Databases

Reactive DB2 Client

The Reactive DB2 Client is a client for DB2 with a straightforward API focusing on scalability and low overhead.

The client is reactive and non-blocking, allowing to handle many database connections with a single thread.

Features

Support for DB2 on Linux, Unix, and Windows
Limited support for DB2 on z/OS
Event driven
Lightweight
Built-in connection pooling
Prepared queries caching
Batch and cursor
Row streaming
RxJava API
Direct memory to object without unnecessary copies
Java 8 Date and Time
SSL/TLS
HTTP/1.x CONNECT, SOCKS4a or SOCKS5 proxy support

Current limitations

No stored procedures support
Some column types (e.g. BLOB and CLOB) are not supported

Usage

To use the Reactive DB2 Client add the following dependency to the dependencies section of your build descriptor:

Maven (in your pom.xml):

<dependency>
 <groupId>io.vertx</groupId>
 <artifactId>vertx-db2-client</artifactId>
 <version>4.5.11</version>
</dependency>

Gradle (in your build.gradle file):

dependencies {
 compile 'io.vertx:vertx-db2-client:4.5.11'
}

Getting started

Here is the simplest way to connect, query and disconnect

DB2ConnectOptions connectOptions = new DB2ConnectOptions()
  .setPort(50000)
  .setHost("the-host")
  .setDatabase("the-db")
  .setUser("user")
  .setPassword("secret");

// Pool options
PoolOptions poolOptions = new PoolOptions()
  .setMaxSize(5);

// Create the client pool
Pool client = DB2Builder.pool()
  .with(poolOptions)
  .connectingTo(connectOptions)
  .build();

// A simple query
client
  .query("SELECT * FROM users WHERE id='julien'")
  .execute()
  .onComplete(ar -> {
  if (ar.succeeded()) {
    RowSet<Row> result = ar.result();
    System.out.println("Got " + result.size() + " rows ");
  } else {
    System.out.println("Failure: " + ar.cause().getMessage());
  }

  // Now close the pool
  client.close();
});

Connecting to DB2

Most of the time you will use a pool to connect to DB2:

DB2ConnectOptions connectOptions = new DB2ConnectOptions()
  .setPort(50000)
  .setHost("the-host")
  .setDatabase("the-db")
  .setUser("user")
  .setPassword("secret");

// Pool options
PoolOptions poolOptions = new PoolOptions()
  .setMaxSize(5);

// Create the pooled client
SqlClient client = DB2Builder.client()
  .with(poolOptions)
  .connectingTo(connectOptions)
  .using(vertx)
  .build();

The pooled client uses a connection pool and any operation will borrow a connection from the pool to execute the operation and release it to the pool.

If you are running with Vert.x you can pass it your Vertx instance:

DB2ConnectOptions connectOptions = new DB2ConnectOptions()
  .setPort(50000)
  .setHost("the-host")
  .setDatabase("the-db")
  .setUser("user")
  .setPassword("secret");

// Pool options
PoolOptions poolOptions = new PoolOptions()
  .setMaxSize(5);
// Create the pooled client
SqlClient client = DB2Builder.client()
  .with(poolOptions)
  .connectingTo(connectOptions)
  .using(vertx)
  .build();

You need to release the client when you don’t need it anymore:

client.close();

When you need to execute several operations on the same connection, you need to acquire a connection from a pool

You can easily get one from the pool:

DB2ConnectOptions connectOptions = new DB2ConnectOptions()
  .setPort(50000)
  .setHost("the-host")
  .setDatabase("the-db")
  .setUser("user")
  .setPassword("secret");

// Pool options
PoolOptions poolOptions = new PoolOptions()
  .setMaxSize(5);

// Create the pooled client
Pool client = DB2Builder.pool()
  .with(poolOptions)
  .connectingTo(connectOptions)
  .using(vertx)
  .build();

// Get a connection from the pool
client.getConnection().compose(conn -> {
  System.out.println("Got a connection from the pool");

