SQL 客户端

Flink 的 Table & SQL API 可以处理 SQL 语言编写的查询语句,但是这些查询需要嵌入用 Java 或 Scala 编写的表程序中。此外,这些程序在提交到集群前需要用构建工具打包。这或多或少限制了 Java/Scala 程序员对 Flink 的使用。

SQL 客户端 的目的是提供一种简单的方式来编写、调试和提交表程序到 Flink 集群上,而无需写一行 Java 或 Scala 代码。SQL 客户端命令行界面(CLI) 能够在命令行中检索和可视化分布式应用中实时产生的结果。

Animated demo of the Flink SQL Client CLI running table programs on a cluster

入门

本节介绍如何在命令行里启动(setup)和运行你的第一个 Flink SQL 程序。

SQL 客户端捆绑在常规 Flink 发行版中,因此可以直接运行。它仅需要一个正在运行的 Flink 集群就可以在其中执行表程序。有关设置 Flink 群集的更多信息,请参见集群和部署部分。如果仅想试用 SQL 客户端,也可以使用以下命令启动本地集群:

  1. ./bin/start-cluster.sh

启动 SQL 客户端命令行界面

SQL Client 脚本也位于 Flink 的 bin 目录中。用户可以通过启动嵌入式 standalone 进程或通过连接到远程 SQL Gateway 来启动 SQL 客户端命令行界面。SQL 客户端默认使用 embedded 模式,你可以通过以下方式启动 CLI:

  1. ./bin/sql-client.sh

或者显式使用 embedded 模式:

  1. ./bin/sql-client.sh embedded

若想使用 gateway 模式,你可以通过以下命令启动 SQL 客户端:

  1. ./bin/sql-client.sh gateway --endpoint <gateway address>

Note SQL 客户端目前只支持和 REST API 版本大于 v1 的 REST Endpoint 通信。

参阅 SQL Client startup options 了解更多启动命令。

执行 SQL 查询

命令行界面启动后,你可以使用 HELP 命令列出所有可用的 SQL 语句。输入第一条 SQL 查询语句并按 Enter 键执行,可以验证你的设置及集群连接是否正确:

  1. SELECT 'Hello World';

该查询不需要 table source,并且只产生一行结果。CLI 将从集群中检索结果并将其可视化。按 Q 键退出结果视图。

CLI 为维护和可视化结果提供三种模式

表格模式(table mode)在内存中实体化结果,并将结果用规则的分页表格可视化展示出来。执行如下命令启用:

  1. SET 'sql-client.execution.result-mode' = 'table';

变更日志模式(changelog mode)不会实体化和可视化结果,而是由插入(+)和撤销(-)组成的持续查询产生结果流。

  1. SET 'sql-client.execution.result-mode' = 'changelog';

Tableau模式(tableau mode)更接近传统的数据库,会将执行的结果以制表的形式直接打在屏幕之上。具体显示的内容会取决于作业 执行模式的不同(execution.type):

  1. SET 'sql-client.execution.result-mode' = 'tableau';

注意当你使用这个模式运行一个流式查询的时候,Flink 会将结果持续的打印在当前的屏幕之上。如果这个流式查询的输入是有限的数据集, 那么Flink在处理完所有的数据之后,会自动的停止作业,同时屏幕上的打印也会相应的停止。如果你想提前结束这个查询,那么可以直接使用 CTRL-C 按键,这个会停掉作业同时停止屏幕上的打印。

你可以用如下查询来查看三种结果模式的运行情况:

  1. SELECT name, COUNT(*) AS cnt FROM (VALUES ('Bob'), ('Alice'), ('Greg'), ('Bob')) AS NameTable(name) GROUP BY name;

此查询执行一个有限字数示例:

变更日志模式 下,看到的结果应该类似于:

  1. + Bob, 1
  2. + Alice, 1
  3. + Greg, 1
  4. - Bob, 1
  5. + Bob, 2

表格模式 下,可视化结果表将不断更新,直到表程序以如下内容结束:

  1. Bob, 2
  2. Alice, 1
  3. Greg, 1

Tableau模式 下,如果这个查询以流的方式执行,那么将显示以下内容:

