Skip to content
Fetching contributors…
Cannot retrieve contributors at this time
127 lines (108 sloc) 5.96 KB
================================================================================
Licensed to the Apache Software Foundation (ASF) under one or more
contributor license agreements. See the NOTICE file distributed with
this work for additional information regarding copyright ownership.
The ASF licenses this file to You under the Apache License, Version 2.0
(the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
================================================================================
$Id$
=================================
Apache Tomcat 6.0 Patch Proposals
=================================
RELEASE SHOWSTOPPERS:
PATCHES ACCEPTED TO BACKPORT:
[ start all new proposals below, under PATCHES PROPOSED. ]
PATCHES PROPOSED TO BACKPORT:
[ New proposals should be added at the end of the list ]
* Fix the maven stuff for the maven repo.
Before it does't find tomcat-juli.jar and the remoteRepository seems broken .
http://people.apache.org/~jfclere/patches/maven.patch.100711
+1: jfclere
-1: fhanik - easier to pass in the root path (lib/bin) to the macro instead of hacking around it
if we remove the SCP auto feature, then there should be something to replace it with
(http://ant.apache.org/manual/OptionalTasks/scp.html)
+1: kkolinko: +1 for the updated patch (maven.patch.100711)
I still think that it is not so good to remove the old code of <remoteRepository/>,
because it ignores ${maven.repo.url} parameter provided by deploy-snapshot,
deploy-staging and deploy-release targets. Maybe leave old code as a comment
and fix it later if needed? I think that actually nobody besides the release manager
uses this, so I am letting this pass.
* Add StuckThreadDetectionValve
https://github.com/sylvainlaurent/tomcat60/commit/252334f958877221ecb2dc64ee0fd12bb77e360b
+1: slaurent
+1: kkolinko: several comments are below
-1:
kkolinko: To view/download this as a patch file you add ".diff" to the URL above. That is:
https://github.com/sylvainlaurent/tomcat60/commit/252334f958877221ecb2dc64ee0fd12bb77e360b.diff
kkolinko: Minor glitches:
- In mbeans-descriptors.xml:
- There are no properties "asyncSupported", "stateName" in TC6 version of this valve.
- In valve.xml:
- s/tomcat log/Tomcat log/ (or Apache Tomcat log)
- In StuckThreadDetectionValve.java:
- private classes CompletedStuckThread, MonitoredThread: make them static (r1175897)
- Use Long.valueOf() and explicit boxing/unboxing (r1138132, r1175897)
- getStuckThreadIds() returns a list of ids. It might be useful to
have a similar method that returns Thread.getName() names.
* Fix https://issues.apache.org/bugzilla/show_bug.cgi?id=50570
Apply FIPS mode patch from TC7:
http://svn.apache.org/viewvc?rev=1199985&view=rev
http://svn.apache.org/viewvc?rev=1200004&view=rev
http://svn.apache.org/viewvc?rev=1200448&view=rev
+1: schultz, jfclere
-1: kkolinko: It needs to include more preceding and followup
changes. Updated patch proposed below.
* Backport AprLifecycleListener improvements from TC7
This includes:
- Use consistent external synchronization object for private
initialization methods. Turn initialization flags to false after APR
has been terminated. That is, to terminate it only once.
(rev. 946671 + several after it)
- Use ExceptionUtils.handleThrowable() for fatal errors
(rev. 1001904)
- Do not allow to change SSL options if SSL engine has already been
initialised.
(rev. 1199985 + several after it)
- Fix https://issues.apache.org/bugzilla/show_bug.cgi?id=50570
Enable FIPS mode to be set in AprLifecycleListener.
Based upon a patch from Chris Beckey.
(rev. 1199985 + several after it)
kkolinko: I tested only that it does not break things with the current
released tcnative version 1.1.22 that does not have FIPS mode support.
http://people.apache.org/~kkolinko/patches/2011-11-12_tc6_AprLifecycleListener.patch
+1: kkolinko
-1:
* Fix autodeployment of applications that have configuration errors.
If autodeployment fails, create DeployedApplication object and register
what we deployed (xml or war or dir - a single file) as redeployResource.
If it is updated the application will be redeployed.
+ ContainerBase patch adds display of failed apps in manager. They can be
listed and undeployed.
http://people.apache.org/~kkolinko/patches/2011-11-14_tc6_HostConfig.patch
http://people.apache.org/~kkolinko/patches/2011-11-14_tc6_ContainerBase.patch
+1: kkolinko
-1:
PATCHES/ISSUES THAT ARE STALLED
* Backport JSP unloading patch (BZ48358).
The patch has substantially changed since the original version.
Original revisions are: 937787, 1028377, 1028389, 1028396, 1028861, 1028862, 1028863,
1028935, 1028939, 1028940, 1028944, 1028950, 1030014, 1030037
Combined TC 6 patch:
http://people.apache.org/~rjung/patches/BZ48358-JSP_Unloading-TC6_20101118.patch
+1: rjung
-0: markt - http://markmail.org/thread/777be426ulcfmdd3 suggests there may be
a memory leak in this code somewhere. I'd like to get to the
bottom of that before porting this
rjung: I started a discussion about JSPs and memory on the dev list.
The updated patch fixes a race condition.
We can stall this item until we get some feedback about 7.0.5.
markt - r1172614 needs to be included in this proposal. With that patch,
my testing shows that the unloading works as designed
-1:
Jump to Line
Something went wrong with that request. Please try again.