  // All operations execute on the same connection
  return conn
    .query("SELECT * FROM users WHERE id='julien'")
    .execute()
    .compose(res -> conn
      .query("SELECT * FROM users WHERE id='emad'")
      .execute())
    .onComplete(ar -> {
      // Release the connection to the pool
      conn.close();
    });
}).onComplete(ar -> {
  if (ar.succeeded()) {

    System.out.println("Done");
  } else {
    System.out.println("Something went wrong " + ar.cause().getMessage());
  }
});

Once you are done with the connection you must close it to release it to the pool, so it can be reused.

Pool versus pooled client

The DB2Builder allows you to create a pool or a pooled client

SqlClient client = DB2Builder.client()
  .with(poolOptions)
  .connectingTo(connectOptions)
  .using(vertx)
  .build();

// Pipelined
Future<RowSet<Row>> res1 = client.query(sql).execute();

// Connection pool
Pool pool = DB2Builder.pool()
  .with(poolOptions)
  .connectingTo(connectOptions)
  .using(vertx)
  .build();

// Not pipelined
Future<RowSet<Row>> res2 = pool.query(sql).execute();

pool operations are not pipelined, only connections acquired from the pool are pipelined
pooled client operations are pipelined, you cannot acquire a connection from a pooled client

Pool sharing

You can share an pool between multiple verticles or instances of the same verticle. Such pool should be created outside a verticle otherwise it will be closed when the verticle that created it is undeployed

Pool pool = DB2Builder.pool()
  .with(new PoolOptions().setMaxSize(maxSize))
  .connectingTo(database)
  .using(vertx)
  .build();
vertx.deployVerticle(() -> new AbstractVerticle() {
  @Override
  public void start() throws Exception {
    // Use the pool
  }
}, new DeploymentOptions().setInstances(4));

You can also create a shared pool in each verticle:

vertx.deployVerticle(() -> new AbstractVerticle() {
  Pool pool;
  @Override
  public void start() {
    // Get or create a shared pool
    // this actually creates a lease to the pool
    // when the verticle is undeployed, the lease will be released automaticaly
    pool = DB2Builder.pool()
      .with(new PoolOptions()
        .setMaxSize(maxSize)
        .setShared(true)
        .setName("my-pool"))
      .connectingTo(database)
      .using(vertx)
      .build();
  }
}, new DeploymentOptions().setInstances(4));

The first time a shared pool is created it will create the resources for the pool. Subsequent calls will reuse this pool and create a lease to this pool. The resources are disposed after all leases have been closed.

By default, a pool reuses the current event-loop when it needs to create a TCP connection. The shared pool will therefore randomly use event-loops of verticles using it.

You can assign a number of event loop a pool will use independently of the context using it

Pool pool = DB2Builder.pool()
  .with(new PoolOptions()
    .setMaxSize(maxSize)
    .setShared(true)
    .setName("my-pool")
    .setEventLoopSize(4))
  .connectingTo(database)
  .using(vertx)
  .build();

Configuration

There are several alternatives for you to configure the client.

data object

A simple way to configure the client is to specify a DB2ConnectOptions data object.

DB2ConnectOptions connectOptions = new DB2ConnectOptions()
  .setPort(50000)
  .setHost("the-host")
  .setDatabase("the-db")
  .setUser("user")
  .setPassword("secret");

// Pool Options
PoolOptions poolOptions = new PoolOptions().setMaxSize(5);

// Create the pool from the data object
Pool pool = DB2Builder.pool()
  .with(poolOptions)
  .connectingTo(connectOptions)
  .using(vertx)
  .build();

pool.getConnection()
  .onComplete(ar -> {
    // Handling your connection
  });

You can also configure the generic properties with the setProperties or addProperty methods. Note setProperties will override the default client properties.