  1. +-----+----------------------+----------------------+
  2. | +/- | name | cnt |
  3. +-----+----------------------+----------------------+
  4. | + | Bob | 1 |
  5. | + | Alice | 1 |
  6. | + | Greg | 1 |
  7. | - | Bob | 1 |
  8. | + | Bob | 2 |
  9. +-----+----------------------+----------------------+
  10. Received a total of 5 rows

如果这个查询以批的方式执行,显示的内容如下:

  1. +-------+-----+
  2. | name | cnt |
  3. +-------+-----+
  4. | Alice | 1 |
  5. | Bob | 2 |
  6. | Greg | 1 |
  7. +-------+-----+
  8. 3 rows in set

这几种结果模式在 SQL 查询的原型设计过程中都非常有用。这些模式的结果都存储在 SQL 客户端 的 Java 堆内存中。为了保持 CLI 界面及时响应,变更日志模式仅显示最近的 1000 个更改。表格模式支持浏览更大的结果,这些结果仅受可用主内存和配置的最大行数sql-client.execution.max-table-result.rows)的限制。

注意 在批处理环境下执行的查询只能用表格模式或者Tableau模式进行检索。

定义查询语句后,可以将其作为长时间运行的独立 Flink 作业提交给集群。配置部分解释如何声明读取数据的 table source,写入数据的 sink 以及配置其他表程序属性的方法。

Key-strokes

There is a list of available key-strokes in SQL Client

Key-Stroke (Linux, Windows(WSL))Key-Stroke (Mac)Description
alt-b, ctrl+⍇Esc-bBackward word
alt-f, Ctrl+⍈Esc-fForward word
alt-cEsc-cCapitalize word
alt-lEsc-lLowercase word
alt-uEsc-uUppercase word
alt-dEsc-dKill word
alt-nEsc-nHistory search forward (behaves same as down line from history in case of empty input)
alt-pEsc-pHistory search backward (behaves same as up line from history in case of empty input)
alt-tEsc-tTranspose words
ctrl-a⌘-aTo the beginning of line
ctrl-e⌘-eTo the end of line
ctrl-b⌘-bBackward char
ctrl-f⌘-fForward char
ctrl-d⌘-dDelete char
ctrl-h⌘-hBackward delete char
ctrl-t⌘-tTranspose chars
ctrl-i⌘-iInvoke completion
ctrl-j⌘-jSubmit a query
ctrl-m⌘-mSubmit a query
ctrl-k⌘-kKill the line to the right from the cursor
ctrl-w⌘-wKill the line to the left from the cursor
ctrl-u⌘-uKill the whole line
ctrl-l⌘-lClear screen
ctrl-n⌘-nDown line from history
ctrl-p⌘-pUp line from history
ctrl-r⌘-rHistory incremental search backward
ctrl-s⌘-sHistory incremental search forward

Getting help

The documentation of the SQL Client commands can be accessed by typing the HELP command.

See also the general SQL documentation.

Configuration

SQL Client startup options

The SQL Client can be started with the following optional CLI commands. They are discussed in detail in the subsequent paragraphs.

