set HBase properties for Spark Job using spark-submit

Question

During Hbase data migration I have encountered ajava.lang.IllegalArgumentException: KeyValue size too large

In long term :

I need to increase the properties hbase.client.keyvalue.maxsize (from 1048576 to 10485760) in the /etc/hbase/conf/hbase-site.xml but I can't change this file now (I need validation).

In short term :

I have success to import data using command :

hbase org.apache.hadoop.hbase.mapreduce.Import \
  -Dhbase.client.keyvalue.maxsize=10485760 \
  myTable \
  myBackupFile

Now I need to run a Spark Job using spark-submit

What is the better way :

Prefix the HBase properties with 'spark.' (I'm not sure it's possible and if it's works)

spark-submit \
  --conf spark.hbase.client.keyvalue.maxsize=10485760

Using 'spark.executor.extraJavaOptions' and 'spark.driver.extraJavaOptions' to explicitly transmit HBase properties

spark-submit \
  --conf spark.executor.extraJavaOptions=-Dhbase.client.keyvalue.maxsize=10485760 \
  --conf spark.driver.extraJavaOptions=-Dhbase.client.keyvalue.maxsize=10485760

I don't think the spark prefix will work, you' need to use extraJavaOptions. — Ben Watson

VS_FF VS_FF · Accepted Answer · 2020-02-04T11:35:22

If you can change your code, you should be able to set these properties programmatically. I think something like this used to work for me in the past in Java:

Configuration conf = HBaseConfiguration.create();
conf.set("hbase.client.scanner.timeout.period", SCAN_TIMEOUT); // set BEFORE you create the connection object below:
Connection conn = ConnectionFactory.createConnection(conf);

set HBase properties for Spark Job using spark-submit

1 Answers