This is kind of a dumb question, and I'm not sure how best to express
it, but I'll try:
If you clone all the save sets on a given tape as:
nsrclone -b 'clone pool' -S -f file
where 'file' contains a list of all the ssids on the tape, and the save
sets clone successfully to the target tape, then have you really
validated anything?
In other words, let's suppose instead that you cloned maybe half of the
save sets using the same method. In this case, because the save sets are
interleaved with others that may not be in the list, it's going to have
to demultiplex those in order to read them so that it can then write
them to the target clone tape, right? In this situation, I can see that
NW is actually reading the requested data and writing it to the clone
target volume.
BUT, in the first scenario, wherein you specify all the ssids, NW is
able to more efficiently clone them because it doesn't have to
demultiplex anything, correct? In other words, it can preserve the
multiplexing, making the operation more efficient over all, although
more data would have to be cloned than in the second scenario. BUT, in
this case, is it really reading anything like it does in the second?
Just seems that the second scenario would be a more thorough test?
Am I wrong here? Is my understanding skewed?
This came up because I clone certain backups manually (cloning is not
enabled for the affected groups), and I what I used to do was to launch
a clone command (like the one above) wherein I would clone all the save
sets from the tapes that were already full, skipping any that spanned
onto appendable volumes, and then level full backups could also run at
the same time to those appendable volumes and so on and so forth
throughout the week. Finally, once the final fulls were done, I'd then
clone those. Obviously, this involved a number of separate clone
operations, and each had to clone a subset of the overall save sets, so
it was more inefficient than simply waiting until all the fulls were
done and then running one clone command for everything as this will
better preserve the multiplexing. BUT, in retrospect, I'm wondering if
the former method is a better test or validation as it has to
demultiplex since it's working on a subset? OR, is running on the whole
thing good enough?
Thanks.
George
--
George Sinclair
Voice: (301) 713-3284 x210
- The preceding message is personal and does not reflect any official or
unofficial position of the United States Department of Commerce -
- Any opinions expressed in this message are NOT those of the US Govt. -
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
|