  1. ./sql-client [MODE] [OPTIONS]
  2. The following options are available:
  3. Mode "embedded" (default) submits Flink jobs from the local machine.
  4. Syntax: [embedded] [OPTIONS]
  5. "embedded" mode options:
  6. -f,--file <script file> Script file that should be
  7. executed. In this mode, the
  8. client will not open an
  9. interactive terminal.
  10. -h,--help Show the help message with
  11. descriptions of all options.
  12. -hist,--history <History file path> The file which you want to save
  13. the command history into. If not
  14. specified, we will auto-generate
  15. one under your user's home
  16. directory.
  17. -i,--init <initialization file> Script file that used to init
  18. the session context. If get
  19. error in execution, the sql
  20. client will exit. Notice it's
  21. not allowed to add query or
  22. insert into the init file.
  23. -j,--jar <JAR file> A JAR file to be imported into
  24. the session. The file might
  25. contain user-defined classes
  26. needed for the execution of
  27. statements such as functions,
  28. table sources, or sinks. Can be
  29. used multiple times.
  30. -l,--library <JAR directory> A JAR file directory with which
  31. every new session is
  32. initialized. The files might
  33. contain user-defined classes
  34. needed for the execution of
  35. statements such as functions,
  36. table sources, or sinks. Can be
  37. used multiple times.
  38. -pyarch,--pyArchives <arg> Add python archive files for
  39. job. The archive files will be
  40. extracted to the working
  41. directory of python UDF worker.
  42. For each archive file, a target
  43. directory be specified. If the
  44. target directory name is
  45. specified, the archive file will
  46. be extracted to a directory with
  47. the specified name. Otherwise,
  48. the archive file will be
  49. extracted to a directory with
  50. the same name of the archive
  51. file. The files uploaded via
  52. this option are accessible via
  53. relative path. '#' could be used
  54. as the separator of the archive
  55. file path and the target
  56. directory name. Comma (',')
  57. could be used as the separator
  58. to specify multiple archive
  59. files. This option can be used
  60. to upload the virtual
  61. environment, the data files used
  62. in Python UDF (e.g.,
  63. --pyArchives
  64. file:///tmp/py37.zip,file:///tmp
  65. /data.zip#data --pyExecutable
  66. py37.zip/py37/bin/python). The
  67. data files could be accessed in
  68. Python UDF, e.g.: f =
  69. open('data/data.txt', 'r').
  70. -pyclientexec,--pyClientExecutable <arg> The path of the Python
  71. interpreter used to launch the
  72. Python process when submitting
  73. the Python jobs via "flink run"
  74. or compiling the Java/Scala jobs
  75. containing Python UDFs.
  76. -pyexec,--pyExecutable <arg> Specify the path of the python
  77. interpreter used to execute the
  78. python UDF worker (e.g.:
  79. --pyExecutable
  80. /usr/local/bin/python3). The
  81. python UDF worker depends on
  82. Python 3.7+, Apache Beam
  83. (version == 2.43.0), Pip
  84. (version >= 20.3) and SetupTools
  85. (version >= 37.0.0). Please
  86. ensure that the specified
  87. environment meets the above
  88. requirements.
  89. -pyfs,--pyFiles <pythonFiles> Attach custom files for job. The
  90. standard resource file suffixes
  91. such as .py/.egg/.zip/.whl or
  92. directory are all supported.
  93. These files will be added to the
  94. PYTHONPATH of both the local
  95. client and the remote python UDF
  96. worker. Files suffixed with .zip
  97. will be extracted and added to
  98. PYTHONPATH. Comma (',') could be
  99. used as the separator to specify
  100. multiple files (e.g., --pyFiles
  101. file:///tmp/myresource.zip,hdfs:
  102. ///$namenode_address/myresource2
  103. .zip).
  104. -pyreq,--pyRequirements <arg> Specify a requirements.txt file
  105. which defines the third-party
  106. dependencies. These dependencies
  107. will be installed and added to
  108. the PYTHONPATH of the python UDF
  109. worker. A directory which
  110. contains the installation
  111. packages of these dependencies
  112. could be specified optionally.
  113. Use '#' as the separator if the
  114. optional parameter exists (e.g.,
  115. --pyRequirements
  116. file:///tmp/requirements.txt#fil
  117. e:///tmp/cached_dir).
  118. -s,--session <session identifier> The identifier for a session.
  119. 'default' is the default
  120. identifier.
  121. -u,--update <SQL update statement> Deprecated Experimental (for
  122. testing only!) feature:
  123. Instructs the SQL Client to
  124. immediately execute the given
  125. update statement after starting
  126. up. The process is shut down
  127. after the statement has been
  128. submitted to the cluster and
  129. returns an appropriate return
  130. code. Currently, this feature is
  131. only supported for INSERT INTO
  132. statements that declare the
  133. target sink table.Please use
  134. option -f to submit update
  135. statement.
  136. Mode "gateway" mode connects to the SQL gateway for submission.
  137. Syntax: gateway [OPTIONS]
  138. "gateway" mode options:
  139. -e,--endpoint <SQL Gateway address> The address of the remote SQL Gateway
  140. to connect.
  141. -f,--file <script file> Script file that should be executed.
  142. In this mode, the client will not
  143. open an interactive terminal.
  144. -h,--help Show the help message with
  145. descriptions of all options.
  146. -hist,--history <History file path> The file which you want to save the
  147. command history into. If not
  148. specified, we will auto-generate one
  149. under your user's home directory.
  150. -i,--init <initialization file> Script file that used to init the
  151. session context. If get error in
  152. execution, the sql client will exit.
  153. Notice it's not allowed to add query
  154. or insert into the init file.
  155. -s,--session <session identifier> The identifier for a session.
  156. 'default' is the default identifier.
  157. -u,--update <SQL update statement> Deprecated Experimental (for testing
  158. only!) feature: Instructs the SQL
  159. Client to immediately execute the
  160. given update statement after starting
  161. up. The process is shut down after
  162. the statement has been submitted to
  163. the cluster and returns an
  164. appropriate return code. Currently,
  165. this feature is only supported for
  166. INSERT INTO statements that declare
  167. the target sink table.Please use
  168. option -f to submit update statement.

