diff options
| author | Aaron Davidson <aaron@databricks.com> | 2014-04-08 14:40:20 -0700 |
|---|---|---|
| committer | Patrick Wendell <pwendell@gmail.com> | 2014-04-08 14:40:20 -0700 |
| commit | e25b593447a2e0aab9e5066f755e41be9068ecdc (patch) | |
| tree | b2d5b304abd0e926f0e069cfd56e6ad10b8924d8 /bin | |
| parent | a8d86b080ae26c96b078ba14dc60f3b528c07787 (diff) | |
| download | spark-e25b593447a2e0aab9e5066f755e41be9068ecdc.tar.gz spark-e25b593447a2e0aab9e5066f755e41be9068ecdc.tar.bz2 spark-e25b593447a2e0aab9e5066f755e41be9068ecdc.zip | |
SPARK-1445: compute-classpath should not print error if lib_managed not found
This was added to the check for the assembly jar, forgot it for the datanucleus jars.
Author: Aaron Davidson <aaron@databricks.com>
Closes #361 from aarondav/cc and squashes the following commits:
8facc16 [Aaron Davidson] SPARK-1445: compute-classpath should not print error if lib_managed not found
Diffstat (limited to 'bin')
| -rwxr-xr-x | bin/compute-classpath.sh | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/bin/compute-classpath.sh b/bin/compute-classpath.sh index be37102dc0..2a2bb376fd 100755 --- a/bin/compute-classpath.sh +++ b/bin/compute-classpath.sh @@ -63,7 +63,7 @@ fi # built with Hive, so first check if the datanucleus jars exist, and then ensure the current Spark # assembly is built for Hive, before actually populating the CLASSPATH with the jars. # Note that this check order is faster (by up to half a second) in the case where Hive is not used. -num_datanucleus_jars=$(ls "$FWDIR"/lib_managed/jars/ | grep "datanucleus-.*\\.jar" | wc -l) +num_datanucleus_jars=$(ls "$FWDIR"/lib_managed/jars/ 2>/dev/null | grep "datanucleus-.*\\.jar" | wc -l) if [ $num_datanucleus_jars -gt 0 ]; then AN_ASSEMBLY_JAR=${ASSEMBLY_JAR:-$DEPS_ASSEMBLY_JAR} num_hive_files=$(jar tvf "$AN_ASSEMBLY_JAR" org/apache/hadoop/hive/ql/exec 2>/dev/null | wc -l) |