Availability Group Listeners, Client Connectivity, and Application Failover (SQL Server)
Topic Status: Some information in this topic is preview and subject to change in future releases. Preview information describes new features or changes to existing features in Microsoft SQL Server 2016 Community Technology Preview 2 (CTP2).
This topic contains information about considerations for AlwaysOn Availability Groups client connectivity and application-failover functionality.
For the majority of the common listener configurations, you can create the first availability group listener simply by using Transact-SQL statements or PowerShell cmdlets. For more information, see Related Tasks, later in this topic.
In This Topic:
You can provide client connectivity to the database of a given availability group by creating an availability group listener. An availability group listener is a virtual network name (VNN) to which clients can connect in order to access a database in a primary or secondary replica of an AlwaysOn availability group. An availability group listener enables a client to connect to an availability replica without knowing the name of the physical instance of SQL Server to which the client is connecting. The client connection string does not need to be modified to connect to the current location of the current primary replica.
An availability group listener consists of a Domain Name System (DNS) listener name, listener port designation, and one or more IP addresses. Only the TCP protocol is supported by availability group listener. The DNS name of the listener must also be unique in the domain and in NetBIOS. When you create a new availability group listener it becomes a resource in a cluster with an associated virtual network name (VNN), virtual IP (VIP), and availability group dependency. A client uses DNS to resolve the VNN into multiple IP addresses and then tries to connect to each address, until a connection request succeeds or until the connection requests time out.
If read-only routing is configured for one or more readable secondary replicas, read-intent client connections to the primary replica are redirected to a readable secondary replica. Also, if the primary replica goes offline on one instance of SQL Server, and a new primary replica comes online on another instance of SQL Server, the availability group listener enables clients to connect to the new primary replica.
For essential information about availability group listeners, see Create or Configure an Availability Group Listener (SQL Server).
In This Section:
Availability Group Listener Configuration
An availability group listener is defined by the following:
Selecting an Availability Group Listener Port
When configuring an availability group listener, you must designate a port. You can configure the default port to 1433 in order to allow for simplicity of the client connection strings. If using 1433, you do not need to designate a port number in a connection string. Also, since each availability group listener will have a separate virtual network name, each availability group listener configured on a single WSFC can be configured to reference the same default port of 1433.
You can also designate a non-standard listener port; however this means that you will also need to explicitly specify a target port in your connection string whenever connecting to the availability group listener. You will also need to open permission on the firewall for the non-standard port.
If you use the default port of 1433 for availability group listener VNNs, you will still need to ensure that no other services on the cluster node are using this port; otherwise this would cause a port conflict.
If one of the instances of SQL Server is already listening on TCP port 1433 via the instance listener and there are no other services (including additional instances of SQL Server) on the computer listening on port 1433, this will not cause a port conflict with the availability group listener. This is because the availability group listener can share the same TCP port inside the same service process. However multiple instances of SQL Server (side-by-side)should not be configured to listen on the same port.
To use an availability group listener to connect to the primary replica for read-write access, the connection string specifies the availability group listener DNS name. If an availability group primary replica changes to a new replica, existing connections that use an availability group listener's network name are disconnected. New connections to the availability group listener are then directed to the new primary replica. An example of a basic connection string for the ADO.NET provider (System.Data.SqlClient) is as follows:
You can still choose to directly reference the instance of SQL Server name of the primary or secondary replicas instead of using the availability group listener server name, however if you choose to do so you will lose the benefit of new connections being directed automatically to the current primary replica. You will also lose the benefit of read-only routing.
Read-only routing refers to the ability of SQL Server to route incoming connections to an availability group listener to a secondary replica that is configured to allow read-only workloads. An incoming connection referencing an availability group listener name can automatically be routed to a read-only replica if the following are true:
At least one secondary replica is set to read-only access, and each read-only secondary replica and the primary replica are configured to support read-only routing. For more information, see To Configure Availability Replicas for Read-Only Routing, later in this section.
The connection string references an availability group listener, and the application intent of the incoming connection is set to read-only (for example, by using the Application Intent=ReadOnly keyword in the ODBC or OLEDB connection strings or connection attributes or properties). For more information, see Read-Only Application Intent and Read-Only Routing, later in this section.
To Configure Availability Replicas for Read-Only Routing
A database administrator must configure the availability replicas as follows:
For each availability replica that you want to configure as a readable secondary replica, a database administrator must configure the following settings, which take effect only under the secondary role:
Connection access must be set to "all" or "read only".
The read-only routing URL must be specified.
For each of these replicas, a read-only routing list must be specified for the primary role. Specify one or more server names as routing targets.
Read-Only Application Intent and Read-Only Routing
The application intent connection string property expresses the client application’s request to be directed either to a read-write or read-only version of an availability group database. To use read-only routing, a client must use an application intent of read-only in the connection string when connecting to the availability group listener. Without the read-only application intent, connections to the availability group listener are directed to the database on the primary replica.
The application intent attribute is stored in the client’s session during login and the instance of SQL Server will then process this intent and determine what to do according to the configuration of the availability group and the current read-write state of the target database in the secondary replica.
An example of a connection string for the ADO.NET provider (System.Data.SqlClient) that designates read-only application intent is as follows:
In this connection string example, the client is attempting to connect to an availability group listener named AGListener on port 1433 (you may also omit the port if the availability group listener is listening on 1433). The connection string has the ApplicationIntent property set to ReadOnly, making this a read-intent connection string. Without this setting, the server would not have attempted a read-only routing of the connection.
The primary database of the availability group processes the incoming read-only routing request and attempts to locate an online, read-only replica that is joined to the primary replica and is configured for read-only routing. The client receives back connection information from the primary replica server and connects to the identified read-only replica.