Connection URI

Apart from configuring with a DB2ConnectOptions data object, We also provide you an alternative way to connect when you want to configure with a connection URI:

String connectionUri = "db2://dbuser:[email protected]:50000/mydb";

// Create the pool from the connection URI
Pool pool = DB2Builder.pool()
  .connectingTo(connectionUri)
  .using(vertx)
  .build();

// Create the connection from the connection URI
DB2Connection.connect(vertx, connectionUri)
  .onComplete(res -> {
    // Handling your connection
  });

The URI format for a connection string is:

db2://[user[:[password]]@]host[:port][/database][?<key1>=<value1>[&<key2>=<value2>]]

Currently, the client supports the following parameter keys:

host
port
user
password
database

Configuring parameters in connection URI will override the default properties.

Connect retries

You can configure the client to retry when a connection fails to be established.

options
  .setReconnectAttempts(2)
  .setReconnectInterval(1000);

Running queries

When you don’t need a transaction or run single queries, you can run queries directly on the pool; the pool will use one of its connection to run the query and return the result to you.

Here is how to run simple queries:

client
  .query("SELECT * FROM users WHERE id='andy'")
  .execute()
  .onComplete(ar -> {
  if (ar.succeeded()) {
    RowSet<Row> result = ar.result();
    System.out.println("Got " + result.size() + " rows ");
  } else {
    System.out.println("Failure: " + ar.cause().getMessage());
  }
});

Prepared queries

You can do the same with prepared queries.

The SQL string can refer to parameters by position, using the database syntax `?`

client
  .preparedQuery("SELECT * FROM users WHERE id=?")
  .execute(Tuple.of("andy"))
  .onComplete(ar -> {
  if (ar.succeeded()) {
    RowSet<Row> rows = ar.result();
    System.out.println("Got " + rows.size() + " rows ");
  } else {
    System.out.println("Failure: " + ar.cause().getMessage());
  }
});

Query methods provides an asynchronous RowSet instance that works for SELECT queries

client
  .preparedQuery("SELECT first_name, last_name FROM users")
  .execute()
  .onComplete(ar -> {
  if (ar.succeeded()) {
    RowSet<Row> rows = ar.result();
    for (Row row : rows) {
      System.out.println("User " + row.getString(0) + " " + row.getString(1));
    }
  } else {
    System.out.println("Failure: " + ar.cause().getMessage());
  }
});

or UPDATE/INSERT queries:

client
  .preparedQuery("INSERT INTO users (first_name, last_name) VALUES (?, ?)")
  .execute(Tuple.of("Andy", "Guibert"))
  .onComplete(ar -> {
  if (ar.succeeded()) {
    RowSet<Row> rows = ar.result();
    System.out.println(rows.rowCount());
  } else {
    System.out.println("Failure: " + ar.cause().getMessage());
  }
});

The Row gives you access to your data by index

System.out.println("User " + row.getString(0) + " " + row.getString(1));

Column indexes start at 0, not at 1.

Alternatively, data can be retrieved by name:

System.out.println("User " + row.getString("first_name") + " " + row.getString("last_name"));

The client will not do any magic here and the column name is identified with the name in the table regardless of how your SQL text is.

You can access a wide variety of of types

String firstName = row.getString("first_name");
Boolean male = row.getBoolean("male");
Integer age = row.getInteger("age");

You can use cached prepared statements to execute one-shot prepared queries:

connectOptions.setCachePreparedStatements(true);
client
  .preparedQuery("SELECT * FROM users WHERE id = ?")
  .execute(Tuple.of("julien"))
  .onComplete(ar -> {
    if (ar.succeeded()) {
      RowSet<Row> rows = ar.result();
      System.out.println("Got " + rows.size() + " rows ");
    } else {
      System.out.println("Failure: " + ar.cause().getMessage());
    }
  });

You can create a PreparedStatement and manage the lifecycle by yourself.

