Bug 56717 - MapperListener does not remove its listeners. It matters for repeated starts of embedded Tomcat,
Summary: MapperListener does not remove its listeners. It matters for repeated starts ...
Alias: None
Product: Tomcat 7
Classification: Unclassified
Component: Catalina (show other bugs)
Version: 7.0.54
Hardware: PC All
: P2 normal (vote)
Target Milestone: ---
Assignee: Tomcat Developers Mailing List
Depends on:
Reported: 2014-07-13 23:54 UTC by Konstantin Kolinko
Modified: 2014-07-17 22:39 UTC (History)
0 users


Note You need to log in before you can comment on or make changes to this bug.
Description Konstantin Kolinko 2014-07-13 23:54:21 UTC
Debugged om the current trunk at 1610244, between 8.0.10 and 8.0.11.
Steps to reproduce:
1. Add the following lines to build.properties

2. I added log.info("...", new RuntimeException()) to Mapper.addHost(), Mapper.removeHost()

3. Run the test case.

The test case performs "tomcat.stop(); tomcat.start();" twice.

Actual behaviour:
During the second tomcat.start() the Mapper.addHost() is called twice with the same host. It results in an error message,

> SEVERE [main] org.apache.catalina.mapper.Mapper.addHost Duplicate Host [localhost]. The name is already used by Host [localhost]. This Host will be ignored.

Expected behaviour:
No such message.

From debug stacktraces that I printed in addHost(), I see that during the second start the two addHost() calls have the following stack traces:
	at org.apache.catalina.mapper.Mapper.addHost(Mapper.java:95)
	at org.apache.catalina.mapper.MapperListener.registerHost(MapperListener.java:295)
	at org.apache.catalina.mapper.MapperListener.lifecycleEvent(MapperListener.java:482)
	at org.apache.catalina.util.LifecycleSupport.fireLifecycleEvent(LifecycleSupport.java:117)
	at org.apache.catalina.util.LifecycleBase.fireLifecycleEvent(LifecycleBase.java:90)
	at org.apache.catalina.util.LifecycleBase.setStateInternal(LifecycleBase.java:402)
	at org.apache.catalina.util.LifecycleBase.start(LifecycleBase.java:168)
	at org.apache.catalina.core.ContainerBase$StartChild.call(ContainerBase.java:1396)
	at org.apache.catalina.core.ContainerBase$StartChild.call(ContainerBase.java:1386)
	at java.util.concurrent.FutureTask.run(FutureTask.java:262)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
	at java.lang.Thread.run(Thread.java:745)

	at org.apache.catalina.mapper.Mapper.addHost(Mapper.java:95)
	at org.apache.catalina.mapper.MapperListener.registerHost(MapperListener.java:295)
	at org.apache.catalina.mapper.MapperListener.startInternal(MapperListener.java:110)
	at org.apache.catalina.util.LifecycleBase.start(LifecycleBase.java:150)
	at org.apache.catalina.core.StandardService.startInternal(StandardService.java:450)
	at org.apache.catalina.util.LifecycleBase.start(LifecycleBase.java:150)
	at org.apache.catalina.core.StandardServer.startInternal(StandardServer.java:760)
	at org.apache.catalina.util.LifecycleBase.start(LifecycleBase.java:150)
	at org.apache.catalina.startup.Tomcat.start(Tomcat.java:340)
	at org.apache.catalina.startup.TomcatBaseTest$TomcatWithFastSessionIDs.start(TomcatBaseTest.java:410)
	at org.apache.catalina.core.TestStandardContext.testFlagFailCtxIfServletStartFails(TestStandardContext.java:917)
(... JUnit code...)

The second stack trace with "MapperListener.startInternal()" is the expected one. The same stack trace is observed during the first tomcat.start().

The first stack trace with "MapperListener.lifecycleEvent()" is an unexpected one. It is caused by the lifecycle listener still being registered on Host.

The MapperListener.startInternal() method contains line
> addListeners(engine);

As listeners are stored in a list, it means that the same listener will be added the second time. I expect that if one adds the thirds tomcat.start() call to the test case, it will cause Mapper.addHost() to be called three times.

That said, I think there is not much of a problem from calling addHost() more than once for the same host (e.g. Mapper.addContextVersion() may call it silently). Thus I'll change Mapper.addHost() to do not treat such repeated calls as an error.
Comment 1 Konstantin Kolinko 2014-07-15 00:48:34 UTC
Fixed in trunk by r1610562 and r1610564. It will be in 8.0.11 onwards.
Comment 2 Konstantin Kolinko 2014-07-17 22:39:24 UTC
Fixed in tc7.0.x by r1611495 and r1611496. It will be in 7.0.55 onwards.