[Bug 59659] New: WsSession leaks (is not GC) if error occurs during the closing process

https://bz.apache.org/bugzilla/show_bug.cgi?id=59659 Bug ID: 59659 Summary: WsSession leaks (is not GC) if error occurs during the closing process Product: Tomcat 8 Version: 8.5.2 Hardware: PC Status: NEW

[Bug 59661] New: MailSessionFactory ignores system properties

https://bz.apache.org/bugzilla/show_bug.cgi?id=59661 Bug ID: 59661 Summary: MailSessionFactory ignores system properties Product: Tomcat 9 Version: 9.0.0.M6 Hardware: All OS: All Status: NEW Severity: norm

[Bug 59661] MailSessionFactory ignores system properties

https://bz.apache.org/bugzilla/show_bug.cgi?id=59661 Remy Maucherat changed: What|Removed |Added Status|NEW |RESOLVED Resolution|---

[Bug 59661] MailSessionFactory ignores system properties

https://bz.apache.org/bugzilla/show_bug.cgi?id=59661 Michael Osipov <1983-01...@gmx.net> changed: What|Removed |Added Resolution|WONTFIX |FIXED --- Comment

[Bug 59661] MailSessionFactory ignores system properties

https://bz.apache.org/bugzilla/show_bug.cgi?id=59661 Michael Osipov <1983-01...@gmx.net> changed: What|Removed |Added Resolution|FIXED |--- S

Re: Early Access builds of JDK 9 b118 & JDK 9 with Project Jigsaw, b118 (#4987) are available on java.net

On 01/06/2016 15:45, Rory O'Donnell wrote: > Hi Mark, > > JDK-8158237 : JVMTI > hides critical debug information for memory leak tracing > > Is closed as a duplicate of JDK-8033735 > : make > Thr

Re: Early Access builds of JDK 9 b118 & JDK 9 with Project Jigsaw, b118 (#4987) are available on java.net

Thanks Mark, I'll update the bug. Rgds,Rory On 03/06/2016 12:32, Mark Thomas wrote: On 01/06/2016 15:45, Rory O'Donnell wrote: Hi Mark, JDK-8158237 : JVMTI hides critical debug information for memory leak tracing Is closed as a duplicate of

HTTP/2 optimizations and edge cases

Hi, With direct connect having been hacked in (err, I mean, "implemented"), it is (a lot) easier to do meaningful performance tests. h2load is a drop in replacement of ab that uses HTTP/2, and it allowed doing some easy profiling. The good news is that the code seems to be well optimized already

svn commit: r1746718 - /tomcat/trunk/java/org/apache/catalina/ha/context/ReplicatedContext.java

Author: markt Date: Fri Jun 3 13:53:11 2016 New Revision: 1746718 URL: http://svn.apache.org/viewvc?rev=1746718&view=rev Log: Fix Eclipse Neon nag Modified: tomcat/trunk/java/org/apache/catalina/ha/context/ReplicatedContext.java Modified: tomcat/trunk/java/org/apache/catalina/ha/context/Re

Re: HTTP/2 optimizations and edge cases

On 03/06/2016 14:36, Rémy Maucherat wrote: > Hi, > > With direct connect having been hacked in (err, I mean, "implemented"), it > is (a lot) easier to do meaningful performance tests. h2load is a drop in > replacement of ab that uses HTTP/2, and it allowed doing some easy > profiling. > > The goo

svn commit: r1746721 - /tomcat/trunk/test/org/apache/coyote/http11/TestHttp11Processor.java

Author: markt Date: Fri Jun 3 14:27:35 2016 New Revision: 1746721 URL: http://svn.apache.org/viewvc?rev=1746721&view=rev Log: Rename. Nature of Servlet changed as the test evolved. Modified: tomcat/trunk/test/org/apache/coyote/http11/TestHttp11Processor.java Modified: tomcat/trunk/test/org/

svn commit: r1746722 - in /tomcat/tc8.5.x/trunk: ./ test/org/apache/coyote/http11/TestHttp11Processor.java

