Veritas-bu

[Veritas-bu] tape eject after backup - a script that might work for you too....

2000-11-30 16:32:22
Subject: [Veritas-bu] tape eject after backup - a script that might work for you too....
From: John_Wang AT enron DOT net John_Wang AT enron DOT net
Date: Thu, 30 Nov 2000 15:32:22 -0600
Hello David

Seems to be a pretty good library though the 18 slot mail slot doesn't have a
detachable carrier.   Would've been nice if all 18 cartridges could be pulled
from the drawer and swapped with another carrier but that doesn't seem to be the
case.    Also the drawer doesn't seem to have any locks on it so it is possible
to pull the drawer out before the robot is really prepared to let you have it
and you wind up with an annoying alarm sounding off.

Even though the robot rescans all the barcodes if the library door is opened and
then closed, that information doesn't make it into the Netbackup database until
you do so manually from the media manager so it's possible for someone to open
the door, rearrange tapes, and upon closing, find Netbackup picking the wrong
tapes for use.    If some of the tapes are new, this results in tapes with rv's
that don't match their ev's since Netbackup will try and label unlabelled tapes.
Of course, this can be avoided by deactivating all the classes before opening
the door, and updating the database before activating the classes as a matter of
procedure (it would be nicer if Netbackup had a temporary hold feature on the
scheduler).

The only problems I've had with the library so far has been that some of the
SCSI cables were bad (field engineer bent a pin) and the default erase option
for the Sony AIT2 tape drives only writes out 300 eod's rather than writing out
eod's to the end of the partition (really a problem with the drive/OS
configuration not a problem with the library).

Since AIT tapes have cleaning headers, I can't comment on how well the cleaning
cycles work.

There's a truly unnecessary clock built into the LCD panel which must be reset
manually for every time change (I just let it run free).   The plastic buttons
for the LCD control panel tends to fall off (they would've been better off with
foil bubbles).   The cabinet is surprisingly deep, I guess that's to accomodate
the track of the robot and the shifting centre of gravity when the arm moves.

Although the picker seems to move very fast, there's a noticeable time delay
between when I issue a command and when the arm starts to move.   Also, I'm not
conviced that speed is really an issue with an automation product i.e.: any
speed is better than paging me at night for a tape change.

Injecting tapes through the mail slot seems to require specifying a destination
slot for each tape, probably a Netbackup limitation and seems to be irrelevant
of the barcodes, probably another Netbackup limitation as the library does scan
the barcodes in the mail slot each time the drawer is closed.

The unit has a marketing mode where it just moves the arm around and looks busy
during which the entire unit is offline and unusable.   Management and
particularly sales types tend to leave the unit in marketing mode whenever they
can which is really annoying cause then no production can occur.    It would've
been nice if an actual request could interrupt marketing mode to service the
request.

The keys for the front door is different from the keys for the back door.   A
bit unnecessary in my opinion.

When power is dropped, the arm free falls to the bottom with the only drag being
that of the pulleys.   It would've been nice if the arm had been better counter
weighted or had some kind of brake to support it's weight in the power off
situation, at least nicer than hearing that unsettling whine followed by a
thunk.

The tape library was selected before I started here so I wasn't involved with
the selection process, still it seems to be a reasonable choice.

Regards,
John I Wang
Sr. Systems Engineer
Steverson Information Professionals

---
Enron Broadband Services
Enron Building 1472c
ph (713) 345-4291
fax (713) 646-8063


|--------+----------------------->
|        |          david@datasta|
|        |          ff.com       |
|        |                       |
|        |          11/30/00     |
|        |          02:55 PM     |
|        |          Please       |
|        |          respond to   |
|        |          david        |
|        |                       |
|--------+----------------------->
  >-----------------------------------------------------------------------|
  |                                                                       |
  |       To:     kolinsky AT colltech DOT com                                  
 |
  |       cc:     John Wang/Contractor/Enron Communications@Enron         |
  |       Communications, DONALDG AT tc.gc DOT ca, shyamdayal AT usa DOT net,   
        |
  |       veritas-bu AT mailman.eng.auburn DOT edu                              
 |
  |       Subject:     Re: [Veritas-bu] tape eject after backup - a script|
  |       that might work for  you too....                                |
  >-----------------------------------------------------------------------|



Ah, yes...that looks good.

some robots (I think older ones) may have required the -rc1 0 for
positioning purposes.  It just one of those things thats hard to break
when you have been doing it for years, plus I haven't had it 'blow up' on
me either.

However, I will be eval'ing a Scalar 1000 very shortly.

John: What's your take on that unit, first glance?

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
David A. Chapa
Consulting Manager
DataStaff, Inc.
847 413 1144

On Thu, 30 Nov 2000, Lee Kolinsky wrote:

> John,
>
> I think you need to drop the "-rc1 0" part of the command
> You are basically telling vmchange to put each tape in the
> slot.  The robot is smart enough to know that that slot has
> been used and will automatically go to the next one.
>
> Here is the command I use:
>
> $media_type="dlt"
> foreach $media_id (000123 000124 000125 ...)
>
>          vmchange -res -m $media_id -mt $media_type -v OFFSITE -e
>
> end
>
> --
>
> --Lee
>
> Client Site: EATON Corp. Cleveland OH (440) 954-5626
> ----------------------_-----------------------_--------------------------
>  Lee S. Kolinsky     |_| C o l l e c t i v e |_|   kolinsky AT colltech DOT 
> com
>  Pager: 800/946-4646 |___   Technologies    ___|  http://www.colltech.com
>  Pin #: 148-3408         O                 O     "The Power Of Many Minds"
> -------------------------------------------------------------------------
>          What the large print giveth, the small print taketh away.
> _______________________________________________
> Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
>