svn commit: r1036286 - /tomcat/trunk/java/org/apache/catalina/session/ManagerBase.java

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 23:36:25 2010 New Revision: 1036286 URL: http://svn.apache.org/viewvc?rev=1036286&view=rev Log: Narrow the manager-wide sync Modified: tomcat/trunk/java/org/apache/catalina/session/ManagerBase.java Modified: tomcat/trunk/java/org/apache/catalina/session/Manager

svn commit: r1036281 - in /tomcat/trunk/java/org/apache/catalina/session: ManagerBase.java StandardSession.java

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 23:24:25 2010 New Revision: 1036281 URL: http://svn.apache.org/viewvc?rev=1036281&view=rev Log: Refactor the calculation of session manager statistics to the session manager as suggested by kkolinko Modified: tomcat/trunk/java/org/apache/catalina/session/Manage

Re: svn commit: r1036021 - /tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java

2010-11-17 Thread Mark Thomas
On 17/11/2010 19:49, Konstantin Kolinko wrote: My impression looking on the above cited code of StandardSession is that all that arithmetics should be done inside the manager. Then the manager can be optimized how to store it. +1, although it is slightly complicated by the PersistentManager th

DO NOT REPLY [Bug 50288] Uploading a war file that already exists should auto-undeploy the old one

2010-11-17 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=50288 David Becker changed: What|Removed |Added CC||da...@sogeeky.net -- Configure bug

DO NOT REPLY [Bug 50288] New: Uploading a war file that already exists should auto-undeploy the old one

2010-11-17 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=50288 Summary: Uploading a war file that already exists should auto-undeploy the old one Product: Tomcat 6 Version: unspecified Platform: All OS/Version: All Status

svn commit: r1036232 - /tomcat/site/trunk/docs/images/tomcat.gif

2010-11-17 Thread mturk
Author: mturk Date: Wed Nov 17 21:21:52 2010 New Revision: 1036232 URL: http://svn.apache.org/viewvc?rev=1036232&view=rev Log: Logo with tm. Required a litte bit larger image Modified: tomcat/site/trunk/docs/images/tomcat.gif Modified: tomcat/site/trunk/docs/images/tomcat.gif URL: http://sv

svn commit: r1036231 - /tomcat/site/trunk/xdocs/images/tomcat.gif

2010-11-17 Thread mturk
Author: mturk Date: Wed Nov 17 21:21:26 2010 New Revision: 1036231 URL: http://svn.apache.org/viewvc?rev=1036231&view=rev Log: Logo with tm. Required a litte bit larger image Modified: tomcat/site/trunk/xdocs/images/tomcat.gif Modified: tomcat/site/trunk/xdocs/images/tomcat.gif URL: http://

DO NOT REPLY [Bug 50228] PermGen leak when calling JspFragment.invoke() with a custom Writer

2010-11-17 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=50228 --- Comment #3 from Konstantin Kolinko 2010-11-17 16:01:33 EST --- I will have to revert r1032232 It breaks evaluation of tag bodies when there is no writer, e.g. value MyVar: [${myvar}] results in "MyVar: []" being printed. A working

RE: setclasspath vs. setenv

2010-11-17 Thread Caldarale, Charles R
> From: Thangalin [mailto:thanga...@gmail.com] > Subject: setclasspath vs. setenv > I find it odd that nowhere in that file is the classpath environment > variable actually set. That's because the CLASSPATH environment variable is inherently evil. Don't ever use it (goes double for Tomcat). A

Re: setclasspath vs. setenv

2010-11-17 Thread Konstantin Kolinko
2010/11/17 Thangalin : > Hi, > > Judging by its name alone, I would have thought that "setclasspath" (.bat or > .sh) would be a place to put custom code that sets the classpath before > Tomcat starts, or stops. > Setting CLASSPATH explicitly is a bad idea. The cases when it is needed in Tomcat ar

Re: setclasspath vs. setenv

2010-11-17 Thread Pid
On 17/11/2010 20:35, Thangalin wrote: > Hi, > > Judging by its name alone, I would have thought that "setclasspath" (.bat or > .sh) would be a place to put custom code that sets the classpath before > Tomcat starts, or stops. > > I find it odd that nowhere in that file is the classpath environmen

setclasspath vs. setenv

2010-11-17 Thread Thangalin
Hi, Judging by its name alone, I would have thought that "setclasspath" (.bat or .sh) would be a place to put custom code that sets the classpath before Tomcat starts, or stops. I find it odd that nowhere in that file is the classpath environment variable actually set. The script appears to ensur

Re: svn commit: r1036021 - /tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java

