Deploying Cisco Nexus Dashboard Data Broker Software in Clusters

Beginning with Release 3.10.1, Cisco Nexus Data Broker (NDB) has been renamed to Cisco Nexus Dashboard Data Broker. However, some instances of NDB are present in this document, to correspond with the GUI, and installation folder structure. References of NDB/ Nexus Data Broker/ Nexus Dashboard Data Broker, can be used interchangeably.

This chapter contains the following details:

Installing a Cisco Nexus Dashboard Data Broker Cluster

Use this procedure to install a Cisco Nexus Dashboard Data Broker (NDDB) cluster.

Before you begin

Prerequisites:

  • Cisco Nexus Dashboard Data Broker (NDDB) supports 3-node clusters.

  • All IP addresses must be reachable and capable of communicating with each other.

  • All switches in the cluster must connect to all the controllers.

  • All controllers must have the same HA clustering configuration information in the config.ini files.

  • All the NDDB instances should be of the same NDDB version to form the cluster.

  • If using cluster passwords, all controllers must have the same password configured in the ndbjgroups.xml file. See Password Protecting for HA Clusters section in the Cisco Nexus Dashboard Data Broker Configuration Guide.


Note

All the NDDB instances to form the cluster should be of the same NDDB version.

Procedure

Step 1

In a web browser, navigate to www.cisco.com.

Step 2

Scroll down and click Downloads.

Step 3

In the Select a Product search box, enter Nexus Dashboard Data Broker and you are automatically taken to the latest release Software Download screen.

The file information for Release 3.10.4 is displayed: Cisco Nexus Data Broker Software Application: ndb1000-sw-app-k9-3.10.4.zip

Step 4

Download the Cisco Nexus Data Broker application bundle. If prompted, enter your Cisco.com username and password to login.

Step 5

Create a directory in your Linux machine where you plan to install the Data Broker.

For example, in your Home directory, create CiscoNDB.

Step 6

Copy the Cisco Nexus Dashboard Data Broker zip file to the created NDDB directory.

Step 7

Unzip the Data Broker zip file.

The Data Broker software is installed in a directory called ndb. The directory contains the following:

  • runndb.sh file—file to launch NDDB.

  • version.properties file—NDDB build version.

  • configuration directory—contains the NDDB initialization files. This directory also contains the startup subdirectory where configurations are saved.

  • bin directory—contains the NDDB file that has the common CLI.

  • etc directory—contains profile information.

  • lib directory—contains NDDB Java libraries.

  • logs directory—contains NDDB logs.

    Note

     

    The logs directory is created after the NDDB application is started.

  • plugins directory—The directory that contains the NDDB plugins.

  • work directory—webserver working directory.

Step 8

Navigate to the ndb/configuration directory that was created when you installed the software.

Step 9

Use any text editor to open the config.ini file and locate the following text: 


# HA Clustering configuration (semi-colon-separated IP addresses of all controllers that are part of the cluster.)
# supernodes=<ip1>;<ip2>;<ip3>

If a standby node is available: 
#supernodes=<ip1>;<ip2>;<ip3>;<ip4>-standby

Step 10

Uncomment the line which consists of supernodes and replace <ip*> with NDDB server IPs.


IPv4 example:
# HA Clustering configuration (semi-colon-separated IP addresses of all controllers that are part of the cluster.)
supernodes=10.1.1.1;10.2.1.1;10.3.1.1

IPv6 example:
# HA Clustering configuration (semi-colon-separated IP addresses of all controllers that are part of the cluster.)
supernodes=2001:22:11::1;2001:33::44::1;2001:55:66::1

Step 11

Save the file and exit the editor.

Step 12

Repeat steps 5 to 11 in all the Linux machines where the NDDB is installed.

Step 13

Start the primary NDDB server using the ./runndb.sh -start command.

Step 14

After the GUI of the primary NDDB server is up, start the other NDDB servers using the ./runndb.sh -start command.

After the primary is up, await the confirmation messsage displayed on the GUI before starting the members of the cluster. The displayed message reads, Primary is ready, bring up the members.

