Skip to Content.
Sympa Menu

grouper-users - Re: [grouper-users] ldappc-ng jars, gsh, and memory

Subject: Grouper Users - Open Discussion List

List archive

Re: [grouper-users] ldappc-ng jars, gsh, and memory


Chronological Thread 
  • From: Scott Koranda <>
  • To: Chris Hyzer <>
  • Cc: "" <>
  • Subject: Re: [grouper-users] ldappc-ng jars, gsh, and memory
  • Date: Fri, 1 Oct 2010 16:31:25 -0500

Hi,

> I think it means that commons-logging jar is not compatible with slf4j jar.
> Can you search and make sure there is only one of each of those jars in
> all the lib dirs?

Actually, the two builds resulted in different versions of
"slf4j" jars being built.

Last week's build produced

ldappcng-1.6.1/lib/jcl-over-slf4j-1.5.10.jar
ldappcng-1.6.1/lib/jul-to-slf4j-1.5.10.jar

but today's build produced

ldappcng-1.6.1/lib/jcl-over-slf4j-1.6.1.jar
ldappcng-1.6.1/lib/jul-to-slf4j-1.6.1.jar

I think there is something buggy in today's build.

I am going to revert to downloading the binary from the
Grouper web page and hope for the best.

Thanks,

Scott



>
> Thanks,
> Chris
>
> -----Original Message-----
> From:
>
>
> [mailto:]
> On Behalf Of Scott Koranda
> Sent: Friday, October 01, 2010 5:08 PM
> To:
>
> Subject: Re: [grouper-users] ldappc-ng jars, gsh, and memory
>
> > Hi,
> >
> > I have two Grouper 1.6.1 testbeds, one with 1 GB of RAM and
> > one with 8 GB of RAM.
> >
> > I was able to successfully build, deploy, and run ldappc-ng on
> > the testbed with 8 GB of RAM.
> >
> > On the testbed with 1 GB of RAM, when I try to start gsh.sh
> > (for any reason, not just to run ldappc-ng) I see this:
> >
> > Using GROUPER_HOME: /opt/grouper/grouper.api-1.6.1
> > Using GROUPER_CONF: /opt/grouper/grouper.api-1.6.1/conf
> > Using JAVA: java
> > using MEMORY: 64m-512m
> > Exception in thread "main" java.lang.NoSuchMethodError:
> > org.slf4j.spi.LocationAwareLogger.log(Lorg/slf4j/Marker;Ljava/lang/String;ILjava/lang/String;[Ljava/lang/Object;Ljava/lang/Throwable;)V
> > at
> > org.apache.commons.logging.impl.SLF4JLocationAwareLog.debug(SLF4JLocationAwareLog.java:133)
> > at
> > net.sf.ehcache.config.ConfigurationHelper.createCacheManagerEventListener(ConfigurationHelper.java:300)
> > at net.sf.ehcache.CacheManager.configure(CacheManager.java:283)
> > at net.sf.ehcache.CacheManager.init(CacheManager.java:225)
> > at net.sf.ehcache.CacheManager.<init>(CacheManager.java:186)
> > at
> > edu.internet2.middleware.grouper.cache.EhcacheController.initialize(EhcacheController.java:242)
> > at
> > edu.internet2.middleware.grouper.cache.EhcacheController.getCache(EhcacheController.java:175)
> > at
> > edu.internet2.middleware.grouper.cache.GrouperCache.<init>(GrouperCache.java:81)
> > at
> > edu.internet2.middleware.grouper.util.GrouperUtil.resourcePropertiesCache(GrouperUtil.java:2328)
> > at
> > edu.internet2.middleware.grouper.util.GrouperUtil.propertiesFromResourceName(GrouperUtil.java:6969)
> > at
> > edu.internet2.middleware.grouper.util.GrouperUtil.propertiesFromResourceName(GrouperUtil.java:6774)
> > at
> > edu.internet2.middleware.grouper.util.GrouperUtil.logDirsCreateIfNotDone(GrouperUtil.java:504)
> > at
> > edu.internet2.middleware.grouper.util.GrouperUtil.getLog(GrouperUtil.java:484)
> > at
> > edu.internet2.middleware.grouper.util.GrouperUtil.<clinit>(GrouperUtil.java:6174)
> > at
> > edu.internet2.middleware.grouper.app.gsh.GrouperShell.<clinit>(GrouperShell.java:409)
> > at
> > edu.internet2.middleware.grouper.app.gsh.GrouperShellWrapper.main(GrouperShellWrapper.java:16)
> >
> > If I remove the long list of ldappc-ng jars from
> > GROUPER_HOME/lib/custom then gsh.sh starts fine again.
> >
> > Is there anything I can do to work around this problem? Are
> > all of the jars (124 of them) necessary for ldappc-ng to run?
> >
>
> Sorry, the memory difference between the two machines is a red
> herring.
>
> I swapped the set of ldappc-ng jars (both built with maven
> from source) and the behavior swapped.
>
> The two OS installations are the same.
>
> The only difference is that one set of jars was built last
> week and one set was built today.
>
> Since maven is being used to build from source, is it possible
> that today's build process downloaded some broken code that
> was not downloaded last week?
>
> Thanks,
>
> Scott
>



Archive powered by MHonArc 2.6.16.

Top of Page