didn't get any CPU time for 85640 second
(the time that the system was down) since the JVM process (nor anything
else) was running.

Thoughts?

R.