HPlogo High Availability FailOver/iX Manual > Chapter 4 Configuration

Configuration Planning

MPE documents

Complete PDF
Table of Contents

E0803 Edition 2
E1100 Edition 1 ♥

Detailed planning of the hardware configuration of the array itself and the actual HAFO configuration (in SYSGEN) is required. Before taking any action on the array or in SYSGEN, the user should have a complete map and plan for the setup. The path layout and primary and alternate paths for all the ldevs to be covered by HAFO must be determined in advance. Both user and system volumes may be configured for high availability failover, but all ldevs covered by HAFO must be in the XP256 array.

Before beginning HAFO configuration in SYSGEN, the array must already be cabled and the hardware configured to work with the HAFO utilities. As mentioned previously, the array hardware itself allows for two controllers to be configured to talk to the same ldevs.

HAFO Configuration Restrictions

  1. HAFO works with pairs of controllers. That is, all ldevs on a given primary path/controller, for example (path A) must be configured to fail over to the same alternate path (path B). By the same token, any ldevs that happen to have path B as their primary path, must be configured to have path A as their alternate path.

  2. All ldevs on a given path must be configured the same. That is, if any ldev on a path is configured for failover, then all ldevs on the path must be configured for failover to the same alternate path.

  3. The maximum number of ldevs supported is seven ldevs per path. That is, one path in the pair can have seven ldevs and the other path in the pair can have seven ldevs. This limits the configuration to a total of 14 ldevs on a single path should a failover event occur. However, this may have negative performance implications.

Configuring HAFO in Pairs

Figure 4-1 "Configuring HAFO in Pairs" shows the ldevs in the top path could be configured to failover to the bottom path. If any of the ldevs 1, 20, or 21 are configured to failover to the bottom path, then all must be configured to failover to the bottom path. Further, since in this example, some ldevs indeed exist on the bottom path, they must all (ldevs 101, 102 and 103) be configured to failover to the top path.

Figure 4-1 Configuring HAFO in Pairs

[Configuring HAFO in Pairs]

NOTE: The SYSGEN (io) configured SCSI target (id) that is being used on a primary path cannot be configured in the (io) section of SYSGEN on the alternate path. For example, if the path for ldev 1 were 8.0.0, and the alternate path were 48, then one should not use 48.0.0 for any of the ldevs on that path. None of the ldevs 101-103 could be configured as 48.0.0. A correct configuration would to make ldevs 1, 20, and 21 be 8.0.0, 8.1.0, and 8.2.0. Then make ldevs 101, 102, 103 be 48.3.0, 48.4.0, and 48.5.0. Now ldevs 1, 20, and 21 can failover to path 48 and ldevs 101-103 can failover to path 8 with no conflicts.

Again, HAFO does not address the hardware configuration of the array itself. In the example above, the array itself would have to be pre-configured to allow the top controller to talk to the ldevs on the bottom and to allow the bottom controller to talk to the ldevs on the top.

Each controller on the array that is to be used must be connected to a single host adapter card. All host adapters to be used in HAFO configuration must be configured in SYSGEN and initialized during boot.

Configuration Map

Creation of a configuration map is crucial. Proper planning, mapping, and maintenance of the map are critical to the successful use of HAFO. Without a proper plan and up-to-date map, troubleshooting will be difficult and system availability will not be optimized.

The following is a checklist of the steps required to enable HAFO:

  1. Select the array and volume-set(s) to be configured for HAFO. Ldev 1 and the system volume set can be configured for HAFO.

  2. Identify the logical data paths of the fast-wide SCSI Host adapters for each pair of controllers in the array. One path will be the primary path, fully qualified, and the other will be the alternate data path.

  3. Ensure that all paths (I/O adapter cards and ldevs) are configured and recognized when the system is booted.

  4. Create a map of exactly how the layout will look, including the path pairs and each ldev on each path. Include the addresses of each card and ldev, and clearly show the failover scheme.

  5. Ensure that the array itself is configured to allow each primary path controller to talk to the ldevs on its corresponding alternate path.

  6. When the entire plan and map are complete, configure the primary and alternate data paths in SYSGEN.

CAUTION: Mirrored disk/iX and HAFO products cannot be used together for the same volume set.

Mirrored disk and HAFO can be used on the same system with multiple volumes and multiple arrays. However, these products must NOT be used simultaneously on the same user-volume set. For example, HAFO could be configured for the system volume set (which is on an array), and a user-volume of the same or different array. Mirrored disk/iX could be used on standalone disks in towers or in enclosures.



Chapter 4 Configuration

Accessing the HAFO Configuration Commands