SQL Client Configuration

You can configure the SQL Client by setting the options below, or any valid Flink configuration entry:

  1. SET 'key' = 'value';
KeyDefaultTypeDescription
sql-client.display.max-column-width

Streaming
30IntegerWhen printing the query results, this parameter determines the number of characters shown on screen before truncating.This only applies to columns with variable-length types (e.g. STRING) in streaming mode.Fixed-length types and all types in batch mode are printed using a deterministic column width
sql-client.execution.max-table-result.rows

Batch Streaming
1000000IntegerThe number of rows to cache when in the table mode. If the number of rows exceeds the specified value, it retries the row in the FIFO style.
sql-client.execution.result-mode

Batch Streaming
TABLE

Enum

Determines how the query result should be displayed.

Possible values:
  • “TABLE”: Materializes results in memory and visualizes them in a regular, paginated table representation.
  • “CHANGELOG”: Visualizes the result stream that is produced by a continuous query.
  • “TABLEAU”: Display results in the screen directly in a tableau format.
sql-client.verbose

Batch Streaming
falseBooleanDetermine whether to output the verbose output to the console. If set the option true, it will print the exception stack. Otherwise, it only output the cause.

SQL Client result modes

The CLI supports three modes for maintaining and visualizing results.

The table mode materializes results in memory and visualizes them in a regular, paginated table representation. It can be enabled by executing the following command in the CLI:

  1. SET 'sql-client.execution.result-mode' = 'table';

The result of a query would then look like this, you can use the keys indicated at the bottom of the screen as well as the arrows keys to navigate and open the various records:

  1. name age isHappy dob height
  2. user1 20 true 1995-12-03 1.7
  3. user2 30 true 1972-08-02 1.89
  4. user3 40 false 1983-12-23 1.63
  5. user4 41 true 1977-11-13 1.72
  6. user5 22 false 1998-02-20 1.61
  7. user6 12 true 1969-04-08 1.58
  8. user7 38 false 1987-12-15 1.6
  9. user8 62 true 1996-08-05 1.82
  10. Q Quit + Inc Refresh G Goto Page N Next Page O Open Row
  11. R Refresh - Dec Refresh L Last Page P Prev Page

The changelog mode does not materialize results and visualizes the result stream that is produced by a continuous query consisting of insertions (+) and retractions (-).

  1. SET 'sql-client.execution.result-mode' = 'changelog';

The result of a query would then look like this:

  1. op name age isHappy dob height
  2. +I user1 20 true 1995-12-03 1.7
  3. +I user2 30 true 1972-08-02 1.89
  4. +I user3 40 false 1983-12-23 1.63
  5. +I user4 41 true 1977-11-13 1.72
  6. +I user5 22 false 1998-02-20 1.61
  7. +I user6 12 true 1969-04-08 1.58
  8. +I user7 38 false 1987-12-15 1.6
  9. +I user8 62 true 1996-08-05 1.82
  10. Q Quit + Inc Refresh O Open Row
  11. R Refresh - Dec Refresh

The tableau mode is more like a traditional way which will display the results in the screen directly with a tableau format. The displaying content will be influenced by the query execution type (execution.type).

