On Sat, Jan 25, 2003 at 12:34:47PM -0500, John R. Jackson wrote:
> 
> The standard "trick" is to call either "amtape <CONFIG> eject" or
> (what I do) "amtape <CONFIG> slot advance" after the amdump run. For
> instance:
> 
>   amdump <CONFIG> ; amtape <CONFIG> slot advance

So I did this last night, and today all tapes are still in their drives.
I would have expected to see one of them with a nice green 'operate
handle' light :-(

> Without the amtape call, the next amdump run will do a "load current"
> (which doesn't do anything because the tape is still loaded), check
> the current tape (the one that was written "yesterday"), decide it
> does not like it and then do a "load next", which ejects current and
> goes on to the next tape. That works but, as you noted, the last tape
> from a run sits in the drive between runs.

Hmm. I've just read this again. You say a "load next" ejects the current
tape - which would imply that I should be seeing an ejected tape
eventually. This in fact never happens.

"amtape <config> eject" works fine, however.

Should I have 'needeject' set in my chg-multi.conf? It's the only way I
can see of having amanda eject a tape when moving on to the next. But
the comment in chg-multi.conf says, "If you are using multiple drives as
a tape changer, you don't want to do this..."

I'm thinking that I *do* want to do this. Would I be right?

Simon