[Microsoft][ODBC Driver Manager] Invalid string or buffer length exception for 64 bit SQL Server

this is a java bug, upgrade to java 1.7.70 at least

java and access

to connect access form java, i get this error. Error: java.sql.SQLException: [Microsoft][ODBC Driver Manager] Data source name not found and no default driver specified To connect java 64 bit with access, it 's necesary install driver 64 bit download and install http://download.microsoft.com/download/0/B/8/0B8EFDD9-8CE5-405C-A28D-A108A63AE2FC/AccessDatabaseEngine_X64.exe And if you want read all query and launch this query "SELECT [Name] FROM MSysObjects WHERE Name Like 'MSys*' ORDER BY [Name]" you get this error: error: READ en 'MSysObjects' You must execute a macro http://stackoverflow.com/questions/19971082/no-read-permission-on-msysobjects strDdl = "GRANT SELECT ON MSysObjects TO Admin;" CurrentProject.Connection.Execute strDdl If you get this error "ODBC Driver Manager 'invalid string or buffer length' error " , my solver is update java 1.7.70 at least.

install jdk7 in ubuntu

To install jdk7 in ubuntu follow next step

sudo apt-get install python-software-properties

if dont install this error is show "Ubuntu: add-apt-repository: command not found" when execute next step

 sudo add-apt-repository ppa:webupd8team/java

 sudo apt-get update

 sudo apt-get install oracle-java7-installer

make a mirror dbpedia in 5 steps

to make a mirror, you follow 5 simples steps. you must install virtuoso database. and change virtuoso.ini to read files

DirsAllowed              = ., /path/files/script 

1º download files from dbpedia with command

wget -r -np -nd -nc -A'*.nt.bz2' http://downloads.dbpedia.org/3.8/en/

2º decompress files

for i in `ls ./download/*.bz2` ; do
  bzcat $i | gzip --fast > ${i%.bz2}.gz && mv ./download/*.gz ./decompress/ && rm $i ;
done &

3º deletes triplets more long 1024 chars.

cd decompress;
for i in *.gz  ; do
  echo -n "cleaning $i..."
  zcat $i | grep -v -E '^<.+> <.+> <.{1025,}> \.$'  > $i.clean
  mv *clean ../fixed/
  echo "done."
done

4º split files in small files.

cd fixed
for i in *clean ; do

  split_size=$(stat -c%s "$i")

  echo "Size = $split_size"

  echo -n "slipting  $i..."

  if [ $split_size -gt 5000000 ]; then
        echo "File is large."

        echo "Performing split on file $i"

        split -a 10 -l 50000 $i "split/$i" ;
  else
        mv $i split/ ;
   fi

done;

5º execute insert command in virtuoso database

echo "creating load statement"

cd  slipt

for file in `ls *`
do
        load_target="/home/user/split/$file"
        load_query="EXEC=TTLP_MT(file_to_string_output('$load_target'), '', 'http://dbpedia.org', 255);commit WORK;"
        isql-vt 1111  dba dba  "$load_query"
     #  echo "$virt_isql" "$virt_port" "$virt_userName" "$virt_passWord" "$load_query"
 if [ $?  -ne 1 ]; then
  #echo "execution ok. move data"
  ls  split/ | wc -l
  mv "$load_target"  process-files/
  echo "procesando $file"
 else
   echo "error to process file $file"
   exit;
 fi
done;
echo "done"

Eclipse + maven: error compilation suddenly

Sometimes, you can see this error in eclipse.

[INFO] -------------------------------------------------------------
[ERROR] COMPILATION ERROR :
[INFO] -------------------------------------------------------------
[ERROR] Unable to locate the Javac Compiler in:
  C:\Program Files (x86)\Java\jre7\..\lib\tools.jar
Please ensure you are using JDK 1.4 or above and
not a JRE (the com.sun.tools.javac.Main class is required).
In most cases you can change the location of your Java
installation by setting the JAVA_HOME environment variable.
[INFO] 1 error

This error happen when you compile eclipse and maven with jre. You must change to JDK compile in Eclise.

read a resource inisde a jar

In both examples the file has locate in root of jar file

this.getClass().getClassLoader().getResourceAsStream("file.xml"))
this.getClass().getResourceAsStream("/file.xml"))



this.getClass().getClassLoader().getResource("file.xml").getFile())

creacion de un cluster con Lily en cloudera

Encontre hace poco un proyecto muy interesante http://www.lilyproject.org/lily/index.html
Para instalarlo en clustes hay que seguir la guía pero te encontrarás con algún problema.
http://lilyproject.org/books/lily_book--2_0--20130124-120744/publications/html-chunked/output/dsy432-lily.html#dsy432-lily Recientemente he creado un cluster con 3 equipos en rack de Lily. Primero instalo cloudera en cada uno de los 3 equipos pero me pedia actualizar a la version de 12 de ubuntu.
Una vez actualizado me han ido dando distintos errores:

sncronizacion de los relojes.
La sincronización de los relojes de los equipos, tiene que dar una diferencia menor de 3 segundos. Con un servicio de NTP en ubuntu, esto se ha corregido. El resto de los equipos se conecta a el. Si no se sincroniza correctamente verifica la fecha hora de tu equipo. Verifica que si tienes un dominio, estén todos en dominio, ya que si uno de ellos no lo está, fallará constantemente y no usará el servicio de NTP.

hdfs al arrancar da un error.
se soluciona borrando la carpeta a mano.

mapreduce-traker da un error de versiones distinta, se solucion actualizan todo el sistema.

HBase no arranca bien,
HRegionServer Failed deleting my ephemeral node org.apache.zookeeper.KeeperException$SessionExpiredException: KeeperErrorCode = Session expired esto se debe a un error de sincronización.

Para poder lanzar Lily en Cloudera se debe tener autenticacion con Kerberos si te da un error parecido a este "Authentication failed. Cannot get kdc for realm" es que no tienes kerberos instalados. http://www.cloudera.com/content/cloudera-content/cloudera-docs/CDH4/latest/CDH4-Security-Guide/cdh4sg_topic_11_1.html tutorial de kerberos https://we.riseup.net/debian/kerberos-for-birds-not-dummies primero se debe instalar kerberos install kerbero atp-get install krb5-admin-server