Setting Up a Georedundant Non-CMP Cluster

Note: Georedundancy requires that you configure the CMP system with Manage Geo-Redundant enabled. See the CMP Wireless User's Guide for more information.
Before defining a cluster, ensure the following conditions are met:
  • The server software is installed on all servers in the cluster.
  • The servers have been configured with network time protocol (NTP), domain name server (DNS), IP Routing, and OAM IP addresses.
  • The server IP connection is active.
  • The server application is running on at least one server.
A georedundant non-CMP cluster is one of the following server types:
  • MPE
  • MRA
  • Mediation server
Note: The list of available server types depends on the CMP modes configured. See CMP Modes for more information.
Note: If your system is not set up for georedundancy, see Setting Up a Non-CMP Cluster.

To set up a georedundant non-CMP cluster:

  1. From the Platform Setting section of the navigation pane, select Topology Settings.

    The Cluster Configuration page opens; the initial group is All Clusters.

  2. From the work area, select Add MPE/MRA/Mediation Cluster.

    The Topology Configuration page opens.

  3. In the Cluster Settings section of the page:
    1. (Required) Enter the Name for the site.

      The name can only contain the characters A through Z, a through z, 0 through 9, period (.), hyphen (-), and underline (_). The maximum length is 35 characters.

    2. Select an Appl Type.

      Note: The list of available cluster types to add to the topology depends on the CMP modes configured.

    3. Select the Site Preference.

      Available options are Normal (default) or Reverse. See Georedundant Site Preferences for more information.

    4. Select the type of DSCP Marking (Differentiated Services Code Point) for replication traffic.

      The valid code points are:

      Assured Forwarding
      AF11, AF12, AF13, AF21, AF22, AF23, AF31, AF32, AF33, AF41, AF42, AF43
      Class Selector
      CS1, CS2, CS3, CS4, CS5, CS6, CS7
      Expedited Forwarding
      EF
      Default (for no marking)
      PHB (None)

      For information on DSCP marking, see Diameter Signaling

    5. Select the Replication Stream Count.

      This is the number of redundant TCP/IP socket connections (streams) to carry replication traffic between sites. Up to 8 streams can be configured. The default value is 1 stream.

    6. Select a Replication & Heartbeat network to carry inter-site replication and heartbeat traffic.

      This field only is visible if the system supports georedundancy:

      • None (default)
      • OAM
      • SIG-A
      • SIG-B
      • SIG-C
      • REP
      Note: When saving a configuration using SIG-C, a confirmation appears. Click OK. The RMS option for HW Type is removed until all configured Signaling C VIPs or SIG-C interfaces in static IP are removed.
      A warning icon () indicates that you cannot select a network until you define a static IP address on all servers of both sites.

    7. Select a Backup Heartbeat network to carry inter-site backup heartbeat traffic.

      Note: When saving a configuration using SIG-C, a confirmation message appears. Click OK. The RMS option for HW Type is removed until all configured Signaling C VIPs or SIG-C interfaces in static IP are removed.
      Note: This field only is visible if the system supports georedundancy.

      Available options are:

      • None (default)
      • OAM
      • SIG-A
      • SIG-B
      • SIG-C
      • REP
      A warning icon () indicates that you cannot select a network until you define a static IP address on all servers of both sites.

  4. In the Primary Site Settings section of the page:
    1. Select the Site Name from the list.

      Select Unspecified (default) or the Name of a previously defined site. You can assign multiple clusters to the same site.
      Note: If you select Unspecified, you create a non-georedundant site and cannot add a secondary site.

    2. To import the HW Type and VLAN ID settings from the from the selected site, select Use Site Configuration.

      When Use Site Configuration is selected, the HW Type and VLAN ID settings become read only.

      To edit the fields, uncheck the Use Site Configuration.

      Note: If Unspecified is selected for the site name, the Use Site Configuration option becomes unavailable.

    3. Select the HW Type from the list.

      Available options are:
      • C-Class (default) – HP ProLiant BL460 Gen8 server
      • C-Class (Segregated Traffic) (a configuration where Signaling and other networks are separated onto physically separate equipment) – HP ProLiant BL460 G8
      • Oracle RMS – Oracle Server X5-2
      • RMS (rack-mounted server) – HP ProLiant DL360 Gen8 or HP ProLiant DL380 Gen8 server
      • VM (virtual machine)
      • VM(Automated) (VM managed by NF Agent)

        See Setting Up a VM (Automated) Non-CMP Cluster for details on adding a VM (Automated) cluster.

    4. (Required) To enter up to two OAM VIP (one IPv4 and one IPv6) addresses, click Add New VIP.

      The New OAM VIP dialog box appears.
      1. Enter the OAM VIP address and the Mask.
        This is the IP address the CMP server uses to communicate with a Policy Management cluster.
        Note: Enter the IPv4 address in standard dot format and its subnet mask in CIDR notation from 0 to 32, or the IPv6 address in standard 8-part colon-separated hexadecimal string format and its subnet mask in CIDR notation from 0 to 128.
      2. Click Save

        The OAM VIP address and Mask are saved. Repeat the process for the second OAM VIP.

