AnsweredAssumed Answered

java heap space error in index

Question asked by thor.elpoderoso.esp.technical on Jun 19, 2013
Latest reply on Jun 22, 2013 by thor.elpoderoso.esp.technical
Hello

I'm trying to make a FULL recovery of indexes, but never stops at 60% starts giving me java heap space errors.

Before I start indexing reads:
java.lang.IlegalArgumentException: noderef is a parameter mandaroty

It continues and begins the 10% …. 60% … and JAVA HEAP SPACE.

This is a Redhat with 12GB of RAM and 8cores.
JAVA settings I tested with these two options:

1.- JAVA_OPTS="-server -Xcomp -Xbatch -Xss1m -Xms1g -Xmx2000m -XX:+UseConcMarkSweepGC -XX:+CMSIncrementalMode -XX:NewSize=512m -XX:MaxPermSize=512M -XX:CMSInitiatingOccupancyFraction=80"

2.- JAVA_OPTS="-server -Xms2G -Xmx4G -XX:MaxPermSize=256M -XX:NewSize=1G -Xss1M -verbose:gc -XX:+UseConcMarkSweepGC -XX:+CMSIncrementalMode -XX:CMSInitiatingOccupancyFraction=70"


Also I have this configuration in alfresco-global.propierties

### database connection properties ###
db.driver=org.gjt.mm.mysql.Driver
db.username=admin
db.password=xxxxx
db.name=alfresco
db.url=jdbc:mysql://localhost:3306/alfresco
hibernate.dialect=org.hibernate.dialect.MySQLInnoDBDialect
db.pool.min=20
db.pool.max=275
db.pool.validate.return=true
db.pool.validate.query=select 1
db.pool.evict.idle.min=550000
db.pool.evict.validate=true
db.pool.abandoned.detect=true
db.pool.abandoned.time=550

### lucene configuration ###
lucene.indexer.maxDocsForInMemoryIndex=1000
lucene.indexer.maxDocsForInMemoryMerge=1000

What happens you can do?

Thanks mates




Outcomes