OpenVMS Device Names

OpenVMS uses a standard format for the names of terminals, disks, tape drives, network adapters, and other devices.

The general format of device names is ddcu: where dd are letters representing the device type, c is a letter indicating which controller (adapter) the device is attached to, and u is the unit number on that adapter. The trailing colon is part of the device name.

So, TTB4: represents a terminal (TT) connected to port 4 on the second terminal adapter (B). TTA6: is port 6 on the first adapter, etc. Note that the first actual device on a given adapter may be either unit 0 or 1.

Some of the more common device types are as follows:

     DI - Disk     DK - SCSI Disk     DP - Disk     DU - Disk     DQ - IDE Disk     DS - Disk Stripe Set     ES - Ethernet Network Device     ET - Ethernet Network Device     EZ - Ethernet Network Device     EF - Ethernet Network Device     EX - Ethernet Network Device     EC - Ethernet Network Device     XQ - Ethernet Network Device     XE - Ethernet Network Device     FC - FDDI Network Device     FX - FDDI Network Device     FT  - Network Terminal     LT  - LAT network Terminal     NTY - Network Terminal     OP  - Operators Console (OPA0:)     TT  - Terminal     TX  - Terminal     RT  - DECnet Terminal (SET HOST)     VT  - Virtual Terminal     MK  - SCSI Tape Drive     MT  - Tape Drive     MU  - Tape Drive 

SCSI devices usually have unit numbers that are multiples of 100. The first SCSI disk will be DKA0:, the second DKA100:, then DKA200:, and so forth.

You may see a device name preceded by an underscore, as in _DKB400:. The underscore signifies that the name is a physical device name, not a logical name.

Some systems, including VMScluster members, sometimes have slightly different device names. Some devices may be preceded by the name of the VMScluster node to which the device is attached. Say, for example, that nodes PHOEBE and WILLIS each have a disk device DKA100:.

The disk attached to node PHOEBE may be shown as PHOEBE$DKA100: and the disk attached to WILLIS as WILLIS$DKA100:.

Depending on system configuration, you may instead see an allocation class associated with certain devices. An allocation class is a number identifying the primary data path to a device that has more than one possible path. This mechanism is designed to provide failover transparently should one path fail.

An allocation class is represented by the string $n$ preceding the device name, where n identifies the allocation class. A system manager may assign a number in the range 1 through 255 as the allocation class for a given cluster node or storage controller. A different number may be used for disk versus tape devices.

So, a disk device may be shown as $1$DUA4: rather than simply DUA4:. Should you find allocation classes in use at your site, simply consider the allocation class to be part of the device name and pay no further attention to them.