  1. SET 'sql-client.execution.result-mode' = 'tableau';

The result of a query would then look like this:

  1. +----+--------------------------------+-------------+---------+------------+--------------------------------+
  2. | op | name | age | isHappy | dob | height |
  3. +----+--------------------------------+-------------+---------+------------+--------------------------------+
  4. | +I | user1 | 20 | true | 1995-12-03 | 1.7 |
  5. | +I | user2 | 30 | true | 1972-08-02 | 1.89 |
  6. | +I | user3 | 40 | false | 1983-12-23 | 1.63 |
  7. | +I | user4 | 41 | true | 1977-11-13 | 1.72 |
  8. | +I | user5 | 22 | false | 1998-02-20 | 1.61 |
  9. | +I | user6 | 12 | true | 1969-04-08 | 1.58 |
  10. | +I | user7 | 38 | false | 1987-12-15 | 1.6 |
  11. | +I | user8 | 62 | true | 1996-08-05 | 1.82 |
  12. +----+--------------------------------+-------------+---------+------------+--------------------------------+
  13. Received a total of 8 rows

Note that when you use this mode with streaming query, the result will be continuously printed on the console. If the input data of this query is bounded, the job will terminate after Flink processed all input data, and the printing will also be stopped automatically. Otherwise, if you want to terminate a running query, just type CTRL-C in this case, the job and the printing will be stopped.

All these result modes can be useful during the prototyping of SQL queries. In all these modes, results are stored in the Java heap memory of the SQL Client. In order to keep the CLI interface responsive, the changelog mode only shows the latest 1000 changes. The table mode allows for navigating through bigger results that are only limited by the available main memory and the configured maximum number of rows (sql-client.execution.max-table-result.rows).

Attention Queries that are executed in a batch environment, can only be retrieved using the table or tableau result mode.

Initialize Session Using SQL Files

A SQL query needs a configuration environment in which it is executed. SQL Client supports the -i startup option to execute an initialization SQL file to setup environment when starting up the SQL Client. The so-called initialization SQL file can use DDLs to define available catalogs, table sources and sinks, user-defined functions, and other properties required for execution and deployment.

An example of such a file is presented below.

  1. -- Define available catalogs
  2. CREATE CATALOG MyCatalog
  3. WITH (
  4. 'type' = 'hive'
  5. );
  6. USE CATALOG MyCatalog;
  7. -- Define available database
  8. CREATE DATABASE MyDatabase;
  9. USE MyDatabase;
  10. -- Define TABLE
  11. CREATE TABLE MyTable(
  12. MyField1 INT,
  13. MyField2 STRING
  14. ) WITH (
  15. 'connector' = 'filesystem',
  16. 'path' = '/path/to/something',
  17. 'format' = 'csv'
  18. );
  19. -- Define VIEW
  20. CREATE VIEW MyCustomView AS SELECT MyField2 FROM MyTable;
  21. -- Define user-defined functions here.
  22. CREATE FUNCTION foo.bar.AggregateUDF AS myUDF;
  23. -- Properties that change the fundamental execution behavior of a table program.
  24. SET 'execution.runtime-mode' = 'streaming'; -- execution mode either 'batch' or 'streaming'
  25. SET 'sql-client.execution.result-mode' = 'table'; -- available values: 'table', 'changelog' and 'tableau'
  26. SET 'sql-client.execution.max-table-result.rows' = '10000'; -- optional: maximum number of maintained rows
  27. SET 'parallelism.default' = '1'; -- optional: Flink's parallelism (1 by default)
  28. SET 'pipeline.auto-watermark-interval' = '200'; --optional: interval for periodic watermarks
  29. SET 'pipeline.max-parallelism' = '10'; -- optional: Flink's maximum parallelism
  30. SET 'table.exec.state.ttl' = '1000'; -- optional: table program's idle state time
  31. SET 'restart-strategy.type' = 'fixed-delay';
  32. -- Configuration options for adjusting and tuning table programs.
  33. SET 'table.optimizer.join-reorder-enabled' = 'true';
  34. SET 'table.exec.spill-compression.enabled' = 'true';
  35. SET 'table.exec.spill-compression.block-size' = '128kb';

This configuration:

  • connects to Hive catalogs and uses MyCatalog as the current catalog with MyDatabase as the current database of the catalog,
  • defines a table MyTable that can read data from a CSV file,
  • defines a view MyCustomView that declares a virtual table using a SQL query,
  • defines a user-defined function myUDF that can be instantiated using the class name,
  • uses streaming mode for running statements and a parallelism of 1,
  • runs exploratory queries in the table result mode,
  • and makes some planner adjustments around join reordering and spilling via configuration options.