2010-11-17 Thread Konstantin Kolinko
2010/11/17 Mark Thomas : > On 17/11/2010 13:05, ma...@apache.org wrote: >> Author: markt >> Date: Wed Nov 17 13:05:20 2010 >> New Revision: 1036021 >> >> URL: http://svn.apache.org/viewvc?rev=1036021&view=rev >> Log: >> Increase length of tests to >> a) reduce impact of start-up overhead >> b) make

DO NOT REPLY [Bug 50275] Mod_JK does not set chunk size for response with chunked transfer encoding

2010-11-17 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=50275 --- Comment #2 from Andre Piwoni 2010-11-17 14:20:21 EST --- Tim, Few questions come to mind: (1) If this is not Apache/mod_jk bug, are you saying this is AJP connector bug? (2) If none of the above, are you saying that setting Transfer-

DO NOT REPLY [Bug 50282] Reference from javax.security.auth.login.LoginContext.contextClassLoader initalized in static code block prevents an undeployed Web application from being garbage collected (s

2010-11-17 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=50282 Justin Almquist changed: What|Removed |Added Platform|PC |Macintosh OS/Version|

Re: svn commit: r1036021 - /tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java

2010-11-17 Thread Mark Thomas
On 17/11/2010 13:35, Mark Thomas wrote: > On 17/11/2010 13:05, ma...@apache.org wrote: >> Author: markt >> Date: Wed Nov 17 13:05:20 2010 >> New Revision: 1036021 >> >> URL: http://svn.apache.org/viewvc?rev=1036021&view=rev >> Log: >> Increase length of tests to >> a) reduce impact of start-up over

svn commit: r1036153 - /tomcat/jk/trunk/native/common/jk_ajp_common.c

2010-11-17 Thread mturk
Author: mturk Date: Wed Nov 17 18:20:20 2010 New Revision: 1036153 URL: http://svn.apache.org/viewvc?rev=1036153&view=rev Log: Although jk_shutdown_socket preserves the errno across the call, use the error returned from the jk_tcp_socket_recvfull. This will make sure the rare ocasion with errno

svn commit: r1036150 - /tomcat/trunk/webapps/docs/changelog.xml

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 18:06:20 2010 New Revision: 1036150 URL: http://svn.apache.org/viewvc?rev=1036150&view=rev Log: Add recent performance improvements to changelog Modified: tomcat/trunk/webapps/docs/changelog.xml Modified: tomcat/trunk/webapps/docs/changelog.xml URL: http://svn

svn commit: r1036145 - in /tomcat/trunk: java/org/apache/catalina/session/ManagerBase.java test/org/apache/catalina/session/Benchmarks.java

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 17:59:49 2010 New Revision: 1036145 URL: http://svn.apache.org/viewvc?rev=1036145&view=rev Log: Session manager performance Switch to a queue of randomInputStreams Significant performance improvement on multi-core non-Windows platforms Modified: tomcat/trunk/jav

svn commit: r1036129 - in /tomcat/trunk: java/org/apache/catalina/ha/session/ java/org/apache/catalina/session/ test/org/apache/catalina/session/

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 17:37:19 2010 New Revision: 1036129 URL: http://svn.apache.org/viewvc?rev=1036129&view=rev Log: Session manager performance Switch to only allowing changes to randomFile to take effect when the Manager next starts. This will simplify replacing RandomIS with a queue

svn commit: r1036119 - in /tomcat/trunk/java/org/apache/catalina: ha/session/DeltaManager.java session/PersistentManagerBase.java session/StandardManager.java

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 17:21:13 2010 New Revision: 1036119 URL: http://svn.apache.org/viewvc?rev=1036119&view=rev Log: Tab police - sorry rebuilt my laptop and forgot to configure Eclipse for spaces Modified: tomcat/trunk/java/org/apache/catalina/ha/session/DeltaManager.java tomca

svn commit: r1036100 - /tomcat/trunk/java/org/apache/catalina/session/ManagerBase.java

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 16:27:51 2010 New Revision: 1036100 URL: http://svn.apache.org/viewvc?rev=1036100&view=rev Log: Rename attributes to align with getter/setter to make code easier to read Modified: tomcat/trunk/java/org/apache/catalina/session/ManagerBase.java Modified: tomcat/t

svn commit: r1036099 - in /tomcat/trunk/java/org/apache/catalina: ha/session/BackupManager.java ha/session/DeltaManager.java session/ManagerBase.java session/PersistentManagerBase.java session/Standar

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 16:26:23 2010 New Revision: 1036099 URL: http://svn.apache.org/viewvc?rev=1036099&view=rev Log: Reduce code duplication Initialise the random number generator once in ManagerBase Modified: tomcat/trunk/java/org/apache/catalina/ha/session/BackupManager.java t

DO NOT REPLY [Bug 50285] New: Standard HTTP and AJP connectors silently ignore attribute typos

