Sqoop for database imports in Hadoop - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the output of this Sqoop import command?

Consider the following Sqoop command to import data from a MySQL database into HDFS:

sqoop import \
--connect jdbc:mysql://localhost/employees \
--username user \
--password pass \
--table employees \
--target-dir /user/hadoop/employees_data \
--num-mappers 1

What will be the result of running this command?

A Data from the 'employees' table is imported into HDFS directory '/user/hadoop/employees_data' as SequenceFiles using one mapper.

B Data from the 'employees' table is imported into HDFS directory '/user/hadoop/employees_data' as Avro files using multiple mappers.

C Data from the 'employees' table is imported into HDFS directory '/user/hadoop/employees_data' as text files using one mapper.

D The command will fail because the '--num-mappers' option cannot be set to 1.

💡 Hint

Think about the default file format and the effect of '--num-mappers 1'.

The command imports the 'employees' table data into the specified HDFS directory as text files. The '--num-mappers 1' option means only one mapper is used, so the import is done in a single task. The default file format is text unless specified otherwise.

🧠 Conceptual

intermediate

Which option controls the number of parallel tasks in Sqoop import?

In Sqoop, you want to speed up data import by running multiple parallel tasks. Which command-line option controls how many parallel mappers Sqoop uses during import?

A --num-mappers

B --parallel-tasks

C --mapper-count

D --task-parallelism

💡 Hint

Look for the option that specifies the number of mappers.

The '--num-mappers' option sets how many parallel map tasks Sqoop will use to import data. The other options do not exist in Sqoop.

🔧 Debug

advanced

Why does this Sqoop import command fail with a connection error?

Given this command:

sqoop import \
--connect jdbc:mysql://localhost:3306/employees \
--username user \
--password pass \
--table employees \
--target-dir /user/hadoop/employees_data

The command fails with a connection refused error. What is the most likely cause?

A The username or password is incorrect.

B The MySQL server is not running or not reachable at localhost:3306.

C The '--table' option is misspelled.

D The '--target-dir' path is invalid in HDFS.

💡 Hint

Connection refused usually means the server is unreachable.

A connection refused error means Sqoop cannot reach the MySQL server at the given host and port. This usually means the server is down or blocked by firewall. Other options would cause different errors.

❓ data_output

advanced

What is the number of files created after this Sqoop import?

Run this Sqoop import command:

sqoop import \
--connect jdbc:mysql://localhost/employees \
--username user \
--password pass \
--table employees \
--target-dir /user/hadoop/employees_data \
--num-mappers 4

Assuming the import succeeds, how many part files will be created in the target directory?

A 5 part files, one extra for metadata

B 1 part file, because files are merged

C 0 files, because data is stored in a database

D 4 part files, one per mapper

💡 Hint

Each mapper writes one output file.

Sqoop uses one mapper per parallel task, and each mapper writes one part file in the target directory. So with 4 mappers, 4 part files are created.

🚀 Application

expert

How to import only rows matching a condition using Sqoop?

You want to import only employees with salary greater than 50000 from the 'employees' table using Sqoop. Which option should you use to filter rows during import?

A --where "salary > 50000"

B --condition salary > 50000

C --filter salary > 50000

D --query "SELECT * FROM employees WHERE salary > 50000"

💡 Hint

Look for the option that allows SQL WHERE clause filtering.

The '--where' option lets you specify a SQL WHERE clause to filter rows during import. The '--query' option requires a full query with a placeholder for splitting, so it is more complex to use.