hive sequenceFile table with hdfs lzo

[yuanxiaolong]

I have a sequenceFile hive table (lzo compress hdfs) , hive cli select rows from table is fine。 when I run the same hive sql In HUE , I got this

java.io.IOException: java.lang.RuntimeException: native-lzo library not available

after google the message , found the error is https://github.com/cloudera/hue/blob/master/apps/beeswax/src/beeswax/locale/zh_CN/LC_MESSAGES/django.po

#: src/beeswax/templates/execute.mako:393
msgid "Fetching results ran into the following error(s):"
msgstr "正在提取遇到以下错误的结果："

How can I set the hue env , like spark ? thanks ! thanks !

export SPARK_LIBRARY_PATH=:/usr/lib/hadoop/lib/native/
export SPARK_SUBMIT_LIBRARY_PATH=:/usr/lib/hadoop/lib/native/
export SPARK_CLASSPATH=:/usr/lib/hadoop/hadoop-lzo-0.4.20-SNAPSHOT.jar

[romainr]

By Hive CLI, do you mean with 'beeline'? Do you run 'beeline' from the same machine as Hue?

[yuanxiaolong]

Thanks for reply, Hive CLI just hive command $HIVE_HOME/bin/hive I run hive command and query hive sql On "Hive Metastore Server" or "Hive gateway" (different machine) all fine. "HUE" and "Hive Metastore Server" (same machine) ，but run hive sql with HUE UI , it doesn't work。

[romainr]

Hue uses the same as 'beeline' which uses HiveServer2 (so misconfiguration there, probably need to add the libs to HiveServer2 path)

[yuanxiaolong]

Romainr, appreciate for your help, I set the env for HiveServer2, It works . thanks again! :+1:

LD_LIBRARY_PATH=/usr/local/hadoop/lzo/lib

[romainr]

:)