2010-11-17 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=50285 Summary: Standard HTTP and AJP connectors silently ignore attribute typos Product: Tomcat Connectors Version: unspecified Platform: All OS/Version: All Status

svn commit: r1036039 - /tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 13:56:37 2010 New Revision: 1036039 URL: http://svn.apache.org/viewvc?rev=1036039&view=rev Log: Update figures Modified: tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java Modified: tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java URL:

Re: svn commit: r1036021 - /tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java

2010-11-17 Thread Mark Thomas
On 17/11/2010 13:05, ma...@apache.org wrote: > Author: markt > Date: Wed Nov 17 13:05:20 2010 > New Revision: 1036021 > > URL: http://svn.apache.org/viewvc?rev=1036021&view=rev > Log: > Increase length of tests to > a) reduce impact of start-up overhead > b) make it easier to profile Looks like t

svn commit: r1036021 - /tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 13:05:20 2010 New Revision: 1036021 URL: http://svn.apache.org/viewvc?rev=1036021&view=rev Log: Increase length of tests to a) reduce impact of start-up overhead b) make it easier to profile Modified: tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java

svn commit: r1036019 - in /tomcat/trunk: java/org/apache/catalina/ha/session/ java/org/apache/catalina/session/ test/org/apache/catalina/session/

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 12:59:06 2010 New Revision: 1036019 URL: http://svn.apache.org/viewvc?rev=1036019&view=rev Log: Session manager performance Focused on Windows. Use a queue of Random's to generate session ID to remove sync bottleneck on random.nextBytes(bytes) Timings suggest some b

svn commit: r1035976 - /tomcat/tc6.0.x/trunk/STATUS.txt

2010-11-17 Thread kkolinko
Author: kkolinko Date: Wed Nov 17 10:53:57 2010 New Revision: 1035976 URL: http://svn.apache.org/viewvc?rev=1035976&view=rev Log: Propose update patch for path parameter handling. Remove obsolete comment from Tim's patch. Modified: tomcat/tc6.0.x/trunk/STATUS.txt Modified: tomcat/tc6.0.x/tru

svn commit: r1035975 - in /tomcat/trunk: java/org/apache/catalina/session/ManagerBase.java test/org/apache/catalina/session/Benchmarks.java

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 10:52:58 2010 New Revision: 1035975 URL: http://svn.apache.org/viewvc?rev=1035975&view=rev Log: Session manager performance Focused on Windows. No need for DataInputStream, so remove it. Ensure randomIS is consistent with devRandomSource including when devRandomSour

svn commit: r1035974 - /tomcat/trunk/test/org/apache/catalina/session/Threading.java

2010-11-17 Thread markt
Author: markt Date: Wed Nov 17 10:48:16 2010 New Revision: 1035974 URL: http://svn.apache.org/viewvc?rev=1035974&view=rev Log: Add some test code I used to check thread-safety of FileInputStream on various platforms Added: tomcat/trunk/test/org/apache/catalina/session/Threading.java (with

svn commit: r1035973 - /tomcat/trunk/java/org/apache/catalina/connector/CoyoteAdapter.java

2010-11-17 Thread kkolinko
Author: kkolinko Date: Wed Nov 17 10:41:03 2010 New Revision: 1035973 URL: http://svn.apache.org/viewvc?rev=1035973&view=rev Log: CoyoteAdapter.parsePathParameters: - do not reevaluate start/end, because we already have them in local variables - move warnedEncoding flag outside the loop, because o

Re: [taglibs] Tab police

2010-11-17 Thread sebb
On 17 November 2010 07:07, Jeremy Boynes wrote: > For a one-time reformat I'd probably go with the one in IDEA followed by > manual review for where it messed up line breaks etc. > > For ongoing checks, we could add a checkstyle report to the build using the > same ruleset as Tomcat (or a strict

Re: [taglibs] Tab police

2010-11-17 Thread sebb
On 17 November 2010 03:45, Rex Wang wrote: > What's the formatter you plan to use? Is it public and other committer > obeyed? Otherwise, it will mess up again in future.. > > Anyway, +1 for tab replacement with 4 spaces. The current code asumes tabs consume 8 spaces, at least in the example code

DO NOT REPLY [Bug 48358] JSP-unloading reloaded

2010-11-17 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=48358 --- Comment #11 from Christian Brandel 2010-11-17 03:43:21 EST --- Any chances that this patch might make it into the 6.0.x line? -- Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email --- You are receiving

DO NOT REPLY [Bug 50275] Mod_JK does not set chunk size for response with chunked transfer encoding

2010-11-17 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=50275 Tim Whittington changed: What|Removed |Added Status|NEW |RESOLVED Resolution|