sqlConnection
  .prepare("SELECT * FROM users WHERE id = ?")
  .onComplete(ar -> {
    if (ar.succeeded()) {
      PreparedStatement preparedStatement = ar.result();
      preparedStatement.query()
        .execute(Tuple.of("julien"))
        .onComplete(ar2 -> {
          if (ar2.succeeded()) {
            RowSet<Row> rows = ar2.result();
            System.out.println("Got " + rows.size() + " rows ");
            preparedStatement.close();
          } else {
            System.out.println("Failure: " + ar2.cause().getMessage());
          }
        });
    } else {
      System.out.println("Failure: " + ar.cause().getMessage());
    }
  });

Batches

You can execute prepared batch

List<Tuple> batch = new ArrayList<>();
batch.add(Tuple.of("julien", "Julient Viet"));
batch.add(Tuple.of("emad", "Emad Alblueshi"));
batch.add(Tuple.of("andy", "Andy Guibert"));

// Execute the prepared batch
client
  .preparedQuery("INSERT INTO USERS (id, name) VALUES (?, ?)")
  .executeBatch(batch)
  .onComplete(res -> {
  if (res.succeeded()) {

    // Process rows
    RowSet<Row> rows = res.result();
  } else {
    System.out.println("Batch failed " + res.cause());
  }
});

You can fetch generated keys by wrapping your query in SELECT <COLUMNS> FROM FINAL TABLE ( <SQL> ), for example:

client
  .preparedQuery("SELECT color_id FROM FINAL TABLE ( INSERT INTO color (color_name) VALUES (?), (?), (?) )")
  .execute(Tuple.of("white", "red", "blue"))
  .onComplete(ar -> {
    if (ar.succeeded()) {
      RowSet<Row> rows = ar.result();
      System.out.println("Inserted " + rows.rowCount() + " new rows.");
      for (Row row : rows) {
        System.out.println("generated key: " + row.getInteger("color_id"));
      }
    } else {
      System.out.println("Failure: " + ar.cause().getMessage());
    }
  });

Using connections

Getting a connection

When you need to execute sequential queries (without a transaction), you can create a new connection or borrow one from the pool. Remember that between acquiring the connection from the pool and returning it to the pool, you should take care of the connection because it might be closed by the server for some reason such as an idle time out.

pool
  .getConnection()
  .compose(connection ->
    connection
      .preparedQuery("INSERT INTO Users (first_name,last_name) VALUES (?, ?)")
      .executeBatch(Arrays.asList(
        Tuple.of("Julien", "Viet"),
        Tuple.of("Andy", "Guibert")
      ))
      .compose(res -> connection
        // Do something with rows
        .query("SELECT COUNT(*) FROM Users")
        .execute()
        .map(rows -> rows.iterator().next().getInteger(0)))
      // Return the connection to the pool
      .eventually(v -> connection.close())
  ).onSuccess(count -> {
  System.out.println("Insert users, now the number of users is " + count);
});

Prepared queries can be created:

connection
  .prepare("SELECT * FROM users WHERE first_name LIKE ?")
  .compose(pq ->
    pq.query()
      .execute(Tuple.of("Andy"))
      .eventually(v -> pq.close())
  ).onSuccess(rows -> {
  // All rows
});

Simplified connection API

When you use a pool, you can call withConnection to pass it a function executed within a connection.

It borrows a connection from the pool and calls the function with this connection.

The function must return a future of an arbitrary result.

After the future completes, the connection is returned to the pool and the overall result is provided.

pool.withConnection(connection ->
  connection
    .preparedQuery("INSERT INTO Users (first_name,last_name) VALUES (?, ?)")
    .executeBatch(Arrays.asList(
      Tuple.of("Julien", "Viet"),
      Tuple.of("Andy", "Guibert")
    ))
    .compose(res -> connection
      // Do something with rows
      .query("SELECT COUNT(*) FROM Users")
      .execute()
      .map(rows -> rows.iterator().next().getInteger(0)))
).onSuccess(count -> {
  System.out.println("Insert users, now the number of users is " + count);
});

Using transactions

Transactions with connections

You can execute transaction using SQL BEGIN/COMMIT/ROLLBACK, if you do so you must use a SqlConnection and manage it yourself.

