tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From André Warnier>
Subject Re: Tomcat dying on its own
Date Wed, 08 Dec 2010 23:23:57 GMT
Mark Thomas wrote:
> On 08/12/2010 22:38, Christopher Schultz wrote:
>> Mark,
>> On 12/6/2010 3:49 PM, Mark Thomas wrote:
>>> On 06/12/2010 20:44, Guillaume Carbonneau wrote:
>>>> It happens every day or so.
>>>> Nothing happens before in the log... I'm logging the requests but
>>>> tomcat dies when there is no traffic it seems.
>>>> I even tried to bench it with more than 10000 requests the other day
>>>> and it handled the load pretty fine and no "crash".
>>>> I control all the code in my app, there is no System.exit() call
>>> There have been cases of libraries calling System.exit() on some error
>>> conditions.
>> Isn't there some indication of a System.exit() in catalina.out?
>> If there isn't, we could write an optional Listener that could install a
>> shutdown hook that will complain to stdout if the JVM goes down without
>> Tomcat's "permission".
> Been there, tried that. There is no easy way (I could find) to
> distinguish between a legitimate and illegitimate shut-down.
Naive suggestion : how about some global flag which the legitimate shutdown paths set (and

the others don't know about), and which the "hook" in question checks ?

Other naive suggestion/question : something somewhere already intercepts a System.exit(),

if i judge by the shutdown messages of the Connectors.  How about making the flag a code 
or string, which would indicate which legitimate agent triggered it, and which the things

which print the Connector shutdown messages obtain and duly add to their shutdown message

?  The absence of ditto (or the initialisation content of it) would be a telltale sign.

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message