Monitor Irish Stats WF

Parameters (with examples):
stats_db_name=openaire_beta_stats_20231208
monitor_irish_db_name=openaire_beta_stats_monitor_ie_20231208b
monitor_irish_db_prod_name=openaire_beta_stats_monitor_ie
graph_db_name=openaire_beta_20231208
monitor_irish_db_shadow_name=openaire_beta_stats_monitor_ie_shadow
hive_timeout=150000
hadoop_user_name=dnet.beta
resumeFrom=Step1-buildIrishMonitorDB
This commit is contained in:
dimitrispie 2023-12-22 11:05:24 +02:00
parent 40b98d8182
commit ffdd03d2f4
7 changed files with 525 additions and 0 deletions

View File

@ -0,0 +1,32 @@
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<parent>
<artifactId>dhp-workflows</artifactId>
<groupId>eu.dnetlib.dhp</groupId>
<version>1.2.4-SNAPSHOT</version>
</parent>
<modelVersion>4.0.0</modelVersion>
<artifactId>dhp-stats-monitor-irish</artifactId>
<dependencies>
<dependency>
<groupId>org.apache.spark</groupId>
<artifactId>spark-core_2.11</artifactId>
</dependency>
<dependency>
<groupId>org.apache.spark</groupId>
<artifactId>spark-sql_2.11</artifactId>
</dependency>
</dependencies>
<build>
<plugins>
<plugin>
<groupId>pl.project13.maven</groupId>
<artifactId>git-commit-id-plugin</artifactId>
<version>2.1.11</version>
<configuration>
<failOnNoGitDirectory>false</failOnNoGitDirectory>
</configuration>
</plugin>
</plugins>
</build>
</project>

View File

@ -0,0 +1,30 @@
<configuration>
<property>
<name>jobTracker</name>
<value>${jobTracker}</value>
</property>
<property>
<name>nameNode</name>
<value>${nameNode}</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>hive_jdbc_url</name>
<value>jdbc:hive2://iis-cdh5-test-m3.ocean.icm.edu.pl:10000/;UseNativeQuery=1;?spark.executor.memory=22166291558;spark.yarn.executor.memoryOverhead=3225;spark.driver.memory=15596411699;spark.yarn.driver.memoryOverhead=1228</value>
</property>
<property>
<name>oozie.wf.workflow.notification.url</name>
<value>{serviceUrl}/v1/oozieNotification/jobUpdate?jobId=$jobId%26status=$status</value>
</property>
</configuration>

View File

@ -0,0 +1,70 @@
export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs
export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami)
if ! [ -L $link_folder ]
then
rm -Rf "$link_folder"
ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder}
fi
export HADOOP_USER_NAME=$2
function copydb() {
db=$1
FILE=("hive_wf_tmp_"$RANDOM)
hdfs dfs -mkdir hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/
# change ownership to impala
# hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db
hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/
# copy the databases from ocean to impala
echo "copying $db"
hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/
hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db
# drop tables from db
for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`;
do
`impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "drop table $i;"`;
done
# drop views from db
for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`;
do
`impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "drop view $i;"`;
done
# delete the database
impala-shell -i impala-cluster-dn1.openaire.eu -q "drop database if exists ${db} cascade";
# create the databases
impala-shell -i impala-cluster-dn1.openaire.eu -q "create database ${db}";
impala-shell -q "INVALIDATE METADATA"
echo "creating schema for ${db}"
for (( k = 0; k < 5; k ++ )); do
for i in `impala-shell -d ${db} --delimited -q "show tables"`;
do
impala-shell -d ${db} --delimited -q "show create table $i";
done | sed 's/"$/;/' | sed 's/^"//' | sed 's/[[:space:]]\date[[:space:]]/`date`/g' | impala-shell --user $HADOOP_USER_NAME -i impala-cluster-dn1.openaire.eu -c -f -
done
# load the data from /tmp in the respective tables
echo "copying data in tables and computing stats"
for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`;
do
impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "load data inpath '/tmp/$FILE/${db}.db/$i' into table $i";
impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "compute stats $i";
done
# deleting the remaining directory from hdfs
hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -rm -R /tmp/$FILE/${db}.db
}
MONITOR_DB=$1
#HADOOP_USER_NAME=$2
copydb $MONITOR_DB

View File

@ -0,0 +1,23 @@
export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs
export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami)
if ! [ -L $link_folder ]
then
rm -Rf "$link_folder"
ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder}
fi
SOURCE=$1
PRODUCTION=$2
echo ${SOURCE}
echo ${PRODUCTION}
#echo "Updating ${PRODUCTION} monitor database old cluster"
#impala-shell -q "create database if not exists ${PRODUCTION}"
#impala-shell -d ${PRODUCTION} -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}./" | sed "s/$/;/" | impala-shell -c -f -
#impala-shell -d ${SOURCE} -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -c -f -
echo "Updating ${PRODUCTION} monitor database"
impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}"
impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION} -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}./" | sed "s/$/;/" | impala-shell -i impala-cluster-dn1.openaire.eu -c -f -
impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE} -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -i impala-cluster-dn1.openaire.eu -c -f -
echo "Production monitor db ready!"

