prediction; context problems

Hi people, Sriram, Remy, Costin, Thanks Sriram for the +1 on this! As you guys have read, I think something is up in the the area of Context information; specifiers, parsing, and storage of current deployment state. I've previously been talking about determinism as one of the problems, but

svn commit: r399165 - /tomcat/tc6.0.x/trunk/java/org/apache/catalina/startup/Constants.java

Author: remm Date: Tue May 2 23:21:27 2006 New Revision: 399165 URL: http://svn.apache.org/viewcvs?rev=399165&view=rev Log: - Change some locations. Modified: tomcat/tc6.0.x/trunk/java/org/apache/catalina/startup/Constants.java Modified: tomcat/tc6.0.x/trunk/java/org/apache/catalina/startup

svn commit: r399096 - /tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/group/interceptors/TcpFailureDetector.java

Author: fhanik Date: Tue May 2 16:59:40 2006 New Revision: 399096 URL: http://svn.apache.org/viewcvs?rev=399096&view=rev Log: added small comment Modified: tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/group/interceptors/TcpFailureDetector.java Modified: t

svn commit: r399095 - /tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/group/interceptors/TcpFailureDetector.java

Author: fhanik Date: Tue May 2 16:58:18 2006 New Revision: 399095 URL: http://svn.apache.org/viewcvs?rev=399095&view=rev Log: implemented the heart beat, the failure detector should be ready to test Modified: tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/gro

svn commit: r399092 - /tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/group/interceptors/TcpFailureDetector.java

Author: fhanik Date: Tue May 2 16:51:02 2006 New Revision: 399092 URL: http://svn.apache.org/viewcvs?rev=399092&view=rev Log: More coding, still need to complete the heartbeat() to make sure there isn't a leak Modified: tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalin

svn commit: r399090 - in /tomcat/container/tc5.5.x/modules/groupcom: doc/ src/share/org/apache/catalina/tribes/group/ src/share/org/apache/catalina/tribes/group/interceptors/ src/share/org/apache/cata

Author: fhanik Date: Tue May 2 16:46:22 2006 New Revision: 399090 URL: http://svn.apache.org/viewcvs?rev=399090&view=rev Log: Refactored the membership layer to not be tied to multicasting, so that same logic can be used for other implementations Started implemented a TcpFailureDetector, so that

Re: context problems: not addressed; summary

On 5/2/06, Remy Maucherat <[EMAIL PROTECTED]> wrote: Costin Manolache wrote: > Did you figure out what he is talking about ? Could you summarise - I > can't parse his rant. To be honest, not really. The point is that I simplified the deployer so that only one meaningful information is considered

svn commit: r399064 - /tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/membership/McastMembership.java

Author: fhanik Date: Tue May 2 14:59:20 2006 New Revision: 399064 URL: http://svn.apache.org/viewcvs?rev=399064&view=rev Log: no need to synchronize on a get Modified: tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/membership/McastMembership.java Modified:

Re: context problems: not addressed; summary

Costin Manolache wrote: Did you figure out what he is talking about ? Could you summarise - I can't parse his rant. To be honest, not really. The point is that I simplified the deployer so that only one meaningful information is considered (for example, when deploying an archive the path come

svn commit: r399059 - /tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/membership/McastMembership.java

Author: fhanik Date: Tue May 2 14:42:43 2006 New Revision: 399059 URL: http://svn.apache.org/viewcvs?rev=399059&view=rev Log: Fixed a small bug with soft shutdown in the membership layer Modified: tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/membership/Mcas

svn commit: r399051 - /tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/group/GroupChannel.java

Author: fhanik Date: Tue May 2 14:19:03 2006 New Revision: 399051 URL: http://svn.apache.org/viewcvs?rev=399051&view=rev Log: moved around code and made the heartbeat thread be low priority Modified: tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/group/GroupC

svn commit: r399050 - in /tomcat/container/tc5.5.x/modules/groupcom: src/share/org/apache/catalina/tribes/group/GroupChannel.java to-do.txt

Author: fhanik Date: Tue May 2 14:15:22 2006 New Revision: 399050 URL: http://svn.apache.org/viewcvs?rev=399050&view=rev Log: Implemented a channel interceptor heartbeat, useful for cleaning up Modified: tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/group/Gr

svn commit: r399042 - /tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/membership/MemberImpl.java

Author: fhanik Date: Tue May 2 13:45:02 2006 New Revision: 399042 URL: http://svn.apache.org/viewcvs?rev=399042&view=rev Log: A member is unique to its session, applications should not assume that a crashed member, that has rejoined is guaranteed to be the same application Modified: tomca

svn commit: r399039 - in /tomcat/container/tc5.5.x/modules/groupcom: doc/ src/share/org/apache/catalina/tribes/ src/share/org/apache/catalina/tribes/membership/ src/share/org/apache/catalina/tribes/tr

Author: fhanik Date: Tue May 2 13:30:17 2006 New Revision: 399039 URL: http://svn.apache.org/viewcvs?rev=399039&view=rev Log: Refactored the sender so that its easy to transfer properties, and not missing important settings Moved shutdown payload to the Member interface so that an app can differ

Re: Security on JDBC Realm new features

with all that said, I think the best option for you is to - create an extension of the current realm instead of patching the existing one - submit the code - and if everyone thinks its useful, the code looks good, we can add it to Tomcat and let the user decide if they wanna use it but I don'

May Chun Chew/FEA/PEC is out of the office.

