On Sat, 18 Dec 2004 10:01:41 +0000
Keith Matthews <amanda AT frequentous.co DOT uk> wrote:

> On Sat, 18 Dec 2004 09:17:30 +0000
> Keith Matthews <amanda AT frequentous.co DOT uk> wrote:
> 
> > In the light of messages just posted I'll report this in case it
> > didn't get out. Apologies to those who got it first time, I don't
> > like those who assume that no answer simply maans people don't want
> > to answer either.
> > 
> > I'm having trouble getting a remote backup to work. 
> > 
> > The report states that the disk backup failed due to a timeout.
> > Examination of /var/log/messages at the client shows that amandad
> > exited'status 1' but gives no other indication of the cause of the
> > problem. This is happening with all disks on that client. 
> > 


OK, for the sake of posterity I'd better post some more for this.

The problem seems to be related to the number of entries in the disklist
for the relevant host.

I originally had 

<hostname> wd0a user-tar -1
<hostname> wd0e comp-user-tar -1
<hostname> wd0g comp-user-tar -1
<hostname> /var/amanda user-tar -1
<hostname> /var/backups user-tar -1
<hostname> /var/clamav user-tar -1
<hostname> /var/cron user-tar -1
<hostname> /var/mysql user-tar -1
<hostname> /var/named user-tar -1
<hostname> /var/spool comp-user-tar -1
<hostname> /var/www user-tar -1

(I've replaced the real, fqdn, hostname for security reasons).

After some considerable amount of cut-and-try testing I discovered that
the system worked quite happily with just one disklist entry. Further
testing revealed that it was quite happy with as many as 7 entries, more
would cause the first few to fail, and the whole set would cause the lot
to fail.

Replacing the above with one entry per filesystem (i.e wd0a, wd0e, wd0g)
where the whole filesystem was needed, and the top level directory
(/var) for the other case, with an exclude file to eliminate the
unwanted had the whole set dumping correctly.  I have no idea if this is
a generic Amanda issue or one specific to the OpenBSD port.

Debugging was complicated by the disk entries being tried in reverse
order, something else that does not seem to be mentioned in the
documentation.

In case anyone wonders about the effect of 'inparrallel' I left it at
the default of 4.