test

Sun Java System Message Queue 4.1 Administration Guide

Managing Clusters

The following sections describe how to perform various administrative management tasks for conventional and high-availability clusters, respectively.

Managing Conventional Clusters

The procedures in this section show how to perform the following tasks for a conventional cluster:

Clustering Conventional Brokers

There are two general methods of connecting conventional brokers into a cluster: from the command line (using the -cluster option) or by setting the imq.cluster.brokerlist property in the cluster configuration file. Whichever method you use, each broker that you start attempts to connect to the other brokers in the cluster every five seconds; the connection will succeed once the master broker is started up (if one is configured). If a broker in the cluster starts before the master broker, it will remain in a suspended state, rejecting client connections, until the master broker starts; the suspended broker then will automatically become fully functional. It is therefore a good idea to start the master broker first and then the others, after the master broker has completed its startup.

Note –

Whichever clustering method you use, you must make sure that no broker in the cluster is given an address that resolves to the network loopback IP address (127.0.0.1). Any broker configured with this address will be unable to connect to other brokers in the cluster.

In particular, some Linux installers automatically set the localhost entry to the network loopback address. On such systems, you must modify the system IP address so that all brokers in the cluster can be addressed properly: For each Linux system participating in the cluster, check the /etc/hosts file as part of cluster setup. If the system uses a static IP address, edit the /etc/hosts file to specify the correct address for localhost. If the address is registered with Domain Name Service (DNS), edit the file /etc/nsswitch.conf to change the order of the entries so that DNS lookup is performed before consulting the local hosts file. The line in /etc/nsswitch.conf should read as follows: 

   hosts: dns files
Note –

If you are clustering a Message Queue 4.1 broker together with those from earlier versions of Message Queue, you must set the value of the 4.1 broker’s imq.autocreate.queue.maxNumActiveConsumers property to 1. Otherwise the brokers will not be able to establish a cluster connection.

ProcedureTo Cluster Conventional Brokers from the Command Line

  1. If you are using a master broker, start it with the imqbrokerd command, using the -cluster option to specify the complete list of brokers to be included in the cluster.

    For example, the following command starts the broker as part of a cluster consisting of the brokers running at the default port (7676) on host1, at port 5000 on host2, and at port 9876 on the default host (localhost): 

       imqbrokerd  -cluster host1,host2:5000,:9876

  2. Once the master broker (if any) is running, start each of the other brokers in the cluster with the imqbrokerd command, using the same list of brokers with the -cluster option that you used for the master broker.

    The value specified for the -cluster option must be the same for all brokers in the cluster.

ProcedureTo Cluster Conventional Brokers Using a Cluster Configuration File

An alternative method, better suited for production systems, is to use a cluster configuration file to specify the composition of the cluster:

  1. Create a cluster configuration file that uses the imq.cluster.brokerlist property to specify the list of brokers to be connected.

    If you are using a master broker, identify it with the imq.cluster.masterbroker property in the configuration file.

  2. For each broker in the cluster, set the imq.cluster.url property in the broker’s instance configuration file to point to the cluster configuration file.

  3. Use the imqbrokerd command to start each broker.

    If there is a master broker, start it first, then the others after it has completed its startup.

ProcedureTo Establish Secure Connections Between Brokers

If you want secure, encrypted message delivery between brokers in a cluster, configure the cluster connection service to use an SSL-based transport protocol:

  1. For each broker in the cluster, set up SSL-based connection services, as described in Message Encryption.

  2. Set each broker’s imq.cluster.transport property to ssl, either in the cluster configuration file or individually for each broker.

Adding Brokers to a Conventional Cluster

The procedure for adding a new broker to a conventional cluster depends on whether the cluster uses a cluster configuration file.

ProcedureTo Add a New Broker to a Conventional Cluster Using a Cluster Configuration File

  1. Add the new broker to the imq.cluster.brokerlist property in the cluster configuration file.

  2. Issue the following command to any broker in the cluster: 

       imqcmd reload cls

    This forces each broker to reload the cluster configuration, ensuring that all persistent information for brokers in the cluster is up to date. Note that it is not necessary to issue this command to every broker in the cluster; executing it for any one broker will cause all of them to reload the cluster configuration.