I will be out of the office starting 05/03/2006 and will not return until 05/04/2006. I will respond to your message when I return. For urgent matters, pls contact Lim Siong Pew at [EMAIL PROTECTED] or 63629384 - To unsubsc

DO NOT REPLY [Bug 37098] - Cannot create attributes conditionally in tags

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG· RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT . ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND· INSERTED IN THE BUG DATABASE. http://issues.apache.org/bugzilla/show_bu

DO NOT REPLY [Bug 39468] - ConcurrentModificationException during webapp reload

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG· RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT . ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND· INSERTED IN THE BUG DATABASE. http://issues.apache.org/bugzilla/show_bu

DO NOT REPLY [Bug 39468] New: - ConcurrentModificationException during webapp reload

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG· RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT . ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND· INSERTED IN THE BUG DATABASE. http://issues.apache.org/bugzilla/show_bu

svn commit: r398982 - in /tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/group: ChannelInterceptorBase.java interceptors/MessageDispatchInterceptor.java

Author: fhanik Date: Tue May 2 10:42:43 2006 New Revision: 398982 URL: http://svn.apache.org/viewcvs?rev=398982&view=rev Log: Added helper method to process the option flags, and the message dispatch interceptor uses the hard coded flag, but can be changed Modified: tomcat/container/tc5.5

svn commit: r398973 - in /tomcat/container/tc5.5.x/modules/groupcom: ./ src/share/org/apache/catalina/tribes/transport/ src/share/org/apache/catalina/tribes/transport/bio/ src/share/org/apache/catalin

Author: fhanik Date: Tue May 2 10:36:53 2006 New Revision: 398973 URL: http://svn.apache.org/viewcvs?rev=398973&view=rev Log: Completed the failed ack feature. Its now optional whether it throws the exception or not, default is not Modified: tomcat/container/tc5.5.x/modules/groupcom/src/sh

svn commit: r398970 - in /tomcat/container/tc5.5.x/modules/groupcom: ./ src/share/org/apache/catalina/tribes/ src/share/org/apache/catalina/tribes/membership/ src/share/org/apache/catalina/tribes/tran

Author: fhanik Date: Tue May 2 10:28:06 2006 New Revision: 398970 URL: http://svn.apache.org/viewcvs?rev=398970&view=rev Log: Fix shutdown message, wasnt getting broadcasted since it caches the serialized data for speed purposes Started implementing the fail ack command so that we can have a ack

Re: Security on JDBC Realm new features

Those are very valid questions. 1. Yes so i guess there could be a default value in the code so that it could always be set to 3 unless set by the admin. 2. I am not sure how to handle this, as if you only had read access to the database then there would be no way to set this up, unl

svn commit: r398964 - in /tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes: ./ membership/

Author: fhanik Date: Tue May 2 08:31:11 2006 New Revision: 398964 URL: http://svn.apache.org/viewcvs?rev=398964&view=rev Log: Added in a stop message for a soft shutdown Added in the ability to send a payload with each broadcast package Modified: tomcat/container/tc5.5.x/modules/groupcom/sr

svn commit: r398966 - /tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/membership/McastServiceImpl.java

Author: fhanik Date: Tue May 2 08:35:09 2006 New Revision: 398966 URL: http://svn.apache.org/viewcvs?rev=398966&view=rev Log: Changed logging to info for a soft shutdown Modified: tomcat/container/tc5.5.x/modules/groupcom/src/share/org/apache/catalina/tribes/membership/McastServiceImpl.java

Re: context problems: not addressed; summary

Did you figure out what he is talking about ? Could you summarise - I can't parse his rant. Costin On 5/2/06, Remy Maucherat <[EMAIL PROTECTED]> wrote: Thomas Whitmore wrote: > Perhaps my bug report came across as a rant, because it had covered > several attempts all of which featured Contexts

Re: context problems: not addressed; summary

Thomas Whitmore wrote: Perhaps my bug report came across as a rant, because it had covered several attempts all of which featured Contexts being non-functional or deployment auto-deleting what it shouldn't have, because I'd already read the docs, searched the web, read the NGs, read the bug repo