hi Raihan,
I was in the same situation before.See if this thing helps you.
Hive command writes to a log file and you cat that file for the success pattern.
Let me know if this will help you.Need any further help??
for log in $logdir1/*.hivelog;
do
cat $log | grep "success pattern" #&> /dev/null
if [ $? -eq 0 ]
then
echo $log job suceeded
else
echo $log job failed
fi
done
{ while read myline;do
filePath=$LOCAL_ROOT_DIR$PREFIXS"/"$myline
HIVELOGNAME=`echo $myline | cut -d . -f 1`
( hive -f $fileName.tmp > $LOGDIR/$HIVELOGNAME.hivelog 2>&1 ) &
done } < $LOCAL_ROOT_DIR$PREFIXS"/"$FILE_LIST
}
Regards
Abhishek
On Wed, Aug 15, 2012 at 8:03 PM, Raihan Jamal <[email protected]> wrote:
> Does the hive command make use of an exit status for success vs. failure?
>
> I am asking this question as I am executing my few HiveQL queries from my
> Shell Script and If one HiveQL queries gets failed due to certain reasons, I
> want to abort my shell script at that moment only without executing other
> HiveQL queries.
>
>
>
>
> Raihan Jamal
>