Upgrading a Cisco Nexus Dashboard Data Broker Cluster

Before you begin

Prerequisites:

  • Cisco Nexus Dashboard Data Broker (NDDB) supports 3-node clusters.

  • All IP addresses must be reachable and capable of communicating with each other.

  • All switches in the cluster must connect to all the controllers.

  • All controllers must have the same HA clustering configuration information in the config.ini files.

  • All the Nexus Dashboard Data Broker instances should be of the same Nexus Dashboard Data Broker version to form the cluster.

  • If using cluster passwords, all controllers must have the same password configured in the ndbjgroups.xml file. See Password Protecting for HA Clusters section in the Cisco Nexus Data Broker Configuration Guide.


Note

All the Nexus Dashboard Data Broker instances to form the cluster should be of the same Cisco Nexus Dashboard Data Broker version.

Procedure

Step 1

Login to the Cisco Nexus Dashboard Data Broker primary server.

Step 2

Navigate to Administration > Backup/ Restore.

Step 3

Click Backup Locally to download the configuration file.

Step 4

Stop all Cisco Nexus Dashboard Data Broker instances using the runndb.xh -stop command.

Step 5

If TLS certification is enabled between NDDB server and NDDB Devices, take backup of the tlsTrustStore and tlsKeyStore files from /ndb/configuration.

Step 6

Perform the previous step on all the NDDB cluster instances.

Step 7

In a web browser, navigate to www.cisco.com.

Step 8

Navigate to Support > Products > Downloads.

Step 9

In the Find Products and Downloads search box, enter “Nexus Data Broker” and click on ‘Downloads’ from search response list.

The file information for Release 3.10.4 is displayed: Cisco Nexus Dashboard Data Broker Software Application: ndb1000-sw-app-k9-3.10.4.zip

Step 10

Download the Cisco Nexus Dashboard Data Broker application bundle. When prompted, enter your Cisco.com username and password to login.

Step 11

Create a directory in your Linux machine where you plan to install Cisco Nexus Dashboard Data Broker.

For example, in your Home directory, create Cisco Nexus Dashboard Data Broker.

Step 12

Copy the Cisco Nexus Dashboard Data Broker zip file to the directory that you have created.

Step 13

Unzip the Cisco Nexus Dashboard Data Broker zip file.

The Cisco Nexus Dashboard Data Broker software is installed in a directory called ndb. The directory contains the following:

  • runndb.sh file—file to launch NDDB.

  • version.properties file—NDDB build version.

  • configuration directory—contains the NDDB initialization files. This directory also contains the startup subdirectory where configurations are saved.

  • bin directory—contains the NDDB file that has the NDDB common CLI.

  • etc directory—contains profile information.

  • lib directory—contains NDDB Java libraries.

  • logs directory—contains NDDB logs.

    Note

     

    The logs directory is created after the NDDB application is started.

  • plugins directory—The directory that contains the NDDB plugins.

  • work directory—webserver working directory.

Step 14

Navigate to the ndb/configuration directory that was created when you installed the software.

Step 15

Use any text editor to open the config.ini file and locate the following text:

Step 16

Locate th following text: 

# HA Clustering configuration (semi-colon-separated IP addresses of all controllers that are part of the cluster.)
# supernodes=<ip1>;<ip2>;<ip3>

If a standby node is available: 
#supernodes=<ip1>;<ip2>;<ip3>;<ip4>-standby

Step 17

Uncomment the line which consists of supernodes and replace <ip*> with NDDB Server IPs.


IPv4 example:
# HA Clustering configuration (semi-colon-separated IP addresses of all controllers that are part of the cluster.)
supernodes=10.1.1.1;10.2.1.1;10.3.1.1

IPv6 example:
# HA Clustering configuration (semi-colon-separated IP addresses of all controllers that are part of the cluster.)
supernodes=2001:22:11::1;2001:33::44::1;2001:55:66::1

Step 18

Save the file and exit the editor.