Or you can use the transaction API of SqlConnection:

pool.getConnection()
  // Transaction must use a connection
  .onSuccess(conn -> {
    // Begin the transaction
    conn.begin()
      .compose(tx -> conn
        // Various statements
        .query("INSERT INTO Users (first_name,last_name) VALUES ('Julien','Viet')")
        .execute()
        .compose(res2 -> conn
          .query("INSERT INTO Users (first_name,last_name) VALUES ('Andy','Guibert')")
          .execute())
        // Commit the transaction
        .compose(res3 -> tx.commit()))
      // Return the connection to the pool
      .eventually(v -> conn.close())
      .onSuccess(v -> System.out.println("Transaction succeeded"))
      .onFailure(err -> System.out.println("Transaction failed: " + err.getMessage()));
  });

When the database server reports the current transaction is failed (e.g the infamous current transaction is aborted, commands ignored until end of transaction block), the transaction is rollbacked and the completion future is failed with a TransactionRollbackException:

tx.completion()
  .onFailure(err -> {
    System.out.println("Transaction failed => rolled back");
  });

Simplified transaction API

When you use a pool, you can call withTransaction to pass it a function executed within a transaction.

It borrows a connection from the pool, begins the transaction and calls the function with a client executing all operations in the scope of this transaction.

The function must return a future of an arbitrary result:

when the future succeeds the client will commit the transaction
when the future fails the client will rollback the transaction

After the transaction completes, the connection is returned to the pool and the overall result is provided.

pool.withTransaction(client -> client
  .query("INSERT INTO Users (first_name,last_name) VALUES ('Julien','Viet')")
  .execute()
  .flatMap(res -> client
    .query("INSERT INTO Users (first_name,last_name) VALUES ('Andy','Guibert')")
    .execute()
    // Map to a message result
    .map("Users inserted")))
  .onSuccess(v -> System.out.println("Transaction succeeded"))
  .onFailure(err -> System.out.println("Transaction failed: " + err.getMessage()));

Cursors and streaming

By default, prepared query execution fetches all rows, you can use a Cursor to control the amount of rows you want to read:

connection
  .prepare("SELECT * FROM users WHERE first_name LIKE ?")
  .onComplete(ar0 -> {
  if (ar0.succeeded()) {
    PreparedStatement pq = ar0.result();

    // Cursors require to run within a transaction
    connection
      .begin()
      .onComplete(ar1 -> {
      if (ar1.succeeded()) {
        Transaction tx = ar1.result();

        // Create a cursor
        Cursor cursor = pq.cursor(Tuple.of("julien"));

        // Read 50 rows
        cursor
          .read(50)
          .onComplete(ar2 -> {
          if (ar2.succeeded()) {
            RowSet<Row> rows = ar2.result();

            // Check for more ?
            if (cursor.hasMore()) {
              // Repeat the process...
            } else {
              // No more rows - commit the transaction
              tx.commit();
            }
          }
        });
      }
    });
  }
});

Cursors shall be closed when they are released prematurely:

cursor
  .read(50)
  .onComplete(ar2 -> {
  if (ar2.succeeded()) {
    // Close the cursor
    cursor.close();
  }
});

A stream API is also available for cursors, which can be more convenient, specially with the Rxified version.

connection
  .prepare("SELECT * FROM users WHERE first_name LIKE ?")
  .onComplete(ar0 -> {
  if (ar0.succeeded()) {
    PreparedStatement pq = ar0.result();

    // Streams require to run within a transaction
    connection
      .begin()
      .onComplete(ar1 -> {
      if (ar1.succeeded()) {
        Transaction tx = ar1.result();

        // Fetch 50 rows at a time
        RowStream<Row> stream = pq.createStream(50, Tuple.of("julien"));

        // Use the stream
        stream.exceptionHandler(err -> {
          System.out.println("Error: " + err.getMessage());
        });
        stream.endHandler(v -> {
          tx.commit();
          System.out.println("End of stream");
        });
        stream.handler(row -> {
          System.out.println("User: " + row.getString("last_name"));
        });
      }
    });
  }
});

The stream read the rows by batch of 50 and stream them, when the rows have been passed to the handler, a new batch of 50 is read and so on.