  3. (Optional) Set the value of the imq.cluster.url property in the new broker’s instance configuration file (config.properties) to point to the cluster configuration file.

  4. Start the new broker.

    If you did not perform step 3, use the -D option on the imqbrokerd command line to set the value of imq.cluster.url to the location of the cluster configuration file.

ProcedureTo Add a New Broker to a Conventional Cluster Without a Cluster Configuration File

  1. (Optional) Set the values of the following properties in the new broker’s instance configuration file (config.properties) :

      imq.cluster.brokerlist


      imq.cluster.masterbroker (if necessary)


      imq.cluster.transport (if you are using a secure cluster connection service)


  2. Start the new broker.

    If you did not perform step 1, use the -D option on the imqbrokerd command line to set the property values listed there.

Removing Brokers From a Conventional Cluster

The method you use to remove a broker from a conventional cluster depends on whether you originally created the cluster from the command line or by means of a central cluster configuration file.

ProcedureTo Remove a Broker From a Conventional Cluster Using the Command Line

If you used the imqbrokerd command from the command line to connect the brokers into a cluster, you must stop each of the brokers and then restart them, specifying the new set of cluster members on the command line:

  1. Stop each broker in the cluster, using the imqcmd command.

  2. Restart the brokers that will remain in the cluster, using the imqbrokerd command’s -cluster option to specify only those remaining brokers.

    For example, suppose you originally created a cluster consisting of brokers A, B, and C by starting each of the three with the command 

       imqbrokerd  -cluster A,B,C

    To remove broker A from the cluster, restart brokers B and C with the command 

       imqbrokerd  -cluster B,C

ProcedureTo Remove a Broker From a Conventional Cluster Using a Cluster Configuration File

If you originally created a cluster by specifying its member brokers with the imq.cluster.brokerlist property in a central cluster configuration file, it isn’t necessary to stop the brokers in order to remove one of them. Instead, you can simply edit the configuration file to exclude the broker you want to remove, force the remaining cluster members to reload the cluster configuration, and reconfigure the excluded broker so that it no longer points to the same cluster configuration file:

  1. Edit the cluster configuration file to remove the excluded broker from the list specified for the imq.cluster.brokerlist property.

  2. Issue the following command to each broker remaining in the cluster: 

       imqcmd reload cls

    This forces the brokers to reload the cluster configuration.

  3. Stop the broker you’re removing from the cluster.

  4. Edit that broker’s instance configuration file (config.properties), removing or specifying a different value for its imq.cluster.url property.

Managing the Configuration Change Record

As noted earlier, a conventional cluster can optionally have one master broker, which maintains a configuration change record to keep track of any changes in the cluster’s persistent state. The master broker is identified by the imq.cluster.masterbroker configuration property, either in the cluster configuration file or in the instance configuration files of the individual brokers.

Because of the important information that the configuration change record contains, it is important to back it up regularly so that it can be restored in case of failure. Although restoring from a backup will lose any changes in the cluster’s persistent state that have occurred since the backup was made, frequent backups can minimize this potential loss of information. The backup and restore operations also have the positive effect of compressing and optimizing the change history contained in the configuration change record, which can grow significantly over time.

ProcedureTo Back Up the Configuration Change Record

  1. Use the -backup option of the imqbrokerd command, specifying the name of the backup file.

    For example: 

       imqbrokerd  -backup mybackuplog

ProcedureTo Restore the Configuration Change Record

  1. Shut down all brokers in the cluster.

  2. Restore the master broker’s configuration change record from the backup file.

    The command is 

       imqbrokerd  -restore mybackuplog

  3. If you assign a new name or port number to the master broker, update the imq.cluster.brokerlist and imq.cluster.masterbroker properties accordingly in the cluster configuration file.