Step 19

Repeat the steps 7 to 18 in all the Linux machines where the NDDB is installed.

Step 20

Start the Primary NDDB server using the ./runndb.sh -start command.

Step 21

After the GUI of the Primary NDDB Server is up, start the other NDDB servers using the ./runndb.sh -start command.

Step 22

Login to the Primary Server NDDB GUI.

Step 23

Navigate to Administration > Backup/Restore > Actions > Restore Locally and upload the configuration you had earlier downloaded.

Step 24

Stop all instances of NDDB in the cluster using the ./runndb.sh -stop command.

Step 25

If TLS certification is enabled between NDDB server and NDDB switches, copy the tls TrustStore and tlSKeyStore files to ndb/configuration taken from the step 5 for all NDDB instances.

Step 26

Start the primary NDDB server using the ./runndb.sh -start command.

If TLS certification is enabled, use below commands to start NDDB servers.

./runndb.sh -tls -tlskeystore ./configuration/tlsKeyStore -tlstruststore ./configuration/tlsTrustStore
bin/ndb config-keystore-passwords --user <NDB_username> --password <NDB_password> --url https://<Cluster_NDB_IP>:8443 --verbose --prompt --keystore-password <keystore-password> --truststore-password <truststore-password>

After the primary is up, await the confirmation messsage displayed on the GUI before starting the members of the cluster. The displayed message reads, Primary is ready, bring up the members.

Step 27

Start the member NDDB server(s) using the ./runndb.sh -start command.

If TLS certification is enabled, use below commands to start NDDB servers.

./runndb.sh -tls -tlskeystore ./configuration/tlsKeyStore -tlstruststore ./configuration/tlsTrustStore
bin/ndb config-keystore-passwords --user <NDB_username> --password <NDB_password> --url https://<Cluster_NDB_IP>:8443 --verbose --prompt --keystore-password <keystore-password> --truststore-password <truststore-password>

Upgrading the Application Software with TLS-enabled for HA-clustered Controller

Use this procedure for upgrading the Nexus Dashboard Data Borker (NDDB) application software in centralized mode, using the GUI, when the TLS certification is enabled in a HA-clustered controller.

Procedure

Step 1

Log in to the existing NDDB GUI instance using https://server IP:8443.

Step 2

Navigate to the Administration > Backup/ Restore tab.

Step 3

Click Backup now Locally to download the configuration as a zip file.

Step 4

Stop the current NDDB instance(s) using the ./runndb.sh -stop command.

Step 5

After the NDDB instances are stopped, navigate to the /ndb/configuration folder, and copy the tlsTrustStore and tlsKeyStore files to local/common folder.

Step 6

Download the NDDB 3.10.4 software from the standard Cisco.com Downloads page and configure the cluster mode using the "supernodes" configuration in the config.ini file and start the new NDDB 3.10.4 cluster using the ./runndb.sh -start command on all the controllers.

Step 7

On the primary controller, navigate to the Administration > Backup/ Restore tab.

Step 8

Click Restore Locally to upload the configuration file which you had earlier downloaded (Step 3, above).

After the configuration is uploaded successfully, you will see a success message on the GUI.

Step 9

Copy the tlsTrustStore and tlsKeyStore files to NDB 3.10.4 /ndb/configuration folder (which was copied to the local/common folder in step 5).

Step 10

Stop the NDDB 3.10.4 instances on all the controllers using the ./runndb.sh -stop command.

Step 11

Start the NDDB instance on the primary controller using the ./runndb.sh -tls -tlskeystore ./configuration/tlsKeyStore -tlstruststore ./configuration/tlsTrustStore command.

Wait for a few minutes; a ready message is displayed.

Step 12

Start the NDDB instance on the other controllers of the cluster using the ./runndb.sh -tls -tlskeystore ./configuration/tlsKeyStore -tlstruststore ./configuration/tlsTrustStore command.

To start the instances on the other controllers, you need not mention the whole command again (with the TLS keywords). You can use the standard ./runndb.sh -start .