The stream can be resumed or paused, the loaded rows will remain in memory until they are delivered and the cursor will stop iterating.

Tracing queries

The SQL client can trace query execution when Vert.x has tracing enabled.

The client reports the following client spans:

Query operation name
tags
db.system: the database management system product
db.user: the database username
db.instance: the database instance
db.statement: the SQL query
db.type: sql

The default tracing policy is PROPAGATE, the client will only create a span when involved in an active trace.

You can change the client policy with setTracingPolicy, e.g you can set ALWAYS to always report a span:

options.setTracingPolicy(TracingPolicy.ALWAYS);

DB2 type mapping

Currently the client supports the following DB2 types

BOOLEAN (java.lang.Boolean) (DB2 LUW only)
SMALLINT (java.lang.Short)
INTEGER (java.lang.Integer)
BIGINT (java.lang.Long)
REAL (java.lang.Float)
DOUBLE (java.lang.Double)
DECIMAL (io.vertx.sqlclient.data.Numeric)
CHAR (java.lang.String)
VARCHAR (java.lang.String)
ENUM (java.lang.String)
DATE (java.time.LocalDate)
TIME (java.time.LocalTime)
TIMESTAMP (java.time.LocalDateTime)
BINARY (byte[])
VARBINARY (byte[])
ROWID (io.vertx.db2client.impl.drda.DB2RowId or java.sql.RowId) (DB2 z/OS only)

Some types that are currently NOT supported are:

XML
BLOB
CLOB
DBCLOB
GRAPHIC / VARGRAPHIC

For a further documentation on DB2 data types, see the following resources:

Tuple decoding uses the above types when storing values, it also performs on the fly conversion of the actual value when possible:

pool
  .query("SELECT an_int_column FROM exampleTable")
  .execute().onSuccess(rowSet -> {
    Row row = rowSet.iterator().next();

    // Stored as INTEGER column type and represented as java.lang.Integer
    Object value = row.getValue(0);

    // Convert to java.lang.Long
    Long longValue = row.getLong(0);
  });

Using Java enum types

You can map Java enum types to these column types:

Strings (VARCHAR, TEXT)
Numbers (SMALLINT, INTEGER, BIGINT)

client.preparedQuery("SELECT day_name FROM FINAL TABLE ( INSERT INTO days (day_name) VALUES (?), (?), (?) )")
  .execute(Tuple.of(Days.FRIDAY, Days.SATURDAY, Days.SUNDAY))
  .onComplete(ar -> {
    if (ar.succeeded()) {
      RowSet<Row> rows = ar.result();
      System.out.println("Inserted " + rows.rowCount() + " new rows");
      for (Row row : rows) {
        System.out.println("Day: " + row.get(Days.class, "day_name"));
      }
    } else {
      System.out.println("Failure: " + ar.cause().getMessage());
    }
  });
client.preparedQuery("SELECT day_num FROM FINAL TABLE ( INSERT INTO days (day_num) VALUES (?), (?), (?) )")
  .execute(Tuple.of(Days.FRIDAY.ordinal(), Days.SATURDAY.ordinal(), Days.SUNDAY.ordinal()))
  .onComplete(ar -> {
    if (ar.succeeded()) {
      RowSet<Row> rows = ar.result();
      System.out.println("Inserted " + rows.rowCount() + " new rows");
      for (Row row : rows) {
        System.out.println("Day: " + row.get(Days.class, "day_num"));
      }
    } else {
      System.out.println("Failure: " + ar.cause().getMessage());
    }
  });

The String type is matched with the Java enum’s name returned by the name() method.

