diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/copyDataToImpalaCluster.sh b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/copyDataToImpalaCluster.sh index 8b5ff7f02..989eeae84 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/copyDataToImpalaCluster.sh +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/copyDataToImpalaCluster.sh @@ -10,9 +10,8 @@ export PROD_USAGE_STATS_DB="openaire_prod_usage_stats" # Set the active HDFS node of OCEAN and IMPALA cluster. -OCEAN_DOMAIN='iis-cdh5-test-m1.ocean.icm.edu.pl' -OCEAN_HDFS_NODE='hdfs://'${OCEAN_DOMAIN}':8020' -echo "Active OCEAN HDFS Node: ${OCEAN_HDFS_NODE}" +OCEAN_HDFS_NODE='hdfs://nameservice1' +echo "OCEAN HDFS virtual-name which resolves automatically to the active-node: ${OCEAN_HDFS_NODE}" IMPALA_HDFS_NODE='' if hdfs dfs -test -e hdfs://impala-cluster-mn1.openaire.eu >/dev/null 2>&1; then @@ -71,7 +70,7 @@ function copydb() { # Since there are many DBs handled in this script and many more may be added, we cannot easily keep track of the views of all those tables, so leave this extra loop for now. for (( k = 1; k <= 2; k++ )); do - echo -e "Create tables iteration_${k}" + echo -e "\nCreate tables iteration_${k}\n" for i in `hive -e "use $db; show tables;" | sed 's/WARN:.*//g'`; do CURRENT_PRQ_FILE=`hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -ls -C "/tmp/${TEMP_SUBDIR}/${db}.db/${i}/*.parq" | head -1` if [[ -z "$CURRENT_PRQ_FILE" ]]; then