Networker

Re: [Networker] Creating a dump of a legato tape?

2003-04-17 17:56:32
Subject: Re: [Networker] Creating a dump of a legato tape?
From: George Sinclair <George.Sinclair AT NOAA DOT GOV>
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Date: Thu, 17 Apr 2003 17:56:21 -0400
Well, yes, I see your point. If someone were to recycle the original
tape then NetWorker would get pretty confused if you later popped in the
copy, but let me ask you this. Is this situation not identical to one
wherein you remove the volume from the volumes listing (thus removing
all saveset entries from the media database) but later scan in the tape
since you still have it?

Yeah, in our case the label name of the original copy will NOT be
reused. We can guarantee that. Furthermore, the tape itself will most
likely never be recycled. So it just seems like 'dd' would be the most
efficient means to create a copy of the tape. I don't care about
un-multi-plexing the data, so that advantage of cloning is moot. Also,
the advantage to making a copy that I see is that I don't have to worry
about continuation savesets and whether or not the savesets will fit on
the copy because a copy is a copy, assuming, of course, the copy tape
has enough native capacity. I guess there could be issues with various
device compressions, though. So, if I were going from LTO to SDLT or
vice versa the device compression could differ.

George

Robert Maiello wrote:
>
> On Thu, 17 Apr 2003 16:30:36 -0400, George Sinclair
> <George.Sinclair AT NOAA DOT GOV> wrote:
> If one makes a duplicate of a tape Networker has no way of tracking it.  The
> orignal copy can have a location of "Jukebox" or "Offsite" but Networker
> will have no knowledge where the copy is .
>
> Also,  if Networker recycles the original tape, called tape001, and new data
> is written to it, it still has no knowledge of the copy.   Its a bit
> confusing now because we have a tape call tape001 from say March and a tape
> called tape001 from April... if you try to scan in the old tape to recover
> it will conflict with the current tape called tape001.  Very confusing.
>
> I guess if you know the original volume name will never be reused a copy is
> good ..ie.  a second or third copy of an evidentiary backup.
>
> Robert Maiello
> Thomson Healthcare
>
> >So I guess the point is that if you can use utilities like 'dd', or
> >'tcopy' on Solaris, then why ever use the clone volume feature? Other
> >than the fact that cloning removes the multi-plexing (nice), wouldn't
> >'dd' or 'tcopy' be faster and thus preferred?
> >
> >George
> >
> >Yura Pismerov wrote:
> >>
> >> George Sinclair wrote:
> >> >
> >> > We all know that you can clone a NetWorker volume which is really just
> >> > cloning all the savesets, but does anyone if you can use something like
> >> > 'dd' to dump the contents of a NetWorker labeled tape to something like
> >> > tar and then re - 'dd' it back to another "blank" tape?
> >>
> >>         Why don't run
>
> dd directly tape-to-tape with bs=64k (that is what
> >> Networker uses) ?
> >>
> >> >
> >> > In essence, I'm trying to determine if it's possible to create an exact
> >> > copy of a NetWorker labeled tape without using NetWorker. Seems it would
> >> > be faster than cloning since it would not have to un-multi-plex
> >> > anything. It could just read and write. Data is data so why wouldn't it
> >> > work? Obviously, NetWorker would not have to be aware that the duplicate
> >> > tape existed, until you used it in which case it would just think it was
> >> > the original. Would they not be interchangeable?
> >> >
> >> > Alternatively, if the NetWorker label on the source tape proves a
> >> > problem then might it be possible to somehow skip past this part of the
> >> > source tape, create a new labeled tape and then dump everything else on
> >> > the original tape that comes after the label to the new labeled (copy)
> >> > tape.
> >> >
> >> > Would be curious to know if anyone has experimented with this and how
> >> > you did it.
> >> >
> >> > Thanks.
> >> >
> >> > George
> >> >
> >> > --
> >> > 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.
> >> > =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
> >>
> >> --
> >> Yuri Pismerov, Sr. System Administrator,
> >> TUCOWS.COM INC. (416) 535-0123  ext. 1352
> >
> >--
> >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.
> =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
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.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

<Prev in Thread] Current Thread [Next in Thread>