Number types are matched with the Java enum’s ordinal returned by the ordinal() method and the row.get() method returns the corresponding enum’s name() value at the ordinal position of the integer value retrieved.

Collector queries

You can use Java collectors with the query API:

Collector<Row, ?, Map<Long, String>> collector = Collectors.toMap(
  row -> row.getLong("id"),
  row -> row.getString("last_name"));

// Run the query with the collector
client.query("SELECT * FROM users")
  .collecting(collector)
  .execute()
  .onComplete(ar -> {
    if (ar.succeeded()) {
      SqlResult<Map<Long, String>> result = ar.result();

      // Get the map created by the collector
      Map<Long, String> map = result.value();
      System.out.println("Got " + map);
    } else {
      System.out.println("Failure: " + ar.cause().getMessage());
    }
  });

The collector processing must not keep a reference on the Row as there is a single row used for processing the entire set.

The Java Collectors provides many interesting predefined collectors, for example you can create easily create a string directly from the row set:

Collector<Row, ?, String> collector = Collectors.mapping(
  row -> row.getString("last_name"),
  Collectors.joining(",", "(", ")")
);

// Run the query with the collector
client
  .query("SELECT * FROM users")
  .collecting(collector)
  .execute()
  .onComplete(ar -> {
    if (ar.succeeded()) {
      SqlResult<String> result = ar.result();

      // Get the string created by the collector
      String list = result.value();
      System.out.println("Got " + list);
    } else {
      System.out.println("Failure: " + ar.cause().getMessage());
    }
  });

Using SSL/TLS

To configure the client to use SSL connection, you can configure the DB2ConnectOptions like a Vert.x NetClient.

DB2ConnectOptions options = new DB2ConnectOptions()
  .setPort(50001)
  .setHost("the-host")
  .setDatabase("the-db")
  .setUser("user")
  .setPassword("secret")
  .setSsl(true)
  .setTrustStoreOptions(new JksOptions()
      .setPath("/path/to/keystore.p12")
      .setPassword("keystoreSecret"));

DB2Connection.connect(vertx, options)
  .onComplete(res -> {
    if (res.succeeded()) {
      // Connected with SSL
    } else {
      System.out.println("Could not connect " + res.cause());
    }
  });

More information can be found in the Vert.x documentation.

Using a proxy

You can also configure the client to use an HTTP/1.x CONNECT, SOCKS4a or SOCKS5 proxy.

More information can be found in the Vert.x documentation.

Advanced pool configuration

Server load balancing

You can configure the pool with a list of servers instead of a single server.

Pool pool = DB2Builder.pool()
  .with(options)
  .connectingTo(Arrays.asList(server1, server2, server3))
  .using(vertx)
  .build();

The pool uses a round-robin load balancing when a connection is created to select different servers.

this provides load balancing when the connection is created and not when the connection is borrowed from the pool.

Pool connection initialization

You can use the withConnectHandler to interact with a connection after it has been created and before it is inserted in a pool.

builder.withConnectHandler(conn -> {
  conn.query(sql).execute().onSuccess(res -> {
    // Release the connection to the pool, ready to be used by the application
    conn.close();
  });
});

Once you are done with the connection, you should simply close it to signal the pool to use it.

Dynamic connection configuration

You can configure the pool connection details using a Java supplier instead of an instance of SqlConnectOptions.

Since the supplier is asynchronous, it can be used to provide dynamic pool configuration (e.g. password rotation).

Pool pool = DB2Builder.pool()
  .with(poolOptions)
  .connectingTo(() -> {
    Future<SqlConnectOptions> connectOptions = retrieveOptions();
    return connectOptions;
  })
  .using(vertx)
  .build();

Unresolved directive in index.adoc - include::override/rxjava3.adoc[]

MySQL

Oracle