View File

@ -0,0 +1,28 @@
export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs
export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami)
if ! [ -L $link_folder ]
then
rm -Rf "$link_folder"
ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder}
fi
export SOURCE=$1
export TARGET=$2
export SHADOW=$3
export SCRIPT_PATH=$4
export GRAPHDB=$5
export HIVE_OPTS="-hiveconf mapred.job.queue.name=analytics -hiveconf hive.spark.client.connect.timeout=120000ms -hiveconf hive.spark.client.server.connect.timeout=300000ms -hiveconf spark.executor.memory=19166291558 -hiveconf spark.yarn.executor.memoryOverhead=3225 -hiveconf spark.driver.memory=11596411699 -hiveconf spark.yarn.driver.memoryOverhead=1228"
export HADOOP_USER_NAME="oozie"
echo "Getting file from " $4
hdfs dfs -copyToLocal $4
#update Monitor DB IRISH
#cat CreateDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2/g1" | sed "s/GRAPHDB/$3/g1" > foo
cat buildIrishMonitorDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2/g1" | sed "s/GRAPHDB/$5/g1" > foo
hive $HIVE_OPTS -f foo
echo "Hive shell finished"

View File

@ -0,0 +1,224 @@
drop database if exists TARGET cascade;
create database if not exists TARGET;
create view if not exists TARGET.category as select * from SOURCE.category;
create view if not exists TARGET.concept as select * from SOURCE.concept;
create view if not exists TARGET.context as select * from SOURCE.context;
create view if not exists TARGET.country as select * from SOURCE.country;
create view if not exists TARGET.countrygdp as select * from SOURCE.countrygdp;
create view if not exists TARGET.creation_date as select * from SOURCE.creation_date;
--create view if not exists TARGET.funder as select * from SOURCE.funder;
create view if not exists TARGET.fundref as select * from SOURCE.fundref;
create view if not exists TARGET.rndexpenditure as select * from SOURCE.rndexpediture;
create view if not exists TARGET.rndgdpexpenditure as select * from SOURCE.rndgdpexpenditure;
create view if not exists TARGET.doctoratestudents as select * from SOURCE.doctoratestudents;
create view if not exists TARGET.totalresearchers as select * from SOURCE.totalresearchers;
create view if not exists TARGET.totalresearchersft as select * from SOURCE.totalresearchersft;
create view if not exists TARGET.hrrst as select * from SOURCE.hrrst;
drop table if exists TARGET.irish_funders;
create TEMPORARY table TARGET.irish_funders as
select distinct xpath_string(fundingtree[0].value, '//funder/name') as funder from GRAPHDB.project
where xpath_string(fundingtree[0].value, '//funder/jurisdiction')='IE';
--create TEMPORARY table TARGET.irish_funders as
--select distinct name as funder from SOURCE.fundref where country='IE';
drop table if exists TARGET.result;
create table TARGET.result stored as parquet as
select distinct * from SOURCE.result r where r.id in (select ro.id from SOURCE.result_organization ro
left outer join SOURCE.organization o on o.id=ro.organization
left outer join SOURCE.result_projects rp on rp.id=ro.id
left outer join SOURCE.project p on p.id=rp.project
join TARGET.irish_funders irf
where o.country='IE' or p.funder=irf.funder);
create view if not exists TARGET.category as select * from SOURCE.category;
create view if not exists TARGET.concept as select * from SOURCE.concept;
create view if not exists TARGET.context as select * from SOURCE.context;
create view if not exists TARGET.country as select * from SOURCE.country;
create view if not exists TARGET.countrygdp as select * from SOURCE.countrygdp;
create view if not exists TARGET.creation_date as select * from SOURCE.creation_date;
create table TARGET.funder stored as parquet as select * from SOURCE.funder where country='IE';
create view if not exists TARGET.fundref as select * from SOURCE.fundref;
create view if not exists TARGET.rndexpenditure as select * from SOURCE.rndexpediture;
create view if not exists TARGET.rndgdpexpenditure as select * from SOURCE.rndgdpexpenditure;
create view if not exists TARGET.doctoratestudents as select * from SOURCE.doctoratestudents;
create view if not exists TARGET.totalresearchers as select * from SOURCE.totalresearchers;
create view if not exists TARGET.totalresearchersft as select * from SOURCE.totalresearchersft;
create view if not exists TARGET.hrrst as select * from SOURCE.hrrst;
--create view if not exists TARGET.graduatedoctorates as select * from SOURCE.graduatedoctorates;
create table TARGET.result_citations stored as parquet as select * from SOURCE.result_citations orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_references_oc stored as parquet as select * from SOURCE.result_references_oc orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_citations_oc stored as parquet as select * from SOURCE.result_citations_oc orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_classifications stored as parquet as select * from SOURCE.result_classifications orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_apc stored as parquet as select * from SOURCE.result_apc orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_concepts stored as parquet as select * from SOURCE.result_concepts orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_datasources stored as parquet as select * from SOURCE.result_datasources orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_fundercount stored as parquet as select * from SOURCE.result_fundercount orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_gold stored as parquet as select * from SOURCE.result_gold orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_greenoa stored as parquet as select * from SOURCE.result_greenoa orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_languages stored as parquet as select * from SOURCE.result_languages orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_licenses stored as parquet as select * from SOURCE.result_licenses orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.licenses_normalized STORED AS PARQUET as select * from SOURCE.licenses_normalized;