    5. (Optional) To enter up to six Signaling VIPs addresses (up to two each for each of SIG-A, SIG-B,and SIG-C), click Add New VIP.

      The signaling VIP is the IP address a PCEF device uses to communicate with the cluster. A non-CMP cluster supports redundant communication channels, named SIG-A and SIG-B, for carriers who use redundant signaling channels.

      The New Signaling VIP dialog box appears.
      1. Enter the Signaling VIP address and the Mask.
        This is the IP address the CMP server uses to communicate with an external signaling network.
        Note: Enter the IPv4 address in standard dot format and its subnet mask in CIDR notation from 0 to 32, or the IPv6 address in standard 8-part colon-separated hexadecimal string format and its subnet mask in CIDR notation from 0 to 128.
      2. Select the Interface from the list.
        Available options are:
        • SIG-A
        • SIG-B
        • SIG-C
      3. Click Save.

        The Signaling VIP address and Mask are saved.

    6. If the hardware type is C-Class, C-Class(Segregated Traffic), or Oracle RMS, configure the General Network settings:

      1. Enter the OAM VLAN ID.

        The default value is 3.

      2. Enter the SIG-A VLAN ID.

        The default value is 5.

      3. (Optional) Enter the SIG-B VLAN ID.

        The default value is 6.

      4. (Optional) Enter the SIG-C VLAN ID.

        The default value is 7.

      Note: Virtual LAN (VLAN) IDs are in the range of 1 through 4095.

    7. If the hardware type is C-Class or C-Class(Segregated Traffic), for the User Defined Network, enter the REP VLAN ID.

      Note: Virtual LAN (VLAN) IDs are in the range of 1 through–4095.

  5. To configure Server-A, in the Server-A section of the page:
    1. (Required) To enter the IP address, click Add New IP.

      The Add New IP dialog box appears.
      1. Enter the IP address in either IPv4 or IPv6 format.

        This is the IP address of the server. For an IPv4 address, enter it in the standard IP dot-format. For an IPv6 address, enter it in the standard 8-part colon-separated hexadecimal string format.

      2. Select the IP Preference: IPv4 or IPV6.
        The server will preferentially use the IP address in the specified format for communication.
        • If neither an IPv6 OAM IP nor a static IP address is defined, IPv6 cannot be selected.
        • If neither an IPv4 OAM IP nor a static IP address is defined, IPv4 cannot be selected.

    2. Enter the HostName of the server.

      The name can only contain the characters A through Z, a through z, 0 through 9, period (.), hyphen (-), and underline (_). This must exactly match the host name provisioned for this server (the output of the Linux command uname –n).
      Note: If the server has a configured server IP, you can click Load to retrieve the remote server host name. If the retrieve fails, you must enter the host name.

    3. Select Forced Standby to put Server-A into forced standby.

      By default, Server-A will be the initial active server of the cluster.

    4. In the Path Configuration section, to add a Static IP, click Add New.

      The New Path dialog box appears.

      Note: If an alternate replication path and secondary HA heartbeat path is used, a server Static IP address must be entered in this field.
      1. Enter a Static IP address and Mask.
      2. Select the Interface:
        • SIG-A
        • SIG-B
        • SIG-C
        • REP
        • BKUP
          Note: If the hardware type is C-Class (Segregated Traffic) or Oracle RMS, the BKUP network is available.

  6. (Optional) To configure Server-B, in the Server-B section of the page:
    1. (Required) To enter the IP address, click Add New IP.

      The Add New IP dialog box appears.
      1. Enter the IP address in either IPv4 or IPv6 format.

        The IP address of the server. For an IPv4 address, enter it in the standard IP dot-format. For an IPv6 address, enter it in the standard 8-part colon-separated hexadecimal string format.

      2. Select the IP Preference: IPv4 or IPV6.
        The server will preferentially use the IP address of the specified format for communication.
        • If neither an IPv6 OAM IP nor a static IP address is defined, IPv6 cannot be selected.
        • If neither an IPv4 OAM IP nor a static IP address is defined, IPv4 cannot be selected.

    2. Enter the HostName of the server.

      The name can only contain the characters A through Z, a through z, 0 through 9, period (.), hyphen (-), and underline (_). This must exactly match the host name provisioned for this server (the output of the Linux command uname –n).

      If the server has a configured server IP, you can click Load to retrieve the remote server host name. If the retrieve fails, you must enter the host name.

    3. Select Forced Standby to put Server-B into forced standby.

      By default, Server-A will be the initial active server of the cluster.

    4. In the Path Configuration section, to add a Static IP, click Add New.

      The New Path dialog box appears.

      Note: If an alternate replication path and secondary HA heartbeat path is used, a server Static IP address must be entered in this field.
      1. Enter a Static IP address and Mask.
      2. Select the Interface:
        • SIG-A
        • SIG-B
        • SIG-C
        • REP
        • BKUP
          Note: If the hardware type is C-Class (Segregated Traffic) or Oracle RMS, the BKUP network is available.

  7. Click Save.

    A confirmation message appears.

  8. Click OK.
  9. If you are setting up multiple clusters, repeat this procedure.
The cluster is defined.
Parent topic: About Setting Up a Georedundant Cluster