When using -i <init.sql> option to initialize SQL Client session, the following statements are allowed in an initialization SQL file:

  • DDL(CREATE/DROP/ALTER),
  • USE CATALOG/DATABASE,
  • LOAD/UNLOAD MODULE,
  • SET command,
  • RESET command.

When execute queries or insert statements, please enter the interactive mode or use the -f option to submit the SQL statements.

Attention If SQL Client receives errors during initialization, SQL Client will exit with error messages.

Dependencies

The SQL Client does not require setting up a Java project using Maven, Gradle, or sbt. Instead, you can pass the dependencies as regular JAR files that get submitted to the cluster. You can either specify each JAR file separately (using --jar) or define entire library directories (using --library). For connectors to external systems (such as Apache Kafka) and corresponding data formats (such as JSON), Flink provides ready-to-use JAR bundles. These JAR files can be downloaded for each release from the Maven central repository.

The full list of offered SQL JARs can be found on the connection to external systems page.

You can refer to the configuration section for information on how to configure connector and format dependencies.

Usage

SQL Client allows users to submit jobs either within the interactive command line or using -f option to execute sql file.

In both modes, SQL Client supports to parse and execute all types of the Flink supported SQL statements.

Interactive Command Line

In interactive Command Line, the SQL Client reads user inputs and executes the statement terminated by a semicolon (;).

SQL Client will print success message if the statement is executed successfully. When getting errors, SQL Client will also print error messages. By default, the error message only contains the error cause. In order to print the full exception stack for debugging, please set the sql-client.verbose to true through command SET 'sql-client.verbose' = 'true';.

Execute SQL Files

SQL Client supports to execute a SQL script file with the -f option. SQL Client will execute statements one by one in the SQL script file and print execution messages for each executed statements. Once a statement fails, the SQL Client will exit and all the remaining statements will not be executed.

An example of such a file is presented below.

  1. CREATE TEMPORARY TABLE users (
  2. user_id BIGINT,
  3. user_name STRING,
  4. user_level STRING,
  5. region STRING,
  6. PRIMARY KEY (user_id) NOT ENFORCED
  7. ) WITH (
  8. 'connector' = 'upsert-kafka',
  9. 'topic' = 'users',
  10. 'properties.bootstrap.servers' = '...',
  11. 'key.format' = 'csv',
  12. 'value.format' = 'avro'
  13. );
  14. -- set sync mode
  15. SET 'table.dml-sync' = 'true';
  16. -- set the job name
  17. SET 'pipeline.name' = 'SqlJob';
  18. -- set the queue that the job submit to
  19. SET 'yarn.application.queue' = 'root';
  20. -- set the job parallelism
  21. SET 'parallelism.default' = '100';
  22. -- restore from the specific savepoint path
  23. SET 'execution.savepoint.path' = '/tmp/flink-savepoints/savepoint-cca7bc-bb1e257f0dab';
  24. INSERT INTO pageviews_enriched
  25. SELECT *
  26. FROM pageviews AS p
  27. LEFT JOIN users FOR SYSTEM_TIME AS OF p.proctime AS u
  28. ON p.user_id = u.user_id;

This configuration:

  • defines a temporal table source users that reads from a CSV file,
  • set the properties, e.g job name,
  • set the savepoint path,
  • submit a sql job that load the savepoint from the specified savepoint path.

Attention Compared to the interactive mode, SQL Client will stop execution and exits when there are errors.

Execute a set of SQL statements

SQL Client execute each INSERT INTO statement as a single Flink job. However, this is sometimes not optimal because some part of the pipeline can be reused. SQL Client supports STATEMENT SET syntax to execute a set of SQL statements. This is an equivalent feature with StatementSet in Table API. The STATEMENT SET syntax encloses one or more INSERT INTO statements. All statements in a STATEMENT SET block are holistically optimized and executed as a single Flink job. Joint optimization and execution allows for reusing common intermediate results and can therefore significantly improve the efficiency of executing multiple queries.

