tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Johnny Kewl" <j...@kewlstuff.co.za>
Subject Re: Weird stack trace when reloading a webapp
Date Thu, 07 Aug 2008 22:38:29 GMT

----- Original Message ----- 
From: "Olivier Lamy" <olamy@apache.org>
To: "Tomcat Users List" <users@tomcat.apache.org>
Sent: Thursday, August 07, 2008 11:21 PM
Subject: Re: Weird stack trace when reloading a webapp


> Hi,
> Thanks for your reponse but undeploy/redeploy don't save me same issue 
> :-(.

Damn!.... I was hoping ;)

Well... it looks like a struts framework so I think you need to ask some 
question in their mailing lists...
All I can tell from this is that they seem to be hiding xerces in the 
META-INF folder...
and they have a custom classloader trying to access it....
You seem to be making a filter... and the framework seems to want a web app 
for its classloader... but the webapp is down.

So you calling Tiles or something... and that wants to live in a webapp, but 
you in a filter... maybe, but its the kind of thing that should be showing 
in your dev env?
Is it your application? If not ask the supplier...

Some guesses for you... the order of filters is important... maybe yours 
should be behind all the framework filters...
Perhaps <load-on-start> will get the webapp with all this framework stuff in 
it to start and make the cl happy.

Thats the problem with frameworks... always have to ask more than one user 
group ;)

Its a tricky problem I think and you'll have to talk to the guys that made 
your frameworks...

Good luck... Sorry... outta idea's, maybe google will get you there

> Olivier
>
> 2008/8/7 Johnny Kewl <john@kewlstuff.co.za>:
>>
>> ----- Original Message ----- From: "Olivier Lamy" <olamy@apache.org>
>> To: <users@tomcat.apache.org>
>> Sent: Thursday, August 07, 2008 5:17 PM
>> Subject: Weird stack trace when reloading a webapp
>>
>>
>>> Hi,
>>> I use the ant deploy target :
>>>  <deploy url="${url}" username="${tomcat.user}"
>>> password="${tomcat.password}"
>>>          path="/{webPath}" localWar="${warPath}" update="true"
>>> config="${contextPath}" />
>>>
>>> But in catalina.out, I have the following stack trace :
>>>
>>> INFO: Deploying web application archive dotw-trans.war
>>> 2008-08-07 15:04:38,942 INFO  impl.BasicTilesContainer : Initializing
>>> Tiles2 container. . .
>>> 2008-08-07 15:04:39,177 INFO  impl.BasicTilesContainer : Tiles2
>>> container initialization complete.
>>> 2008-08-07 15:04:39,185 INFO  access.TilesAccess : Publishing
>>> TilesContext for context: org.apache.catalina.core.Appli
>>> cationContextFacade
>>> 2008-08-07 15:04:39,247 INFO  filter.DotwFilterDispatcher : init
>>> DotwFilterDispatcher with encoding : UTF-8
>>> Aug 7, 2008 3:04:40 PM org.apache.catalina.loader.WebappClassLoader
>>> findResourceInternal
>>> INFO: Illegal access: this web application instance has been stopped
>>> already.  Could not load META-INF/services/org.ap
>>> ache.xerces.xni.parser.XMLParserConfiguration.  The eventual following
>>> stack trace is caused by an error thrown for de
>>> bugging purposes as well as to attempt to terminate the thread which
>>> caused the illegal access, and has no functional
>>> impact.
>>> Aug 7, 2008 3:04:40 PM org.apache.catalina.core.StandardContext start
>>> SEVERE: Error filterStart
>>> Aug 7, 2008 3:04:40 PM org.apache.catalina.core.StandardContext start
>>> SEVERE: Context [/dotw-trans] startup failed due to previous errors
>>>
>>> Any idea on the Could not load
>>> META-INF/services/org.apache.xerces.xni.parser.XMLParserConfiguration.
>>>
>>> My en is :
>>> - apache-tomcat-6.0.18
>>> - jdk1.6.0_06
>>> - solaris 10
>>>
>>> Any idea ?
>>>
>>> Thanks,
>>> --
>>> Olivier
>>
>> Oliver is that the whole error message or did you abreviate it?
>> Does is say anything about a POST...
>>
>> I tell you why, on complex applications I've seen something similar...
>>
>> Anyway try this.... before you redeploy... open /manager/html and 
>> UNDEPLOY
>> the old one... then run your script...
>> If it is the same thing... you wont get the error because its actually
>> happening in the undeploy...
>> If that works then add another section to your ant script... first
>> undeploy... then redeploy...
>>
>> If it is the same thing... I've only noticed it on very complex apps....
>> Other thing I've noticed is that even when you get this weird message... 
>> the
>> redeploy does work.
>> And you cant debug it... because TC is killing the thing... ide wont trap
>> it...
>> And I think it only happens when there are classloaders in the webapp
>> classloader... which would explain why the webapp class loader cant let
>> go... the child still has references... anyway try the above, and I dont
>> think its too serious... its a last complaint before it gets a bullet
>> between the eyes ;)
>>
>> .... maybe its the same thing...
>> ---------------------------------------------------------------------------
>> HARBOR : http://www.kewlstuff.co.za/index.htm
>> The most powerful application server on earth.
>> The only real POJO Application Server.
>> See it in Action : http://www.kewlstuff.co.za/cd_tut_swf/whatisejb1.htm
>> ---------------------------------------------------------------------------
>>
>>
>>
>>
>> ---------------------------------------------------------------------
>> To start a new topic, e-mail: users@tomcat.apache.org
>> To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
>> For additional commands, e-mail: users-help@tomcat.apache.org
>>
>>
>
> ---------------------------------------------------------------------
> To start a new topic, e-mail: users@tomcat.apache.org
> To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
> For additional commands, e-mail: users-help@tomcat.apache.org
>
> 


---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
For additional commands, e-mail: users-help@tomcat.apache.org


Mime
View raw message