you can set up another appender that writes to a file and get it out by name and write into it. alternatively you can print the stacktrace into a buffer and email the whole thing manually.
usually smtp appenders should buffer X lines or Y seconds and send that out, maybe you can just tweak the config of your appender. -igor On Mon, Jun 29, 2009 at 8:01 AM, Jeremy Levy<[email protected]> wrote: > Igor, > Can you give me some direction on how to go about getting the "e". > > In my own RequestCycle I've done the following: > > �...@override > protected void logRuntimeException(RuntimeException e) { > super.logRuntimeException(e); > log.error(e); > } > > Which doesn't help because I'm just doing the same thing twice: > > 2009-06-29 01:26:09,721 ERROR Web [RequestCycle] : > java.lang.NullPointerException > 2009-06-29 01:26:09,780 ERROR Web [MMRequestCycle] : > java.lang.NullPointerException > > I tried e.printStackTrace(); but log4j SMTPAppender takes each line of the > stacktrace and emails it which generates way to much volume. Any > suggestions? > > Jeremy > > > > On Fri, Jun 19, 2009 at 2:20 PM, Jeremy Levy <[email protected]> wrote: > >> I'll override the method and let you know the results. >> Jeremy >> >> >> On Fri, Jun 19, 2009 at 11:35 AM, Igor Vaynberg >> <[email protected]>wrote: >> >>> right, so where is the stacktrace from the "e" given to the logger? >>> >>> -igor >>> >>> On Fri, Jun 19, 2009 at 6:39 AM, Jeremy Levy<[email protected]> wrote: >>> > Igor, >>> > It's happening on line # 1483 of RequestCycle which corresponds to: >>> > >>> > * Called when an unrecoverable runtime exception during request cycle >>> > handling occurred, which >>> > * will result in displaying a user facing error page. Clients can >>> override >>> > this method in case >>> > * they want to customize logging. NOT called for {...@link >>> > PageExpiredException page expired >>> > * exceptions}. >>> > * >>> > * @param e >>> > * the runtime exception >>> > */ >>> > protected void logRuntimeException(RuntimeException e) >>> > { >>> > log.error(e.getMessage(), e); >>> > } >>> > >>> > Jeremy >>> > >>> > On Thu, Jun 18, 2009 at 11:46 AM, Igor Vaynberg < >>> [email protected]>wrote: >>> > >>> >> that is rather strange, there should be the stack trace. why dont you >>> >> change your logger to show the line numbers so we can see where the >>> >> log statement is coming from. >>> >> >>> >> -igor >>> >> >>> >> On Thu, Jun 18, 2009 at 7:53 AM, Jeremy Levy<[email protected]> wrote: >>> >> > Per, >>> >> > There is no stack dump, that is the entire output. >>> >> > >>> >> > J >>> >> > >>> >> > On Thu, Jun 18, 2009 at 10:48 AM, Per Lundholm < >>> [email protected] >>> >> >wrote: >>> >> > >>> >> >> No. ;-) >>> >> >> >>> >> >> Are you suggesting that the version of Wicket matters? >>> >> >> >>> >> >> How does the stack dump look in your logs? >>> >> >> >>> >> >> /Per >>> >> >> >>> >> >> >>> >> >> On Thu, Jun 18, 2009 at 4:25 PM, Jeremy Levy<[email protected]> >>> wrote: >>> >> >> > I see the following a few times a day, this is with Wicket 1.3.6. >>> It >>> >> >> > results in a 500 being displayed to the user... >>> >> >> > >>> >> >> > 2009-06-18 00:53:09,485 ERROR Web [RequestCycle] : >>> >> >> > java.lang.NullPointerException >>> >> >> > >>> >> >> > I realize this isn't much to go on, any ideas? >>> >> >> > >>> >> >> > j >>> >> >> > >>> >> >> >>> >> >> >>> --------------------------------------------------------------------- >>> >> >> 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] >>> >> >>> >> >>> > >>> >>> --------------------------------------------------------------------- >>> 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]