Note that the application intent can be sent from a client driver to a down-level instance of SQL Server. In this case, application intent of read-only is ignored and the connection proceeds as normal.
You can bypass read-only routing by not setting the application intent connection property to ReadOnly (when not designated, the default is ReadWrite during login) or by connecting directly to the primary replica instance of SQL Server instead of using the availability group listener name. Read-only routing will also not occur if you connect directly to a read-only replica.
While availability group listeners enable support for failover redirection and read-only routing, client connections are not required to use them. A client connection can also directly reference the instance of SQL Server instead of connecting to the availability group listener.
To the instance of SQL Server it is irrelevant whether a connection logs in using the availability group listener or using another instance endpoint. The instance of SQL Server will verify the state of the targeted database and either allow or disallow connectivity based on the configuration of the availability group and the current state of the database on the instance. For example, if a client application connects directly to a instance of SQL Server port and connects to a target database hosted in an availability group, and the target database is in primary state and online, then connectivity will succeed. If the target database is offline or in a transitional state, connectivity to the database will fail.
Alternatively, while migrating from database mirroring to AlwaysOn Availability Groups, applications can specify the database mirroring connection string as long as only one secondary replica exists and it disallows user connections. For more information, see Using Database-Mirroring Connection Strings with Availability Groups, later in this section.
Using Database-Mirroring Connection Strings with Availability Groups
If an availability group possesses only one secondary replica and is not configured to allow read-access to the secondary replica, clients can connect to the primary replica by using a database mirroring connection string. This approach can be useful while migrating an existing application from database mirroring to an availability group, as long as you limit the availability group to two availability replicas (a primary replica and one secondary replica). If you add additional secondary replicas, you will need to create an availability group listener for the availability group and update your applications to use the availability group listener DNS name.
When using database mirroring connection strings, the client can use either SQL Server Native Client or .NET Framework Data Provider for SQL Server. The connection string provided by a client must minimally supply the name of one server instance, the initial partner name, to identify the server instance that initially hosts the availability replica to which you intend to connect. Optionally, the connection string can also supply the name of another server instance, the failover partner name, to identify the server instance that initially hosts the secondary replica as the failover partner name.
For more information about database mirroring connection strings, see Connect Clients to a Database Mirroring Session (SQL Server).
When an availability group failover occurs, existing persistent connections to the availability group are terminated and the client must establish a new connection in order to continue working with the same primary database or read-only secondary database. While a failover is occurring on the server side, connectivity to the availability group may fail, forcing the client application to retry connecting until the primary is brought fully back online.
If the availability group comes back online during a client application’s connection attempt but before the connect timeout period, the client driver may successfully connect during one of its internal retry attempts and no error will be surfaced to the application in this case.
If you are using client libraries that support the MultiSubnetFailover connection option in the connection string, you can optimize availability group failover to a different subnet by setting MultiSubnetFailover to “True” or "Yes", depending on the syntax of the provider you are using.
We recommend this setting for both single and multi-subnet connections to availability groups listeners and to SQL Server Failover Cluster Instance names. Enabling this option adds additional optimizations, even for single-subnet scenarios.
The MultiSubnetFailover connection option only works with the TCP network protocol and is only supported when connecting to an availability group listener and for any virtual network name connecting to SQL Server 2016.
An example of a for the ADO.NET provider (System.Data.SqlClient) connection string that enables multi-subnet failover is as follows:
The MultiSubnetFailover connection option should be set to True even if the availability group only spans a single subnet. This allows you to preconfigure new clients to support future spanning of subnets without any need for future client connection string changes and also optimizes failover performance for single subnet failovers. While the MultiSubnetFailover connection option is not required, it does provide the benefit of a faster subnet failover. This is because the client driver will attempt to open up a TCP socket for each IP address in parallel associated with the availability group. The client driver will wait for the first IP to respond with success and once it does, will then use it for the connection.
When connecting to an availability group listener, if the participating instances of SQL Server use SSL certificates in conjunction with session encryption, the connecting client driver will need to support the Subject Alternate Name in the SSL certificate in order to force encryption. SQL Server driver support for certificate Subject Alternative Name is planned for ADO.NET (SqlClient), Microsoft JDBC and SQL Native Client (SNAC).
A X.509 certificate must be configured for each participating server node in the failover cluster with a list of all availability group listeners set in the Subject Alternate Name of the certificate.
For example, if the WSFC has three availability group listeners with the names AG1_listener.Adventure-Works.com, AG2_listener.Adventure-Works.com, and AG3_listener.Adventure-Works.com, the Subject Alternative Name for the certificate should be set as follows:
CN = ServerFQDN SAN = ServerFQDN,AG1_listener.Adventure-Works.com, AG2_listener.Adventure-Works.com, AG3_listener.Adventure-Works.com
A Server Principal Name (SPN) must be configured in Active Directory by a domain administrator for each availability group listener name in order to enable Kerberos for the client connection to the availability group listener. When registering the SPN, you must use the service account of the server instance that hosts the availability replica . For the SPN to work across all replicas, the same service account must be used for all instances in the WSFC cluster that hosts the availability group.
Use the setspn Windows command line tool to configure the SPN. For example to configure an SPN for an availability group named AG1listener.Adventure-Works.com hosted on a set of instances of SQL Server all configured to run under the domain account corp/svclogin2:
setspn -A MSSQLSvc/AG1listener.Adventure-Works.com:1433 corp/svclogin2
For more information about manual registration of a SPN for SQL Server, see Register a Service Principal Name for Kerberos Connections.