Syntax

  1. EXECUTE STATEMENT SET
  2. BEGIN
  3. -- one or more INSERT INTO statements
  4. { INSERT INTO|OVERWRITE <select_statement>; }+
  5. END;

Attention The statements of enclosed in the STATEMENT SET must be separated by a semicolon (;). The old syntax BEGIN STATEMENT SET; ... END; is deprecated, may be removed in the future version.

SQL CLI

  1. Flink SQL> CREATE TABLE pageviews (
  2. > user_id BIGINT,
  3. > page_id BIGINT,
  4. > viewtime TIMESTAMP,
  5. > proctime AS PROCTIME()
  6. > ) WITH (
  7. > 'connector' = 'kafka',
  8. > 'topic' = 'pageviews',
  9. > 'properties.bootstrap.servers' = '...',
  10. > 'format' = 'avro'
  11. > );
  12. [INFO] Execute statement succeed.
  13. Flink SQL> CREATE TABLE pageview (
  14. > page_id BIGINT,
  15. > cnt BIGINT
  16. > ) WITH (
  17. > 'connector' = 'jdbc',
  18. > 'url' = 'jdbc:mysql://localhost:3306/mydatabase',
  19. > 'table-name' = 'pageview'
  20. > );
  21. [INFO] Execute statement succeed.
  22. Flink SQL> CREATE TABLE uniqueview (
  23. > page_id BIGINT,
  24. > cnt BIGINT
  25. > ) WITH (
  26. > 'connector' = 'jdbc',
  27. > 'url' = 'jdbc:mysql://localhost:3306/mydatabase',
  28. > 'table-name' = 'uniqueview'
  29. > );
  30. [INFO] Execute statement succeed.
  31. Flink SQL> EXECUTE STATEMENT SET
  32. > BEGIN
  33. >
  34. > INSERT INTO pageview
  35. > SELECT page_id, count(1)
  36. > FROM pageviews
  37. > GROUP BY page_id;
  38. >
  39. > INSERT INTO uniqueview
  40. > SELECT page_id, count(distinct user_id)
  41. > FROM pageviews
  42. > GROUP BY page_id;
  43. >
  44. > END;
  45. [INFO] Submitting SQL update statement to the cluster...
  46. [INFO] SQL update statement has been successfully submitted to the cluster:
  47. Job ID: 6b1af540c0c0bb3fcfcad50ac037c862

SQL File

  1. CREATE TABLE pageviews (
  2. user_id BIGINT,
  3. page_id BIGINT,
  4. viewtime TIMESTAMP,
  5. proctime AS PROCTIME()
  6. ) WITH (
  7. 'connector' = 'kafka',
  8. 'topic' = 'pageviews',
  9. 'properties.bootstrap.servers' = '...',
  10. 'format' = 'avro'
  11. );
  12. CREATE TABLE pageview (
  13. page_id BIGINT,
  14. cnt BIGINT
  15. ) WITH (
  16. 'connector' = 'jdbc',
  17. 'url' = 'jdbc:mysql://localhost:3306/mydatabase',
  18. 'table-name' = 'pageview'
  19. );
  20. CREATE TABLE uniqueview (
  21. page_id BIGINT,
  22. cnt BIGINT
  23. ) WITH (
  24. 'connector' = 'jdbc',
  25. 'url' = 'jdbc:mysql://localhost:3306/mydatabase',
  26. 'table-name' = 'uniqueview'
  27. );
  28. EXECUTE STATEMENT SET
  29. BEGIN
  30. INSERT INTO pageview
  31. SELECT page_id, count(1)
  32. FROM pageviews
  33. GROUP BY page_id;
  34. INSERT INTO uniqueview
  35. SELECT page_id, count(distinct user_id)
  36. FROM pageviews
  37. GROUP BY page_id;
  38. END;

Execute DML statements sync/async

By default, SQL Client executes DML statements asynchronously. That means, SQL Client will submit a job for the DML statement to a Flink cluster, and not wait for the job to finish. So SQL Client can submit multiple jobs at the same time. This is useful for streaming jobs, which are long-running in general.