Author: markt Date: Fri Jun 3 14:38:41 2016 New Revision: 1746722 URL: http://svn.apache.org/viewvc?rev=1746722&view=rev Log: Minor clean-up Modified: tomcat/tc8.5.x/trunk/ (props changed) tomcat/tc8.5.x/trunk/test/org/apache/coyote/http11/TestHttp11Processor.java Propchange: tomcat/t

svn commit: r1746724 - /tomcat/trunk/webapps/docs/changelog.xml

Author: markt Date: Fri Jun 3 14:41:07 2016 New Revision: 1746724 URL: http://svn.apache.org/viewvc?rev=1746724&view=rev Log: Typo Modified: tomcat/trunk/webapps/docs/changelog.xml Modified: tomcat/trunk/webapps/docs/changelog.xml URL: http://svn.apache.org/viewvc/tomcat/trunk/webapps/docs

svn commit: r1746725 - in /tomcat/tc8.5.x/trunk: ./ java/org/apache/coyote/http11/Http11Processor.java test/org/apache/coyote/http11/TestHttp11Processor.java webapps/docs/changelog.xml

Author: markt Date: Fri Jun 3 14:43:07 2016 New Revision: 1746725 URL: http://svn.apache.org/viewvc?rev=1746725&view=rev Log: If an async dispatch results in the completion of request processing, ensure that any remaining request body is swallowed before starting the processing of the next requ

svn commit: r1746726 - /tomcat/tc8.0.x/trunk/java/org/apache/jasper/compiler/TagLibraryInfoImpl.java

Author: markt Date: Fri Jun 3 14:49:17 2016 New Revision: 1746726 URL: http://svn.apache.org/viewvc?rev=1746726&view=rev Log: Silence an Eclipse nag Modified: tomcat/tc8.0.x/trunk/java/org/apache/jasper/compiler/TagLibraryInfoImpl.java Modified: tomcat/tc8.0.x/trunk/java/org/apache/jasper/

svn commit: r1746728 - in /tomcat/tc8.0.x/trunk: ./ java/org/apache/coyote/http11/AbstractHttp11Processor.java

Author: markt Date: Fri Jun 3 14:53:55 2016 New Revision: 1746728 URL: http://svn.apache.org/viewvc?rev=1746728&view=rev Log: Simplify. If getErrorState().isError() is true then response.getErrorException() will always be null. Modified: tomcat/tc8.0.x/trunk/ (props changed) tomcat/t

svn commit: r1746729 - /tomcat/tc8.0.x/trunk/test/org/apache/coyote/http11/TestAbstractHttp11Processor.java

Author: markt Date: Fri Jun 3 14:58:37 2016 New Revision: 1746729 URL: http://svn.apache.org/viewvc?rev=1746729&view=rev Log: Minor clean-up Modified: tomcat/tc8.0.x/trunk/test/org/apache/coyote/http11/TestAbstractHttp11Processor.java Modified: tomcat/tc8.0.x/trunk/test/org/apache/coyote/

Re: HTTP/2 optimizations and edge cases

2016-06-03 16:11 GMT+02:00 Mark Thomas : > On 03/06/2016 14:36, Rémy Maucherat wrote: > > Hi, > > > > With direct connect having been hacked in (err, I mean, "implemented"), > it > > is (a lot) easier to do meaningful performance tests. h2load is a drop in > > replacement of ab that uses HTTP/2, a

svn commit: r1746732 - in /tomcat/tc8.0.x/trunk: ./ java/org/apache/coyote/AbstractProtocol.java java/org/apache/coyote/http11/AbstractHttp11Processor.java test/org/apache/coyote/http11/TestAbstractHt

Author: markt Date: Fri Jun 3 15:31:07 2016 New Revision: 1746732 URL: http://svn.apache.org/viewvc?rev=1746732&view=rev Log: If an async dispatch results in the completion of request processing, ensure that any remaining request body is swallowed before starting the processing of the next requ

