Getting Read time out when connecting to hive/spark sql through squirrel sql

3
votes

I'm trying to use Squirrel SQL to connect to spark-sql thriftserver using jdbc. For smaller resultset I'm getting the response. But for queries which are running for more than 30s, Squirrel throws the below exception:

Error: org.apache.thrift.transport.TTransportException:
java.net.SocketTimeoutException: Read timed out
SQLState: 08S01
ErrorCode: 0

But when I check spark job page on port 4040 it shows that the query execution is successful. Even when I run the same query thru beeline I'm able to see the results. Looks like there is a client side timeout for squirrel sql. Kindly let me know how to increase the timeout parameter.

1
jdbcapache-spark-sqlsquirrel-sql
Did you ever find the answer for this? Having the same problem - Matt
Not found the solution yet. - Gayatri Mahesh

1 Answers

0
votes

Go to File -> Global Preferences -> SQL Tab -> Change "Login Timeout" to 0 -> Press OK

After that reconnect the session by Session -> Reconnect

I think this works because of this bug. https://issues.apache.org/jira/browse/HIVE-14517