SQL Client makes sure that a statement is successfully submitted to the cluster. Once the statement is submitted, the CLI will show information about the Flink job.

  1. Flink SQL> INSERT INTO MyTableSink SELECT * FROM MyTableSource;
  2. [INFO] Table update statement has been successfully submitted to the cluster:
  3. Cluster ID: StandaloneClusterId
  4. Job ID: 6f922fe5cba87406ff23ae4a7bb79044

Attention The SQL Client does not track the status of the running Flink job after submission. The CLI process can be shutdown after the submission without affecting the detached query. Flink’s restart strategy takes care of the fault-tolerance. Please use the job statements to monitor the detached query status or stop the detached query.

However, for batch users, it’s more common that the next DML statement requires waiting until the previous DML statement finishes. In order to execute DML statements synchronously, you can set table.dml-sync option to true in SQL Client.

  1. Flink SQL> SET 'table.dml-sync' = 'true';
  2. [INFO] Session property has been set.
  3. Flink SQL> INSERT INTO MyTableSink SELECT * FROM MyTableSource;
  4. [INFO] Submitting SQL update statement to the cluster...
  5. [INFO] Execute statement in sync mode. Please wait for the execution finish...
  6. [INFO] Complete execution of the SQL update statement.

Attention If you want to terminate the job, just type CTRL-C to cancel the execution.

Start a SQL Job from a savepoint

Flink supports to start the job with specified savepoint. In SQL Client, it’s allowed to use SET command to specify the path of the savepoint.

  1. Flink SQL> SET 'execution.savepoint.path' = '/tmp/flink-savepoints/savepoint-cca7bc-bb1e257f0dab';
  2. [INFO] Session property has been set.
  3. -- all the following DML statements will be restroed from the specified savepoint path
  4. Flink SQL> INSERT INTO ...

When the path to savepoint is specified, Flink will try to restore the state from the savepoint when executing all the following DML statements.

Because the specified savepoint path will affect all the following DML statements, you can use RESET command to reset this config option, i.e. disable restoring from savepoint.

  1. Flink SQL> RESET execution.savepoint.path;
  2. [INFO] Session property has been reset.

For more details about creating and managing savepoints, please refer to Job Lifecycle Management.

Define a Custom Job Name

SQL Client supports to define job name for queries and DML statements through SET command.

  1. Flink SQL> SET 'pipeline.name' = 'kafka-to-hive';
  2. [INFO] Session property has been set.
  3. -- all the following DML statements will use the specified job name.
  4. Flink SQL> INSERT INTO ...

Because the specified job name will affect all the following queries and DML statements, you can also use RESET command to reset this configuration, i.e. use default job names.

  1. Flink SQL> RESET pipeline.name;
  2. [INFO] Session property has been reset.

If the option pipeline.name is not specified, SQL Client will generate a default name for the submitted job, e.g. insert-into_<sink_table_name> for INSERT INTO statements.

Monitoring Job Status

SQL Client supports to list jobs status in the cluster through SHOW JOBS statements.

  1. Flink SQL> SHOW JOBS;
  2. +----------------------------------+---------------+----------+-------------------------+
  3. | job id | job name | status | start time |
  4. +----------------------------------+---------------+----------+-------------------------+
  5. | 228d70913eab60dda85c5e7f78b5782c | kafka-to-hive | RUNNING | 2023-02-11T05:03:51.523 |
  6. +----------------------------------+---------------+----------+-------------------------+

Terminating a Job

SQL Client supports to stop jobs with or without savepoints through STOP JOB statements.

  1. Flink SQL> STOP JOB '228d70913eab60dda85c5e7f78b5782c' WITH SAVEPOINT;
  2. +-----------------------------------------+
  3. | savepoint path |
  4. +-----------------------------------------+
  5. | file:/tmp/savepoint-3addd4-0b224d9311e6 |
  6. +-----------------------------------------+

The savepoint path could be specified with state.savepoints.dir either in the cluster configuration or session configuration (the latter would take precedence).

For more details about stopping jobs, please refer to Job Statements.