Re: HTTP/2 optimizations and edge cases

On 03/06/2016 15:59, Rémy Maucherat wrote: > 2016-06-03 16:11 GMT+02:00 Mark Thomas : > >> On 03/06/2016 14:36, Rémy Maucherat wrote: >>> Hi, >>> >>> With direct connect having been hacked in (err, I mean, "implemented"), >> it >>> is (a lot) easier to do meaningful performance tests. h2load is a

svn commit: r1746734 - /tomcat/tc8.0.x/trunk/test/org/apache/coyote/http11/TestAbstractHttp11Processor.java

Author: markt Date: Fri Jun 3 15:39:49 2016 New Revision: 1746734 URL: http://svn.apache.org/viewvc?rev=1746734&view=rev Log: Remove the debug code Modified: tomcat/tc8.0.x/trunk/test/org/apache/coyote/http11/TestAbstractHttp11Processor.java Modified: tomcat/tc8.0.x/trunk/test/org/apache/

svn commit: r1746735 - in /tomcat/tc7.0.x/trunk: ./ java/org/apache/coyote/AbstractProtocol.java java/org/apache/coyote/http11/AbstractHttp11Processor.java test/org/apache/coyote/http11/TestAbstractHt

Author: markt Date: Fri Jun 3 15:41:38 2016 New Revision: 1746735 URL: http://svn.apache.org/viewvc?rev=1746735&view=rev Log: If an async dispatch results in the completion of request processing, ensure that any remaining request body is swallowed before starting the processing of the next requ

[Bug 59650] NIO2 connector does not respect executor bounds, reuses NIO.2 threads for processing request on connections

https://bz.apache.org/bugzilla/show_bug.cgi?id=59650 Ravi Sanwal changed: What|Removed |Added Resolution|INVALID |--- Status|RESOLVED

[Bug 59650] NIO2 connector does not respect executor bounds, reuses NIO.2 threads for processing request on connections

https://bz.apache.org/bugzilla/show_bug.cgi?id=59650 Remy Maucherat changed: What|Removed |Added Severity|normal |enhancement Resolution|---

svn commit: r1746750 - in /tomcat/tc8.0.x/trunk/java/org/apache: coyote/http11/AbstractHttp11Processor.java jasper/compiler/TagLibraryInfoImpl.java

Author: markt Date: Fri Jun 3 19:31:06 2016 New Revision: 1746750 URL: http://svn.apache.org/viewvc?rev=1746750&view=rev Log: Tab police. Sorry - clean IDE install and I missed some required settings. Modified: tomcat/tc8.0.x/trunk/java/org/apache/coyote/http11/AbstractHttp11Processor.java

[Bug 59650] NIO2 connector does not respect executor bounds, reuses NIO.2 threads for processing request on connections

https://bz.apache.org/bugzilla/show_bug.cgi?id=59650 --- Comment #4 from Ravi Sanwal --- Can it at least give the user some control put a cap on the number of threads? May be I am not explaining my problem correctly. Regardless. Tomcat NIO2 connector still doesn't follow the rules of a defined ex

[Bug 59650] NIO2 connector does not respect executor bounds, reuses NIO.2 threads for processing request on connections

https://bz.apache.org/bugzilla/show_bug.cgi?id=59650 --- Comment #5 from Ravi Sanwal --- Typo in previous comment. It should be. We should at lease fix the connector documentation for this behavior and explain that it is better to not use an external executor *but* an internal one specific to th

[Bug 59650] NIO2 connector does not respect executor bounds, reuses NIO.2 threads for processing request on connections

https://bz.apache.org/bugzilla/show_bug.cgi?id=59650 --- Comment #6 from Remy Maucherat --- So you're saying effectively that the NIO2 connector would be better off if it used *two* separate executors: one ExecutorService internally (NIO2 needs one), and an external executor. And most processing