[comp.sys.apollo] Accessing tape cartridge on other node?

herb@blender.uucp (Herb Peyerl) (02/09/91)

krowitz@RICHTER.MIT.EDU (David Krowitz) writes:
>I/O can be used). You must either "crp", "telnet", or "rlogin" to the machine
>to which the tape is attached before you can access it. This is likely to

Unless you have Workstation Solution's "Backup Accelerator".  You can
then wbak to a remote tape drive using:

tbs_wbak -dev [whatever] -on //other_node

You need to have the mbx_helper running on both nodes as well as
server_process_manager.   If you want to rbak from a remote node 
then you're sort of screwed.

-- 
--------------------------------------------------------------------------
UUCP: herb@blender.UUCP   || #define Janitor Administrator
ICBM: 51 03 N / 114 03 W  || Apollo System_Janitor, Novatel Communications
"I spilled spot remover on my dog and now he's gone..." <Steven Wright>

chris@endgame.gsfc.nasa.gov (Chris Shenton) (02/12/91)

I'm writing a tar backup script for users, and we have only one cartridge
(on a DN4000). I would like them to be run it from any machine, and this
requires mounting the remote device, as in:

mt -f //cartnode/sys/node_data/dev/rct8 rewind

but it complains:

?(mt) Unable to acquire //asylum/sys/node_data/dev/rct8
- device in use (OS/SCSI manager)

even when noone is on the node.

What's the story? Is there any way around this?

Thanks.
--
chris@asylum.gsfc.nasa.gov, ...!uunet!asylum.gsfc.nasa.gov!chris, PITCH::CHRIS

krowitz@RICHTER.MIT.EDU (David Krowitz) (02/12/91)

Unlike disk devices (ESDI and SCSI hard disks, floppies, and opticals),
tape devices can only be accessed from the machine to which they are
physically attached. In fact, this is true for all I/O devices *except*
disk devices. The reason for this situation is that Domain/OS automatically
provides low-level servers for network disk I/O (these are the "netreceive",
"netpaging", and "netrequest" processes which show up when you do a
"/bin/ps -aux" command), but not for any other devices. It would have been
nice if servers for these other devices had been implemented, but Apollo
got caught up in doing Just Like Real Unix (JLRU) and has never had the
time to do anything else. OSF/1 won't even have the low-level disk servers
(it *will* have network disk I/O! Don't get me wrong ... but AFS is a
high-level network disk I/O server like NFS -- ie. your system must already
be booted up and have its TCP/IP servers running *before* the network disk
I/O can be used). You must either "crp", "telnet", or "rlogin" to the machine
to which the tape is attached before you can access it. This is likely to
be the case for the forseable future.


 -- David Krowitz

krowitz@richter.mit.edu   (18.83.0.109)
krowitz%richter.mit.edu@eddie.mit.edu
krowitz%richter.mit.edu@mitvma.bitnet
(in order of decreasing preference)