To come back to my orginal problem, so far I have had good results using a time-limited cache with a capacity set to 1 second. I know this sounds stupid, but for some reason when we set the cache to none thing start to become very unstable.
I am not in favor keeping it this way, but for now this appears to do well. I am wondering if this cause additional overhead or introduce other risks that we don't know yet? thanks, Frank >>> [EMAIL PROTECTED] 09-Jul-03 14:43:37 >>> This one time, at band camp, Mattias Bogeblad said: MB>I'm not sure I've missed out on something here but the feature MB>expireCache is something I have wanted for a long time and integrated MB>myself a while ago. This posting seems to suggest that Vince Adamos MB>patch is incorporated in Castor since 0.9.4 but I cannot find it either MB>in 0.9.4.x or in 0.9.5. Did I misunderstand you comments Bruce? Mattias, Please see the following page in ViewCVS: http://project.exolab.org/cgi-bin/viewcvs.cgi/castor/src/main/org/exolab/castor/persist/?cvsroot=castor Notice the CVS checkin message for ClassMolder, LRU, LockEngine, ObjectLock and TransactionContext. I just noticed late last night that the Javadoc on the website is not up to date and I notified Keith of this. Hopefully he'll take care of it soon. CVS is the place to look for new features ;-). Bruce -- perl -e 'print unpack("u30","<0G)[EMAIL PROTECTED]&5R\"F9E<G)E=\$\!F<FEI+F-O;0\`\`");' ----------------------------------------------------------- If you wish to unsubscribe from this mailing, send mail to [EMAIL PROTECTED] with a subject of: unsubscribe castor-dev ----------------------------------------------------------- If you wish to unsubscribe from this mailing, send mail to [EMAIL PROTECTED] with a subject of: unsubscribe castor-dev
