EMC NetWorker discussion <NETWORKER AT LISTSERV.TEMPLE DOT EDU> wrote on 
06/04/2010 12:45:55 PM:

> You need a list of savesets (probably ssid/cloneid actually), so that 
will
> mean some reporting, most likely with mminfo.
> 
> Once you have the ssid/cloneids, you have a couple options:
> 
> - make NetWorker completely and utterly forget about the saveset by 
using
>    the -d option with nsrmm.  This doesn't just purge the client file 
index
>    entries, it also purges the saveset entry from the media database. If
>    you're saving to tape, the data is still on tape, but NetWorker no
>    longer knows it's there.

We do this with our (AFTD) clone savesets. Once it's cloned to tape, the 
next day (or two, depending), we delete the cloned saveset entries and 
reclaim disk space with nsrstage -C -V.

> 
> or
> 
> - use the `-o recyclable' option to nsrmm for each saveset, to force
>    the saveset to become recyclable, then force an index consistency run
>    with nsrim.  This is bypassing the "safety" of using nsrim -l, so you
>    can shoot yourself in the foot if you're not careful, but this is one
>    way to have a little bit more control on what savesets get purged.

Hmmm ... hadn't considered that. I can see where you'd want to make really 
sure you're marking the right savesets.

> > I was wondering if there was a "nsr-cleanup -s 'savetime>1 year'"
> > kind of command to delete savesets older than 1 year.
> 
> There is not, mostly because that kind of thing really depends on what
> schedule is in use at a site.  If you do quarterly fulls and the full
> for that quarter was 13 months ago, what should the cleanup command do?
> If it purges the 13 month old full, then it also must purge all the
> savesets that depend on it, so you lose savesets are less than a year 
old.

Why would newer savesets depend on the older, in this case? I realize I'm 
a little thickheaded and not feeling well today, but why would a full from 
9 months ago depend on a full from 13 months ago? I could understand if it 
were incrementals or differentials, but isn't each full "stand-alone", so 
to speak? Why would one full depend on an earlier full?

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