  4. Restart all brokers in the cluster.

Managing High-Availability Clusters

This section presents step-by-step procedures for performing a variety of administrative tasks for a high-availability cluster:

Clustering High-Availability Brokers

Because high-availability clusters are self-configuring, there is no need to explicitly specify the list of brokers to be included in the cluster. Instead, all that is needed is to set each broker’s configuration properties appropriately and then start the broker; as long as its properties are set properly, it will automatically be incorporated into the cluster. Table 8–1 shows the required settings. In addition, there may be vendor-specific settings required for a particular vendor’s database; Table 8–2 and Table 8–3 show these vendor-specific settings for Sun’s own HADB and MySQL from MySQLAB, respectively.

Table 8–1 Required Configuration Properties for HA Clusters

Property 

Required Value 

Description 

imq.cluster.ha

true

Broker is part of an HA cluster

imq.cluster.clusterid

 

Cluster identifier 

Must be the same for all brokers in the cluster.  

imq.brokerid

 

Broker identifier 

Must be different for each broker in the cluster 

imq.persist.store

jdbc

Model for persistent data storage 

Only JDBC-based persistence is supported for HA data stores.

imq.persist.jdbc.dbVendor

 

Database vendor for HA persistent store:

    hadb: HADB (Sun Microsystems, Inc.)


    derby: Java DB (Derby, Apache Software Foundation)


    oracle: Oracle Real Application Cluster (Oracle Corporation)


    mysql: MySQL (MySQL AB)

Table 8–2 Vendor-Specific Configuration Properties for HADB Database

Property 

Description 

imq.persist.jdbc.hadb.user

User name for opening database connection 

imq.persist.jdbc.hadb.password

Password for opening database connection 

imq.persist.hadb.property.serverList

JDBC URL of database

Use the command 

   hadbm get JdbcURL

to get the URL; remove the prefix 

   jdbc:sun:hadb

and use  

   host:port,host:port...

for the property value.  

Table 8–3 Vendor-Specific Configuration Properties for MySQL Database

Property 

Description 

imq.persist.jdbc.mysql.user

User name for opening database connection 

imq.persist.jdbc.mysql.password

Password for opening database connection 

imq.persist.jdbc.mysql.property.url

JDBC URL for opening database

The property values can be set separately in each broker’s instance configuration file, or they can be specified in a cluster configuration file that all the brokers share. The procedures are as follows:

ProcedureTo Cluster HA Brokers Using Instance Configuration Files

  1. For each broker in the cluster:

    1. Start the broker with the imqbrokerd command.

      The first time a broker instance is run, an instance configuration file (config.properties) is automatically created.

    2. Shut down the broker.

      Use the imqcmd shutdown bkr command.

    3. Edit the instance configuration file to specify the broker’s HA-related configuration properties.

      Table 8–1 shows the required property values.

    4. Specify any additional, vendor-specific properties that may be needed.

      Table 8–2 and Table 8–3 show the required properties for HADB and MySQL databases, respectively.

  2. Place a copy of, or a symbolic link to, your JDBC driver’s .jar file in the appropriate location, depending on your platform:

      Solaris: /usr/share/lib/imq/ext/


      Linux: /opt/sun/mq/share/lib/


      Windows: IMQ_VARHOME\lib\ext


  3. Create the database schema needed for Message Queue persistence.

    Use the imqdbmgr create tbl command; see Database Manager Utility.

  4. Restart each broker with the imqbrokerd command.

    The brokers will automatically register themselves into the cluster on startup.

ProcedureTo Cluster HA Brokers Using a Cluster Configuration File

An alternative method, better suited for production systems, is to use a cluster configuration file to specify the composition of the cluster:

  1. Create a cluster configuration file specifying the cluster’s HA-related configuration properties.

    Table 8–1 shows the required property values. However, do not include the imq.brokerid property in the cluster configuration file; this must be specified separately for each individual broker in the cluster.

  2. Specify any additional, vendor-specific properties that may be needed.

    Table 8–2 and Table 8–3 show the required properties for HADB and MySQL databases, respectively.

  3. For each broker in the cluster:

    1. Start the broker with the imqbrokerd command.

      The first time a broker instance is run, an instance configuration file (config.properties) is automatically created.

    2. Shut down the broker.

      Use the imqcmd shutdown bkr command.

    3. Edit the instance configuration file to specify the location of the cluster configuration file.

      In the broker’s instance configuration file, set the imq.cluster.url property to point to the location of the cluster configuration file you created in step 1.

