Re: [5.0] Servlet logging and error messages

2003-07-21 Thread Remy Maucherat
Costin Manolache wrote: I don't agree with the last point, but I right now I don't have the time to disagree :-) I'm +1 to have the logs related to a context go to the context log file - after the request is mapped to a context. Everything could go in the context's log and be controlled by a per

RE: [5.0] Servlet logging and error messages

2003-07-21 Thread Shapira, Yoav
Howdy, >The full trace of the root cause is now logged (instead of the full >trace of the exception, which was not very useful IMO). >The new display seems to me significantly easier on users (esp when >using the request dispatcher a lot, which could have created huge stack >traces in the past).

Re: [5.0] Servlet logging and error messages

2003-07-21 Thread Remy Maucherat
Shapira, Yoav wrote: Howdy, - Stripping out all the Catalina related part of the stack trace in the HTML generated by the ErrorReportValve; that eases debugging. Hmm... As long as the full stack trace, including Catalina, is still in the log file, that's OK. I know the Catalina parts of the st

RE: [5.0] Servlet logging and error messages

2003-07-21 Thread sathya narayanan
hi , IS there is any way to Connect Database using DBCP.jar without giving username and password in the server.xml file. Doc tells that we can pass the uname and password in the getConnection() method of DataSource . But it is give me an Exception . Plese help me. sathya

RE: [5.0] Servlet logging and error messages

2003-07-21 Thread Shapira, Yoav
Howdy, >- Stripping out all the Catalina related part of the stack trace in the >HTML generated by the ErrorReportValve; that eases debugging. Hmm... As long as the full stack trace, including Catalina, is still in the log file, that's OK. I know the Catalina parts of the stack trace frequently

Re: [5.0] Servlet logging and error messages

2003-07-20 Thread Costin Manolache
Remy Maucherat wrote: > I'm working on the following changes: > > - Stripping out all the Catalina related part of the stack trace in the > HTML generated by the ErrorReportValve; that eases debugging. +1 > - When an exception occurs, and it is a ServletException, I think the > root cause shoul

[5.0] Servlet logging and error messages

2003-07-18 Thread Remy Maucherat
I'm working on the following changes: - Stripping out all the Catalina related part of the stack trace in the HTML generated by the ErrorReportValve; that eases debugging. - When an exception occurs, and it is a ServletException, I think the root cause should be logged, not the external excepti