IP partnership requirements

• You can only create IP partnerships between two systems.
• A system can only be part of one IP partnership.
• You can create IP partnerships between any two Storwize products on the same layer. In other words, both systems must be at the storage layer or both systems must be at the replication layer.
• You cannot use link-local addressing.
• If you use IPv4 addressing, the management IP addresses on both systems must be IPv4-compliant, and these addresses must have connectivity with each other.
• If you use IPv6 addressing, the management IP addresses on both systems must be IPv6-compliant, and these addresses must have connectivity with each other.
• If your system has IPv4 and IPv6 addressing on the ports, you must configure all remote copy links between the local and remote site for one of those addresses.
• You can configure ports from at most two I/O groups from each system for an IP partnership.
• If your system has 10 Gbps and 1 Gbps Ethernet ports, you must configure all remote copy links between the local and remote site for one of those speeds.
• A system can have simultaneous partnerships over Fibre Channel and IP but with separate systems.
• The IP partner systems must not be visible to each other over Fibre Channel or Fibre Channel over Ethernet (FCoE) connections. Fibre Channel or FCoE ports of either system should not be listed as parameters to the svcinfo lsfabric command.
• Clusters that are configured in active IP partnerships should not be zoned with each other for Fibre Channel or FCoE.
• iSCSI hosts can access volumes over IP ports that are participating in an IP partnership; however, this access might result in an impact on performance.
• VLAN tagging of the IP addresses configured for remote copy is supported. If VLAN is to be configured, ports in the same remote copy group must be on same VLAN on both systems.
• Do not attempt to use Network Access Translation (NAT) with IP partnerships. It will prevent IP partnerships from establishing.
• If you configure two intersite links, you must configure separate remote-copy port groups, one for each link.
• If you have one intersite link, you must configure one remote-copy port group for that link.
• No more than two intersite links/remote-copy port groups are supported.
• If you have one remote-copy port group, then configure one port from each node in one I/O group in that remote-copy port group.
• For systems with more than one I/O group, ports from a second I/O group can be added to the remote-copy port group.
• If you have two remote-copy port groups and one I/O group, then on each system, configure one port from one node in the first remote-copy port group. Then, configure a port from the other node in the second remote-copy port group.
• For systems with more than one I/O group, ports from a second I/O group can be added to each of the two remote-copy port groups.
• Only one port in a node can be configured in an IP partnership.
• If you connect systems by directly attaching them without switches, you must have only two direct-attach links. Both direct-attach links must be on the same I/O group. You should use two port groups, where a port group contains only the two ports that are directly linked.
• Data compression is supported for IPv4 or IPv6 partnerships. To enable data compression, both systems in an IP partnership must be running a software level that supports IP partnership compression.
• To fully enable compression in an IP partnership, each system must enable compression. When compression is enabled on the local system, by using the mkippartnership or chpartnership commands, data sent to the remote system is compressed. To send compressed data to the local system, the remote system in the IP partnership must also enable compression.

Supported IP partnership configurations

The following general configurations are supported, but the number of I/O groups that are configured for each site can be different.

Configuration 1: In this configuration, only a single intersite link is available. Therefore, only a one remote copy port group is configured on each node.

Figure 1. One intersite link, one I/O group per system

Only one port from either of the nodes in each system actively participates in the IP partnership. The other port acts as the failover port. If a critical failure is observed on node H1 in Site H, the IP partnership will fail over to node H2 and continue. Remote copy relationships might stop momentarily during the failover.

Configuration 2: In this configuration, only one intersite link is available. Both I/O groups in each system have a single remote-copy port group configured on each node. However, out of all of the available ports, only one port from either of the nodes in each system actively participates in the IP partnership. The other ports act as failover ports.

Figure 2. One intersite link, two I/O groups per system

If a critical failure is observed on node H1 in Site H, the IP partnership fails over to node H2, H3, or H4 and continue. Remote copy relationships might stop momentarily during the failover.

Configuration 3: In this configuration, eight-node systems are available. However, only two I/O groups are supported in IP partnerships. Both I/O groups in each system have a single remote-copy port group configured on each node. However, out of all the available ports, only one port from either node in each system actively participates in IP partnership. The other ports act as failover ports.

Note: Configuration 3 also applies to systems with four I/O groups. Only two I/O groups in a system can be used for IP partnership.

Figure 3. One intersite link, three I/O groups per system

If a critical failure is observed on node H1 in Site 1, the IP partnership fails over to node H2, H3, or H4 and continue. Remote copy relationships might stop momentarily during the failover.

Configuration 4: In this configuration, two intersite links are available; therefore, two remote-copy port groups are configured. One port from each node in each system actively participates in the IP partnership. If a critical failure is observed on node H1 in Site H, the IP partnership continues over the other port on node H2.

No failure occurs in this scenario; however, the effective bandwidth is reduced to half; only one of the two links is available to facilitate IP partnership traffic. When the failure is corrected, ports will fail back and the IP partnership continues to operate over both links.

Figure 4. Two intersite links, one I/O group per system

Configuration 5: In this multi-node configuration, two intersite links are available; two remote-copy port groups are configured. Four ports (one per node) are configured in remote-copy port groups "1" and "2". Out of the four ports, only two ports actively facilitate the IP partnerships. This port and path selection is maintained by an internal algorithm. The other ports act as failover ports.

Figure 5. Two intersite links, two I/O groups per system

If a critical failure occurs on node H1 in Site H, the IP partnership will fail over to node H3 and continue. The link bandwidth is not affected, as the failover happens immediately and completes quickly while IP partnership traffic continues from node H2.

Configuration 6: In this configuration, eight node systems are available. However, only two I/O groups are currently supported for IP partnerships. Both I/O groups in each system have four ports (one per node) configured in remote-copy port groups "1" and "2". Out of the four ports, only two ports actively facilitate the IP partnerships. This port and path selection is maintained by an internal algorithm. The other ports act as failover ports.

Note: Configuration 6 also applies to systems with four I/O groups. Only two I/O groups in a system can be used for IP partnership.

Figure 6. Two intersite links, three I/O groups per system

If a critical failure is observed on the node H1 in Site H, the IP partnership will fail over to node H3 and continue. The link bandwidth is not affected because the failover happens immediately and completes quickly while IP partnership traffic continues from node H2. The other I/O groups, or all the I/O groups, can be connected over a Fibre Channel partnership with another system.