create table TARGET.result_oids stored as parquet as select * from SOURCE.result_oids orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_organization stored as parquet as select * from SOURCE.result_organization orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_peerreviewed stored as parquet as select * from SOURCE.result_peerreviewed orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_pids stored as parquet as select * from SOURCE.result_pids orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_projectcount stored as parquet as select * from SOURCE.result_projectcount orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_projects stored as parquet as select * from SOURCE.result_projects orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_refereed stored as parquet as select * from SOURCE.result_refereed orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_sources stored as parquet as select * from SOURCE.result_sources orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_topics stored as parquet as select * from SOURCE.result_topics orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_fos stored as parquet as select * from SOURCE.result_fos orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_accessroute stored as parquet as select * from SOURCE.result_accessroute orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_instance stored as parquet as select * from SOURCE.result_instance orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_orcid stored as parquet as select * from SOURCE.result_orcid orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create view TARGET.foo1 as select * from SOURCE.result_result rr where rr.source in (select id from TARGET.result);
create view TARGET.foo2 as select * from SOURCE.result_result rr where rr.target in (select id from TARGET.result);
create table TARGET.result_result STORED AS PARQUET as select distinct * from (select * from TARGET.foo1 union all select * from TARGET.foo2) foufou;
drop view TARGET.foo1;
drop view TARGET.foo2;
-- datasources
create view if not exists TARGET.datasource as select * from SOURCE.datasource;
create view if not exists TARGET.datasource_oids as select * from SOURCE.datasource_oids;
create view if not exists TARGET.datasource_organizations as select * from SOURCE.datasource_organizations;
create view if not exists TARGET.datasource_sources as select * from SOURCE.datasource_sources;
create table TARGET.datasource_results stored as parquet as select id as result, datasource as id from TARGET.result_datasources;
-- organizations
create view if not exists TARGET.organization as select * from SOURCE.organization;
create view if not exists TARGET.organization_datasources as select * from SOURCE.organization_datasources;
create view if not exists TARGET.organization_pids as select * from SOURCE.organization_pids;
create view if not exists TARGET.organization_projects as select * from SOURCE.organization_projects;
create view if not exists TARGET.organization_sources as select * from SOURCE.organization_sources;
-- projects
create view if not exists TARGET.project as select * from SOURCE.project;
create view if not exists TARGET.project_oids as select * from SOURCE.project_oids;
create view if not exists TARGET.project_organizations as select * from SOURCE.project_organizations;
create view if not exists TARGET.project_resultcount as select * from SOURCE.project_resultcount;
create view if not exists TARGET.project_classification as select * from SOURCE.project_classification;
create view if not exists TARGET.project_organization_contribution as select * from SOURCE.project_organization_contribution;
create table TARGET.project_results stored as parquet as select id as result, project as id from TARGET.result_projects;
-- indicators
-- Sprint 1 ----
create table TARGET.indi_pub_green_oa stored as parquet as select * from SOURCE.indi_pub_green_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_grey_lit stored as parquet as select * from SOURCE.indi_pub_grey_lit orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_doi_from_crossref stored as parquet as select * from SOURCE.indi_pub_doi_from_crossref orig where exists (select 1 from TARGET.result r where r.id=orig.id);
-- Sprint 2 ----
create table TARGET.indi_result_has_cc_licence stored as parquet as select * from SOURCE.indi_result_has_cc_licence orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_result_has_cc_licence_url stored as parquet as select * from SOURCE.indi_result_has_cc_licence_url orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_has_abstract stored as parquet as select * from SOURCE.indi_pub_has_abstract orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_result_with_orcid stored as parquet as select * from SOURCE.indi_result_with_orcid orig where exists (select 1 from TARGET.result r where r.id=orig.id);
---- Sprint 3 ----
create table TARGET.indi_funded_result_with_fundref stored as parquet as select * from SOURCE.indi_funded_result_with_fundref orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create view TARGET.indi_result_org_collab as select * from SOURCE.indi_result_org_collab;