    4. Specify the broker identifier.

      Set the imq.brokerid property in the instance configuration file to the broker’s unique broker identifier. This value must be different for each broker.

  4. Place a copy of, or a symbolic link to, your JDBC driver’s .jar file in the appropriate location, depending on your platform:

      Solaris: /usr/share/lib/imq/ext/


      Linux: /opt/sun/mq/share/lib/


      Windows: IMQ_VARHOME\lib\ext


  5. Create the database schema needed for Message Queue persistence.

    Use the imqdbmgr create tbl command; see Database Manager Utility.

  6. Restart each broker with the imqbrokerd command.

    The brokers will automatically register themselves into the cluster on startup.

Adding and Removing Brokers in a High-Availability Cluster

Because HA clusters are self-configuring, the procedures for adding and removing brokers are simpler than for a conventional cluster:

ProcedureTo Add a New Broker to an HA Cluster

  1. Set the new broker’s HA-related properties, as described in the preceding section.

    You can do this either by specifying the individual properties in the broker’s instance configuration file (config.properties) or, if there is a cluster configuration file, by setting the broker’s imq.cluster.url property to point to it.

  2. Start the new broker with the imqbrokerd command.

    The broker will automatically register itself into the cluster on startup.

ProcedureTo Remove a Broker from an HA Cluster

  1. Make sure the broker is not running.

    If necessary, use the command 

       imqcmd shutdown bkr

    to shut down the broker.

  2. Remove the broker from the cluster with the command 

       imqdbmgr remove bkr

Preventing or Forcing Takeover of a Broker

Although the takeover of a failed broker’s persistent data by another broker in an HA cluster is normally automatic, there may be times when you want to prevent such a takeover from occurring. To suppress automatic takeover when shutting down a broker, use the -nofailover option to the imqcmd shutdown bkr subcommand: 

   imqcmd shutdown bkr  -nofailover  -b hostName:portNumber

where hostName and portNumber are the host name and port number of the broker to be shut down.

Conversely, you may sometimes need to force a broker takeover to occur manually. (This might be necessary, for instance, if an automatic takeover broker were to fail before completing the takeover process.) In such cases, you can initiate a takeover manually from the command line: first shut down the broker to be taken over with the -nofailover option, as shown above, then issue the command 

   imqcmd takeover bkr  -n brokerID

where brokerID is the broker identifier of the broker to be taken over. If the specified broker appears to be running, the Command utility will display a confirmation message: 

   The broker associated with brokerID last accessed the database # seconds ago. 
   Do you want to take over for this broker?

You can suppress this message, and force the takeover to occur unconditionally, by using the -f option to the imqcmd takeover bkr command: 

   imqcmd takeover bkr  -f  -n brokerID
Note –

The imqcmd takeover bkr subcommand is intended only for use in failed-takeover situations. You should use it only as a last resort, and not as a general way of forcibly taking over a running broker.

You may also find it useful to quiesce a broker before shutting it down, causing it to refuse any new client connections while continuing to service old ones. This allows the broker’s operations to wind down gradually without triggering a takeover by another broker, for instance in preparation for shutting it down administratively for upgrade or similar purposes; see Quiescing a Broker for more information.

Managing the HA Data Store

When converting to high-availability operation, you can use the Message Queue Database Manager utility (imqdbmgr) subcommand 

   imqdbmgr upgrade hastore

to convert an existing standalone HADB persistent data store to a shared HADB store. You can use this command in the following cases:

Because this command only supports conversion of HADB stores, it cannot be used to convert file-based stores or other JDBC-based stores to a shared HADB store. If you were previously running a 3.x version of Message Queue, you must create an HADB store and then manually migrate your data to that store in order to use the high availability feature.

For durability and reliability, it is a good idea to back up a high-availability cluster’s shared persistent data store periodically to backup files. This creates a snapshot of the data store that you can then use to restore the data in case of catastrophic failure. The command for backing up the data store is 

   imqdbmgr backup  -dir backupDir

where backupDir is the path to the directory in which to place the backup files. To restore the data store from these files, use the command 

   imqdbmgr restore  -restore backupDir