HP3000-L Archives

November 2000, Week 3

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Nizzardini, Al" <[log in to unmask]>
Reply To:
Nizzardini, Al
Date:
Tue, 21 Nov 2000 18:32:53 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (134 lines)
Doug, do not believe a request is pending

Thank you,

Al Nizzardini
Technical Consultant
Computer Design & Integration LLC
696 Route 46 West
Teterboro, NJ 07608
<mailto:[log in to unmask]>
T:  201-931-1420 x252
F:  201-931-0101
P: 973-205-3922


-----Original Message-----
From: Doug Werth [mailto:[log in to unmask]]
Sent: Tuesday, November 21, 2000 5:47 PM
To: Nizzardini, Al; [log in to unmask]
Subject: Re: Turbostore question


Al,

Doesn't the RECALL command show you that there is a tape request pending? I
suppose you could write a job that does a RECALL within a loop.

Here is a crude attempt at such a job with no testing included.

!job WatchBak,manager.sys
!stream fullback.job.sys
!while jobinfo("!hplastjob","exists")
!   purge recallz
!   build recallz;rec=-80,,f,ascii
!   file recallz,old;dev=disc
!   recall >*recallz
!   if finfo("recallz","eof") > 1
!       tellop  Tape Request Pending
!#   or add your own code here to process the output of the RECALL
!   endif
!   pause 60
!endwhile
!# Backup job has finished
!eoj


It would be really nice if TurboStore would continue to put out console
messages until the tape request has been satisfied, similar to the way
Mirror/iX displays console messages independent of the actual REPLY when a
mirrored disk is disabled. (Donna? A SYGSYSMAN item perhaps?)

Doug.

Doug Werth                             Beechglen Development Inc.
[log in to unmask]                               Cincinnati, Ohio

----- Original Message -----
From: "Nizzardini, Al" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Tuesday, November 21, 2000 5:25 PM
Subject: Re: Turbostore question


> I do not think I am making my issue clear. The problem is that when there
is
> a request to mount an additional tape it only appears on the console. If
you
> are not constantly monitoring the console or you have many console
messages
> displayed you will miss this request. At this point the backup will just
> stay in the wait state until you mount a new scratch tape. I am trying to
> make this process as automated as possible (no excuses as to why the
backup
> is still running). I guess one answer is to keep doing a showproc on the
job
> to see if it is getting any cpu time or have glance/sos running to see if
> the job is getting cpu time. Anything else I am missing?
>
> Thank you,
>
> Al Nizzardini
> Technical Consultant
> Computer Design & Integration LLC
> 696 Route 46 West
> Teterboro, NJ 07608
> <mailto:[log in to unmask]>
> T:  201-931-1420 x252
> F:  201-931-0101
> P: 973-205-3922
>
>
> -----Original Message-----
> From: Doug Werth [mailto:[log in to unmask]]
> Sent: Tuesday, November 21, 2000 5:06 PM
> To: [log in to unmask]
> Subject: Re: [HP3000-L] Turbostore question
>
>
> John Burke <[log in to unmask]> writes:
>
> > I agree that two (or more, we use five) tape drives are better than one;
> > however, rather than a sequential backup (still takes as long) or a
> parallel
> > backup (too complex when a restore is required) why not two separate,
> > independent, backup jobs?
> >
> > We actually do five simultaneous backups to five tape drives. The
obvious
> > advantage this has over a sequential backup is the time factor. The
> obvious
> > advantage over a parallel backup is ease of restore. Perhaps less
obvious,
> > since the backups are independent, if one fails for any reason, we still
> > have a good backup of 4/5's of our system
> >
> > The downside to this scheme is you have to manually determine how to
split
> > your system and periodically monitor how much is backed up by each
> process;
> > however, it has served us well for many years and once set up requires
> only
> > minimal attention.
>
> Exactly. My assumption is that this is a single tape backup that was
> spilling over onto a second tape. The *easiest* thing to do (besides just
> manually changing tapes) without introducing other complexities would be a
> sequential backup. Anything else would require deeper analysis and an
> overhaul of the backup procedure.
>
> Doug.
>
> Doug Werth                             Beechglen Development Inc.
> [log in to unmask]                               Cincinnati, Ohio

ATOM RSS1 RSS2