create view TARGET.indi_result_org_country_collab as select * from SOURCE.indi_result_org_country_collab;
create view TARGET.indi_project_collab_org as select * from SOURCE.indi_project_collab_org;
create view TARGET.indi_project_collab_org_country as select * from SOURCE.indi_project_collab_org_country;
create view TARGET.indi_funder_country_collab as select * from SOURCE.indi_funder_country_collab;
create view TARGET.indi_result_country_collab as select * from SOURCE.indi_result_country_collab;
---- Sprint 4 ----
create table TARGET.indi_pub_diamond stored as parquet as select * from SOURCE.indi_pub_diamond orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_in_transformative stored as parquet as select * from SOURCE.indi_pub_in_transformative orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_closed_other_open stored as parquet as select * from SOURCE.indi_pub_closed_other_open orig where exists (select 1 from TARGET.result r where r.id=orig.id);
---- Sprint 5 ----
create table TARGET.indi_result_no_of_copies stored as parquet as select * from SOURCE.indi_result_no_of_copies orig where exists (select 1 from TARGET.result r where r.id=orig.id);
---- Sprint 6 ----
create table TARGET.indi_pub_hybrid_oa_with_cc stored as parquet as select * from SOURCE.indi_pub_hybrid_oa_with_cc orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_bronze_oa stored as parquet as select * from SOURCE.indi_pub_bronze_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_downloads stored as parquet as select * from SOURCE.indi_pub_downloads orig where exists (select 1 from TARGET.result r where r.id=orig.result_id);
create table TARGET.indi_pub_downloads_datasource stored as parquet as select * from SOURCE.indi_pub_downloads_datasource orig where exists (select 1 from TARGET.result r where r.id=orig.result_id);
create table TARGET.indi_pub_downloads_year stored as parquet as select * from SOURCE.indi_pub_downloads_year orig where exists (select 1 from TARGET.result r where r.id=orig.result_id);
create table TARGET.indi_pub_downloads_datasource_year stored as parquet as select * from SOURCE.indi_pub_downloads_datasource_year orig where exists (select 1 from TARGET.result r where r.id=orig.result_id);
---- Sprint 7 ----
create table TARGET.indi_pub_gold_oa stored as parquet as select * from SOURCE.indi_pub_gold_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_hybrid stored as parquet as select * from SOURCE.indi_pub_hybrid orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create view TARGET.indi_org_fairness as select * from SOURCE.indi_org_fairness;
create view TARGET.indi_org_fairness_pub_pr as select * from SOURCE.indi_org_fairness_pub_pr;
create view TARGET.indi_org_fairness_pub_year as select * from SOURCE.indi_org_fairness_pub_year;
create view TARGET.indi_org_fairness_pub as select * from SOURCE.indi_org_fairness_pub;
create view TARGET.indi_org_fairness_year as select * from SOURCE.indi_org_fairness_year;
create view TARGET.indi_org_findable_year as select * from SOURCE.indi_org_findable_year;
create view TARGET.indi_org_findable as select * from SOURCE.indi_org_findable;
create view TARGET.indi_org_openess as select * from SOURCE.indi_org_openess;
create view TARGET.indi_org_openess_year as select * from SOURCE.indi_org_openess_year;
create table TARGET.indi_pub_has_preprint stored as parquet as select * from SOURCE.indi_pub_has_preprint orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_in_subscribed stored as parquet as select * from SOURCE.indi_pub_in_subscribed orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_result_with_pid stored as parquet as select * from SOURCE.indi_result_with_pid orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_impact_measures stored as parquet as select * from SOURCE.indi_impact_measures orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_interdisciplinarity stored as parquet as select * from SOURCE.indi_pub_interdisciplinarity orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_apc_affiliations stored as parquet as select * from SOURCE.result_apc_affiliations orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_is_project_result_after stored as parquet as select * from SOURCE.indi_is_project_result_after orig where exists (select 1 from TARGET.result r where r.id=orig.result_id);
create view TARGET.indi_is_funder_plan_s as select * from SOURCE.indi_is_funder_plan_s;
create view TARGET.indi_funder_fairness as select * from SOURCE.indi_funder_fairness;
create view TARGET.indi_funder_openess as select * from SOURCE.indi_funder_openess;
create view TARGET.indi_funder_findable as select * from SOURCE.indi_funder_findable;
create view TARGET.indi_ris_fairness as select * from SOURCE.indi_ris_fairness;
create view TARGET.indi_ris_openess as select * from SOURCE.indi_ris_openess;
create view TARGET.indi_ris_findable as select * from SOURCE.indi_ris_findable;
create table TARGET.indi_pub_green_with_license stored as parquet as select * from SOURCE.indi_pub_green_with_license orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.result_country stored as parquet as select * from SOURCE.result_country orig where exists (select 1 from TARGET.result r where r.id=orig.id);
create table TARGET.indi_pub_publicly_funded stored as parquet as select * from SOURCE.indi_pub_publicly_funded orig where exists (select 1 from TARGET.result r where r.id=orig.id);

