>No, data should be a real directory under each "file tape".
>There will also be a file called info under each "file tape".
>Have you "amlabelled" each "file tape"?
~~~~~~~~~~~
Yes, I amlabeled each one.
I moved the data symbolic link to the tape and then run amlabel.

$ln -s 01 data
$amlabel DailySet1 DailySet101
$rm data
$ln -s 02 data
$amlabel DailySet1 DailySet102
~~~~~~~~~~~~

~~~~~~~~~~~~~
Then when running amdump.

$ln -s 02 data
$amdump DailySet1
~~~~~~~~~~~~

~~~~~~~~~~~~~~
An ls of /backups/02 looks like this.

drwxrwx---   20 amanda   disk         4096 Jul 11 09:00 ..
-rw-------    1 amanda   disk           10 Jul  6 23:28
00000-DailySet102
-rw-------    1 amanda   disk        32768 Jul  6 23:28
00000.DailySet102
-rw-------    1 amanda   disk           15 Jul  7 00:31
00001-srv1.testing.com._home.0
-rw-------    1 amanda   disk     13453688832 Jul  7 00:31
00001.srv1.testing.com._home.0
-rw-------    1 amanda   disk           10 Jul  7 00:31 00002-TAPEEND
-rw-------    1 amanda   disk        32768 Jul  7 00:31 00002.TAPEEND
~~~~~~~~~~~~~~

~~~~~~~~~~~~~
I am backing up my home directory (~18GB).
So my disklist looks like this.

srv1.testing.com /home hard-disk-tar
~~~~~~~~~~~~~

~~~~~~~~~~~~~
Here are the entries within my amanda.conf

define tapetype HARD-DISK {
    comment "Hard disk instead of tape"
    length 20000 mbytes

define dumptype hard-disk-tar {
    hard-disk-dump
    comment "Back up to hard disk instead of tape - using tar"
    program "GNUTAR"
~~~~~~~~~~~~~

So, any ideas where I am going wrong?  Can I just create the data
directory now in each tape and just move the existing files in?  Then
try amrecover again.

Thank you for your help.

drew
> -----Original Message-----
> From: Jon LaBadie [mailto:jon AT jgcomp DOT com]
> Sent: Friday, July 11, 2003 9:31 PM
> To: Drew Mogan
> Subject: Re: tapeless backup...should i have symbolic links to data?
> 
> 
> On Fri, Jul 11, 2003 at 09:49:55AM -0700, Drew Mogan wrote:
> > Hello,
> > 
> > I am wondering whether or not I should be using symbolic links for
> > "data" in my tapeless setup.
> > I am currently able to run "amdump" but not able to successfully run
> > "amrecover"?
> > 
> > Within my amanda.conf
> > tapedev file:/backups
> > 
> > Directory Structure
> > /backups/01
> > ...
> > /backups/20
> > 
> > When I go to run amdump I just point "data" to the next 
> tape in line.
> > $ln -s 06 data
> > $amdump DailySet1
> 
> No, data should be a real directory under each "file tape".
> There will also be a file called info under each "file tape".
> Have you "amlabelled" each "file tape"?
> 
> > 
> > I have recently seen some docs where the data directory is actually
> > created within the tape.
> > /backups/01/data
> > 
> > Which of these methods are best for a tapeless backup setup?
> > How does this affect how I do "amrecover"?
> > 
> > I am currently having trouble with "amrecover" and am not 
> sure what I
> > should be using for "settape".
> > 
> > I have tried using the following:
> > ln -s data 02
> > (then within amrecover)
> > settatape host:file:/backups/02
> > or
> > settatape host:file:/backups
> > 
> > I have not had luck either way.  I receive this error.
> > ~~~~
> > Extracting files using tape drive file:/backups/02 on host
> > srv1.testing.com.
> > Load tape DailySet102 now
> > Continue [?/Y/n/t]? y
> > amrecover: error reading tape: Connection reset by peer
> > extract_list - child returned non-zero status: 1
> > ~~~~~
> > 
> > Thanks so much for your help.
> > drew
> > 
> > 
> >>> End of included message <<<
> 
> -- 
> Jon H. LaBadie                  jon AT jgcomp DOT com
>  JG Computing
>  4455 Province Line Road        (609) 252-0159
>  Princeton, NJ  08540-4322      (609) 683-7220 (fax)
>