Re: [Networker] To index save or not?

In regard to: Re: [Networker] To index save or not?, MIchael Leone said (at...:

Only if the volume is recycled (and therefore erased) are the database
entries normally removed.  So if you still have the tape, and you expect
to read old data from it, I expect the media database still tracks the
volume and the savesets on it, so 'scanner -m' is unnecessary.


Huh. So ... let's use a specific example ... I have an EOM tape from
2007-04-30. Well beyond browse time, but I do see it in the media
database, and it is listing what savesets are on the tape.


Which means you do NOT need scanner -m for that tape.

So, presuming I
know which specific file of a saveset I want, I should be able to do a
restore by doing a "Saveset Recover", and choosing the 2007-04-30 saveset
(even tho that is well past the retention period).


That's correct.

And no "nsrck -L7" needed,


It's not needed.  However, if you ever find yourself in the situation
where you're beyond the browse period (so the client file index no longer
has the list of files that are contained in the saveset) BUT you have
the necessary client index: savesets available, using nsrck -L7 to reload
the index: entries and then doing a normal recover will be the fastest way
to get the data back, especially for large savesets.

no "scanner -i",


Not needed.  You use scanner -i when

        - you don't have the necessary index: savesets available, so nsrck -L7
          won't work
        - you don't know the exact path(s) to the file(s) you need so you
          can't do a saveset recover AND you don't want to recover the entire
          saveset and pick out the bits you need.

no scanner -m"?


Definitely not needed, because your media database still has the
information about the tape and the savesets that are on the tape.

Just have the
tape in the library, and rock-n-roll? :-) Or would I still need "scanner
-i" (yes, because I need to re-scan in the individual files in the
saveset?).


No, because saveset recover doesn't use the client file index at all.  It
does something *like* scanner -i -- it essentially hunts through the
contents of a saveset, looking for the path(s) you've specified, and
recovers it.  It doesn't rebuild the client file index like scanner -i
does, though.

So when WOULD I need the "scanner -i" or "scanner -m", then?


Hopefully I've explained that above, but essentially

        - scanner -m adds information to the media database about the volume,
          telling it what savesets are on the volume BUT NOT THEIR CONTENTS.

        - scanner -i does everything that scanner -m does PLUS it builds a
          list of the CONTENTS OF THE SAVESETS, which get loaded into the
          client file indexes for the relevant client(s).

I'm back to
being really confused ... would I need to do "scanner -m" on a tape volume
that I do *not* see in the media database?


Yes, assuming you're not planning on doing a scanner -i.  There's NEVER
a reason to do these two steps in succession

        scanner -m
        scanner -i

since scanner -i does everything that scanner -m does.  If you know you're
going to need to use scanner -i, just skip the scaner -m.

Or "scanner -i"?


That depends on whether you want to build up client indexes so they
contain the contents of whats in the savesets.

scanner -i does everything that scanner -m does PLUS it actually goes
through the contents of the savesets on the tape, figures out what paths
are contained within, and builds up the indexes for the client(s) in
question, so that you can browse with "recover".

And would I
then need "nsrck -L7"? (probably not)


That also depends.  If you know there are index: savesets on the tape
for the client you're interested in, it would very likely be faster to
do

        scanner -m

to get JUST the list of savesets into the media database, followed by

        nsrck -L7 -t sometimestamp clientname

to get the index entries back into the online indexes so that you can
use recover.  Those two steps should be faster for most savesets than
doing just

        scanner -i

to rebuild both the media database and the client file indexes at once.

Tim
--
Tim Mooney                                        Tim.Mooney AT ndsu DOT edu
Information Technology Services                   (701) 231-1076 (Voice)
Room 242-J6, IACC Building                        (701) 231-8541 (Fax)
North Dakota State University, Fargo, ND 58105-5164

To sign off this list, send email to listserv AT listserv.temple DOT edu and type 
"signoff networker" in the body of the email. Please write to networker-request 
AT listserv.temple DOT edu if you have any problems with this list. You can access the 
archives at http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER