So, this may not be appropriate...so I'll ask for forgiveness now...but
if you had a virtual tape library system in place, then you would
increase the reliability of your cloning and thus reduce the
interruptions felt as a result of the drive/media issue.  An even better
design would be an embedded path to tape that would allow you to then
move your cloned tape from the virtual space to a physical cartridge for
offsite storage...so take a look at my sig line and I'll end the
commercial now.
 
David A. Chapa * Technical Advisor, Technical Marketing * ADIC *
720.249.5836 * david.chapa AT adic DOT com
The FUTURE is HERE - http://www.adic.com/gopathlightvx
<http://www.adic.com/gopathlightvx> 
-----Original Message-----
From: Maiello, Robert [mailto:robert.maiello AT thomson DOT com] 
Sent: Wednesday, March 17, 2004 9:30 AM
To: David Chapa; Legato NetWorker discussion
Subject: RE: [Networker] Cloning
 
Good question.. 
First my sentence should have read  "nsrclone can hang indefinitely
reading the source 
tape if there is a write error on the target tape."   
I did have the target tape drive replaced after noticing that 3 out 4 of
my write errors were 
on that drive.   That said,  the problem manifested itself with
networker not releasing the 
source drive (drive being read from).  It goes on to use the target tape
drive later in the night 
that had the write error. 
I suppose the drive, if bad, may not send a proper signal to the
nsrclone process?  
Legato seemed to think it belonged to this bug/issue number.   I tend to
agree,  if nsrclone gets 
an error writing to a tape it should deal with it in some coherent way
(fail?).   In my case it 
appears that the root cause of the write errors may have been the drive.

Robert Maiello 
Thomson Healthcare 
 
 -----Original Message----- 
From:   David Chapa [mailto:david.chapa AT adic DOT com] 
Sent:   Wednesday, March 17, 2004 11:13 AM 
To:     Legato NetWorker discussion; Robert Maiello 
Subject:        RE: [Networker] Cloning 
Does this seem to be an nsrclone issue specifically or drive/media 
issue? 
David A. Chapa * Technical Advisor, Technical Marketing * ADIC * 
720.249.5836 * david.chapa AT adic DOT com 
The FUTURE is HERE - http://www.adic.com/gopathlightvx 
 
-----Original Message----- 
From: Legato NetWorker discussion [mailto:NETWORKER AT LISTMAIL.TEMPLE DOT EDU]

On Behalf Of Robert Maiello 
Sent: Tuesday, March 16, 2004 2:18 PM 
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU 
Subject: Re: [Networker] Cloning 
All, 
I had jumped on this thread (sorry) with an issue with my nsrclones 
hanging 
when a SCSI write error was encountered.   I did find the source of my 
write errors to be one drive.   Still, it would seem nsrclone can hang 
indefinitely reading the source tape if the tape tape has a SCSI write 
error. 
Legato has added me (and this behavior) to LGTpa60542. No hotfix for 
7.1.1 
yet..may have to wait for next release. 
 
Robert Maiello 
Thomson Healthcare 
 
On Mon, 8 Mar 2004 11:27:31 -0600, Paula Ehn <pehn AT CC.UMANITOBA DOT CA> 
wrote: 
>Solaris 8/ Networker 7.0 
> 
>We are certainly having a similar problem here only we are running in a

Solaris 
>environment. I had a Legato case open before Christmas but they kept 
blaming the 
>tape drive hardware. At the time, it had only occured once on that 
drive but 
>since has occured many times and at least once on all our drives. I 
admit, our 
>present backup server is under-powered for what it is doing but we are 
upgrading 
>it very shortly. I still can't believe that is it because it was 
working fine 
>until I upgraded from Networker 6.1.3 to 7.0. 
> 
>When I find it in this "hung/SuperSlow" state, I just kill the nsrclone

process 
>and play catchup later by running the nsrclone command manually. It 
annoys 
me to 
>no end when it happens on a weekend! 
> 
>We did just get a bunch of new tapes of which were delivered to us when

it was 
>-40C outside. A couple in the batch have already had issues and been 
marked 
full 
>far too early. I could see how this might be related to the cloning 
problem as 
>it's happened more frequently since we started using tapes from this 
new batch. 
> 
>Paula 
> 
> ----------------------------------------------------------------- 
>| Paula Ehn  VE4MHZ/VE4HZ                    Room 624 Eng.        | 
>| Unix System Administrator                  pehn AT cc.umanitoba DOT ca | 
>| Academic Computing and Networking          (204)474-7849        | 
>| University of Manitoba                     Fax: (204) 474-7920  | 
> ----------------------------------------------------------------- 
> 
>On Fri, 5 Mar 2004 14:53:59 -0500, Robert Maiello 
><robert.maiello AT THOMSON DOT COM> wrote: 
>> 
>>I've found with version 7.1.1 on Unix the cloning behavoir has 
changed. 
>>If we encounter an error on the tape being written to nsrclone 
hangs/never 
>>finishes.  In the morning I see the clone is running, the source tape 
is 
>>reading but nothing is being written to.  A look at the drive being 
read 
>>from reveals its not really being read. 
>> 
>>I know this is quite the problem you have but something definitely 
changed 
>>in this version.  In 6.x the clone would fail on tape write errors and

>>the drives would be free. 
>> 
>>Has anyone notice this or Brian's problem when cloning in version 
7.x?? 
>> 
>> 
>>I also have a case open with Legato. 
>> 
>>Robert Maiello 
>>Thomson Healthcare 
>> 
>> 
>>On Fri, 5 Mar 2004 08:10:21 -0500, Spindler, Brian 
>><brian.spindler AT RUESCH DOT COM> wrote: 
>> 
>>>Networker 7.1, Windows 2000 
>>> 
>>>OK! I need some real help here with Cloning.  My Clone Sets NEVER 
>>>complete.  Every morning when I come in the cloning sessions are 
still 
>>>running and when I look at the status they are "writing, idle" and 
>>>"reading, done".  When I look at the perfmon counters for "Write 
>>>kilobyte / sec" there is no activity whatsoever.  How can I make sure

>>>that my clones continue to run as scheduled?  AND if they are hung 
how 
>>>can I nudge them along?  Any help would be greatly appreciated! 
>>> 
>>> 
>>>Thanks, 
>>>Brian 
> 
>-- 
>Note: To sign off this list, send a "signoff networker" command via 
email 
>to listserv AT listmail.temple DOT edu or visit the list's Web site at 
>http://listmail.temple.edu/archives/networker.html where you can 
>also view and post messages to the list. 
>=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*= 
-- 
Note: To sign off this list, send a "signoff networker" command via 
email 
to listserv AT listmail.temple DOT edu or visit the list's Web site at 
http://listmail.temple.edu/archives/networker.html where you can 
also view and post messages to the list. 
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*= 

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
______________________________________________________________________

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=