[Building Sakai] Starting Nakamura on sakai3sam.caret fails

[Christian Vuerings]

When I tried to build Nakamura with Java 1.6 this morning I received the following error [1]
It's just a missing artifact that couldn't be downloaded.
Usually when I encounter such an error it is "magically" fixed the next day.

I don't really know how this works but is apache building sling and then putting snapshots/artifacts online?
And if something goes wrong, the artifacts aren't released online?

Btw the Nightly K2/Stable UX is also down [2] (503).
The build seemed to have been successful [3] but running it wasn't.


- Christian

[1] http://pastebin.com/f17191f3
[2] http://sakai3-dev.uits.indiana.edu:8080/dev/index.html
[3] http://sakai3-dev.uits.indiana.edu/hybrid/logs/sakai3-build.log.txt

On 20 Feb 2010, at 08:59, Ian Boston wrote:

> What version of java did you build with  and what version of java are you running with.
> IIRC the OSGi BND tool gets the location of javax.trasaction wrong cross jvm version. I don't know why at the moment.
> 
> IIRC the blog post relates to a general OSGi startup problem.
> 
> On 19 Feb 2010, at 20:36, Christian Vuerings wrote:
> 
>> Today I did a complete fresh build of Nakamura on sakai3sam (prototype server).
>> 
>> fresh build:
>> - completely remove the .m2/repository folder
>> - git pull ieb
>> - remove the sling folder
>> 
>> The build went fine, but as soon as I start it I receive the following output [1].
>> 
>> I use the following command to start the server:
>> java -server -Xms256m -Xmx1024m -XX:MaxPermSize=256m -Djava.awt.headless=true -Duser.timezone=Europe/London -jar org.sakaiproject.nakamura.app-0.3-SNAPSHOT.jar -p 8080 -l DEBUG
>> 
>> It just stops at
>> 	org.ops4j.pax.web.service.internal.Activator Started pax http service
>> and keeps giving me the following message over and over again:
>> 	DEBUG: Constraint violation for 3.0 detected; module can see javax.transaction.xa from [0] and javax.transaction.xa from [74.0]
>> This causes the Java process to consume 98% of the CPU.
>> In order to shut down the process I had to use kill -9 2332 with root access (killall java didn't work).
>> 
>> When I googled around to find a solution, I discovered a blogpost from someone who encountered a similar issue before [2]...
>> 
>> 
>> Christian
>> 
>> PS: On my local machine (Mac) everything works fine.
>> 
>> [1] http://pastebin.com/f6f737650
>> [2] http://blog.tfd.co.uk/2008/10/05/osgi-version-resolution/#div-comment-97
>