Getting Started Tutorial

Goal

This tutorial covers how to query a file and a directory on your local file system. Files and directories are like standard SQL tables to Drill. If you install Drill in embedded mode, the installer registers and configures your file system as the dfs instance. You can query these types of files using the default dfs storage plugin:

• Plain text files, such as comma-separated values (CSV) or tab-separated values (TSV) files
• JSON files
• Parquet files

In this tutorial, you query plain text files using the dfs storage plugin. You also create a custom storage plugin to simplify querying plain text files.

Prerequisites

This tutorial assumes that you installed Drill in embedded mode. The first few lessons of the tutorial use a Google file of Ngram data that you download from the internet. The compressed Google Ngram files are 8 and 58MB. To expand the compressed files, you need an additional 448MB of free disk space for this exercise.

To get started, use the SQLLine command to start the Drill command line interface (CLI) on Linux, Mac OS X, or Windows.

Start Drill (Linux or Mac OS X)

To start Drill on Linux or Mac OS X, use the SQLLine command.

1. Open a terminal.

2. Navigate to the Drill installation directory.

   Example: $ cd ~/apache-drill-<version>

3. Issue the following command:

   $ bin/sqlline -u jdbc:drill:zk=local
   

   The Drill prompt appears: 0: jdbc:drill:zk=local

Start Drill (Windows)

To start Drill on Windows, use the SQLLine command.

1. Open the apache-drill-<version> folder.
2. Open the bin folder, and double-click on the sqlline.bat file. The Windows command prompt opens.

3. At the sqlline> prompt, issue the following command, and then press Enter:

   !connect jdbc:drill:zk=local  
   

   The following prompt appears: 0: jdbc:drill:zk=local

Stop Drill

To stop Drill, issue the following command at the Drill prompt.

    0: jdbc:drill:zk=local> !quit

In some cases, such as stopping while a query is in progress, this command does not stop Drill. You need to kill the Drill process. For example, on Mac OS X and Linux, follow these steps:

1. Issue a CTRL Z to stop the query, then start Drill again. If the startup message indicates success, skip the rest of the steps. If not, proceed to step 2.

2. Search for the Drill process IDs.

   $ ps auwx | grep drill
   

3. Kill each process using the process numbers in the grep output. For example:

   $ sudo kill -9 2674