mounting disk

[Riverhawk]

I have an issue when I mounted a new disk drive it would show mounting alloc when others are shown as mounted when i run the show dev d command. What does that mean and what am i doing wrong?

[Steven Schweda · Accepted solution]

> I have an issue [...]

Is that anything like a _problem_?

HELP MOUNT

MOUNT /SYSTEM (or /CLUSTER)?

As usual, showing the actual command(s) used
(and actual command output) might be more
helpful than vague descriptions.

[marsh_1]

riverhawk,

when you don't use the qualifiers that steven has shown - just a straight mount - the device , in this case a disk will be allocated to the process that issued the mount command and show as 'allocated'. if you do a show dev d/full you'll see the difference.

[Riverhawk]

Thanks for the clearification guys. Issue/problem resolved.

[Hoff]

It's perfectly possible and fully supported to privately-mount already mounted (shared) disks, too.

This is intended to allow long-running batch jobs (for instance) to express an interest in retaining a disk in its mounted state.

[Riverhawk]

I have mounted the new disk but when I try to access it using this command dir diskname:[000000] it would show error in device name or inappropriate device type for operation. Have I forgotten to do something else after it has been mounted? Sorry I'm new to the vms world.

[Steven Schweda]

> Have I forgotten to do something [...]

You've forgotten to show the actual
command(s) you used, and the actual output
therefrom. "diskname", in particular, might
have been interesting. "SHOW DEVICE D",
too.

[Hoff]

Please post the DCL commands that were used here, and post the full text of the error and (for completeness) the OpenVMS version and platform.

The following commands can retrieve device information and can display the OpenVMS system version, respectively:

SHOW DEVICE /FULL ddcu:
SHOW SYSTEM /NOPROCESS

I strongly -- strongly -- recommend skimming through the OpenVMS User's Guide, part of the OpenVMS documentation set. Having some background in the terms and concepts and command structure -- will greatly speed your requirements. After you have a grip on the User's Guide, the next step is the OpenVMS System Manager's Essentials manual. This is heavier going, and builds on terms and concepts from the User's Guide.

The manuals are here:

http://www.hp.com/go/openvms/doc/

OpenVMS really isn't set up to be particularly forgiving to an inexperienced system manager, and -- as you've found -- it can be somewhere between arcane and opaque if you're not up to speed with the terms and concepts.

We were all new to OpenVMS once, too. And we all eventually read through the core manuals, as much as we didn't want to.

When in doubt, always post what problem(s) you seek to solve (this in general terms; background information), always post the exact command(s) used, and always post the exact error message(s) received. And (for most cases) information on the OpenVMS version and platform, and (for cases such as this) information on the device and the storage media.

The inclusion of the background in the question -- the problem details, the commands, and the full text of the error messages -- is somewhere between very important and critical to getting a fast (and right) answer to your intended question.

[Riverhawk]

Thanks for the links...attached is the info that was requested.

[Steven Schweda]

> $1$DGA2337: (NRCAVA) Mounted 1 RMADSK08 244352160 1 3

> NRCAVA::Theresa's:::>dir rmadsk08:[000000]
> %DIRECT-E-OPENIN, error opening RMADSK08:[000000]*.*;* as input

You're using the volume label where you
should be using the device name. Try

dir $1$DGA2337:[000000]

There may be some "DISK$volume-label" logical
names which were defined when the disks were
mounted, too.
SHOW LOGICAL DISK$

So, for example, this might work, too:

dir disk$rmadsk08:[000000]

[Steven Schweda]

Reading further, ...

> But when I execute the same command on a
> known device I will receive the following.
> [...]
>
> NRCAVA::Theresa's:::>dir prog$disk:[000000]/tot
> Directory RMADSK05:[000000]
> Total of 54 files, 225245 blocks.

But "prog$disk" is a logical name, and, I'd
guess, so is RMADSK05 (with some slightly
exotic properties).

SHOW LOGICAL /FULL prog$disk, RMADSK05

Presumably, someone has defined RMADSK05 (and
others) somewhere, but you haven't done the
corresponding things for your new disks.

[Hoff]

Ok. OpenVMS Alpha V7.2-2. That is quite old, and one of the earlier OpenVMS releases with Fibre Channel SAN (FC SAN) support. OpenVMS Alpha V8.3 is current.

There are errors showing on some of the FC SAN disks, including one that is apparently involved here. These errors may or may not indicate a problem with these disks; these could be transient errors, or hard errors.

At the root of the failing DIRECTORY command are missing site-specific logical names. There are some logical names defined here, and they're not in place for the disks that (I'm guessing) have been added:

SHOW LOGICAL RMADSK*

The logical names typically created when a disk is mounted can be viewed with the following:

SHOW LOGICAL DISK$*

And to access your disk, you can specify the physical device name here. For instance:

DIRECTORY $1$DGA2337: [000000]

Check the system startup procedures for where these local logical names are defined. It's probably either in SYSTARTUP_VMS.COM or in SYLOGICALS.COM, both of which are usually located in SYS$MANAGER: directory. The DISK$* logical names are typically defined automatically, when the disk is mounted.

A formal configuration review and (if you're in a hurry to get up to speed) some on-site or classroom training is in order here, particularly if (as it might be inferred) you've been handed management of this OpenVMS Alpha server. If there are budgetary constraints and/or if expediency is not a requirement, then the User's Guide and the System Manager's manuals should be high up on the reading list.

I'll leave the resolution of this particular case to others here.