George Sinclair wrote:

Well, I would look under saveset recover for the server and then they
should be listed under index:client at least for 6.1.1 or later, but,
yes, as near as I can tell this is normal. The server only backs up its
own index. The other client indexes get backed up when those clients
actually run their backups. There should be a '.nsr' file under each
index with something like:

+skip: * .?*

I assume the server sees all these skips and just ignores all the
indexes, but it knows to override this when backing up its own index?

No not quite. It never backs up the index directories as plain files,
even for its own index. Think of the index as a database, NetWorker has
an inbuilt module for backing them up while the database is open.

You can force backups of the other indexes by putting all clients in a
special group, e.g. 'index' and then launching backup as: 'savegrp -O -l
level group'. This will also save the bootstrap. I guess you could also
explicitly save all indexes on the server by running the save command
and forcing it to ignore the .nsr files with something like:

save -b pool -l level -i /nsr/indexa

NO you shouldn't do this. It is like backing up an open database and
recovery is not guaranteed. NetWorker will save all relevant indices as
part of normal savegrp operations, but if you want extra index backups
the savegrp -O method is valid for any client in any group.

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