View File

@ -0,0 +1,118 @@
<workflow-app name="Irish Monitor Update" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>stats_db_name</name>
<description>the target stats database name</description>
</property>
<property>
<name>graph_db_name</name>
<description>the graph database name</description>
</property>
<property>
<name>monitor_irish_db_name</name>
<description>the target monitor db name</description>
</property>
<property>
<name>monitor_irish_db_prod_name</name>
<description>the name of the production monitor db</description>
</property>
<property>
<name>monitor_irish_db_shadow_name</name>
<description>the name of the shadow monitor db</description>
</property>
<property>
<name>hive_metastore_uris</name>
<description>hive server metastore URIs</description>
</property>
<property>
<name>hive_jdbc_url</name>
<description>hive server jdbc url</description>
</property>
<property>
<name>hive_timeout</name>
<description>the time period, in seconds, after which Hive fails a transaction if a Hive client has not sent a hearbeat. The default value is 300 seconds.</description>
</property>
<property>
<name>hadoop_user_name</name>
<description>user name of the wf owner</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
<property>
<name>hive.txn.timeout</name>
<value>${hive_timeout}</value>
</property>
<property>
<name>mapred.job.queue.name</name>
<value>analytics</value>
</property>
</configuration>
</global>
<start to="resume_from"/>
<decision name="resume_from">
<switch>
<case to="Step1-buildIrishMonitorDB">${wf:conf('resumeFrom') eq 'Step1-buildIrishMonitorDB'}</case>
<case to="Step2-copyDataToImpalaCluster">${wf:conf('resumeFrom') eq 'Step2-copyDataToImpalaCluster'}</case>
<case to="Step3-finalizeImpalaCluster">${wf:conf('resumeFrom') eq 'Step3-finalizeImpalaCluster'}</case>
<default to="Step1-buildIrishMonitorDB"/>
</switch>
</decision>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="Step1-buildIrishMonitorDB">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>monitor_irish.sh</exec>
<argument>${stats_db_name}</argument>
<argument>${monitor_irish_db_name}</argument>
<argument>${monitor_irish_db_shadow_name}</argument>
<argument>${wf:appPath()}/scripts/buildIrishMonitorDB.sql</argument>
<argument>${graph_db_name}</argument>
<file>monitor_irish.sh</file>
</shell>
<ok to="Step2-copyDataToImpalaCluster"/>
<error to="Kill"/>
</action>
<action name="Step2-copyDataToImpalaCluster">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>copyDataToImpalaCluster.sh</exec>
<argument>${monitor_irish_db_name}</argument>
<argument>${hadoop_user_name}</argument>
<file>copyDataToImpalaCluster.sh</file>
</shell>
<ok to="Step3-finalizeImpalaCluster"/>
<error to="Kill"/>
</action>
<action name="Step3-finalizeImpalaCluster">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>finalizeImpalaCluster.sh</exec>
<argument>${monitor_irish_db_name}</argument>
<argument>${monitor_irish_db_prod_name}</argument>
<argument>${monitor_irish_db_shadow_name}</argument>
<file>finalizeImpalaCluster.sh</file>
</shell>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>