D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

Changes to Promotion Stats WF 

1. Add new cluster host at impala-shell commands
2. Add a step for splitting monitor dbs
3. Update workflow.xml to included the new splitting monitor dbs step
This commit is contained in:
dimitrispie 2023-06-19 09:44:34 +03:00
parent 4c770a5e29
commit 36e0a8fec4
3 changed files with 43 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
dhp-workflows/dhp-stats-promote/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/updateProductionViews.sh
View File

@ -10,7 +10,7 @@ export SOURCE=$1
export PRODUCTION=$2
echo "Updating ${PRODUCTION} database"
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 -
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 -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 -c -f -
echo "Production db ready!"

38
dhp-workflows/dhp-stats-promote/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/updateProductionViewsMonitor.sh Normal file
View File

@ -0,0 +1,38 @@
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 PRODUCTION=$2
echo "Updating ${PRODUCTION}'_funded' database"
impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}'_funded'"
impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION}'_funded' -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}'_funded'./" | sed "s/$/;/" | impala-shell -c -f -
impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE}'_funded' -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}'_funded'.\1 as select * from ${SOURCE}'_funded'.\1;/" | impala-shell -c -f -
echo "Production funded db ready!"
echo "Updating ${PRODUCTION}'_institutions' database"
impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}'_institutions'"
impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION}'_institutions' -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}'_institutions'./" | sed "s/$/;/" | impala-shell -c -f -
impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE}'_institutions' -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}'_institutions'.\1 as select * from ${SOURCE}'_institutions'.\1;/" | impala-shell -c -f -
echo "Production insitutions db ready!"
echo "Updating ${PRODUCTION}'_ris_tail' database"
impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}'_ris_tail'"
impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION}'_ris_tail' -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}'_ris_tail'./" | sed "s/$/;/" | impala-shell -c -f -
impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE}'_RIs_tail' -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}'_ris_tail'.\1 as select * from ${SOURCE}'_ris_tail'.\1;/" | impala-shell -c -f -
echo "Production RIS tail db ready!"
contexts="knowmad::other dh-ch::other enermaps::other gotriple::other neanias-atmospheric::other rural-digital-europe::other covid-19::other aurora::other neanias-space::other north-america-studies::other north-american-studies::other eutopia::other"
for i in ${contexts}
do
tmp=`echo "$i" | sed 's/'-'/'_'/g' | sed 's/'::'/'_'/g'`
impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}'_'${tmp}"
impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION}'_'${tmp} -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}'_'${tmp}./" | sed "s/$/;/" | impala-shell -c -f -
impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE}'_'${tmp} -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}'_'${tmp}.\1 as select * from ${SOURCE}'_'${tmp}.\1;/" | impala-shell -c -f -
echo "Production ${tmp} db ready!"
done

4
dhp-workflows/dhp-stats-promote/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml
View File

@ -80,10 +80,10 @@
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>updateProductionViews.sh</exec>
<exec>updateProductionViewsMonitor.sh</exec>
<argument>${monitor_db_name}</argument>
<argument>${monitor_db_production_name}</argument>
<file>updateProductionViews.sh</file>
<file>updateProductionViewsMonitor.sh</file>
</shell>
<ok to="updateObservatoryViews"/>
<error to="Kill"/>