While looking for the answer to this problem, I did a few web and
Usenet searches, but came up empty handed. This is what I did to
fix the problem. I'm posting this for the archives.
Using the standard FreeBSD package of Amanda, I ran "amcheck
standard" as the operator user and would get "host down?"
failures repeatedly when trying to just back up itself. There
was no amandad.debug file, so I couldn't determine what the
problem was from that.
Before I ran amcheck, 'netstat -a | grep amanda' returned:
tcp4 0 0 *.amandaidx *.* LISTEN
udp4 0 0 *.amanda *.*
After:
tcp4 0 0 *.amandaidx *.* LISTEN
The only indication of a problem that I was able to find was the
following message in /var/log/messages:
amanda/udp server failing (looping), service terminated
After struggling a bit while trying to figure this out, I finally
altered my disklist. Where before it had:
hostname da1a comp-root
I put
localhost da1a comp-root
This seemed to do the trick. (I still had to end up compiling
security in common-src to track down other problems.)
Perhaps in a related note, I have the the following line in my
/etc/hosts.allow file:
amandad: localhost: allow
But since this is over UDP, I don't think TCP wrappers would
matter, would they?
Anyway, I hope this helps someone. I've seen queries in the past
on the mailing list, but no "Yes! that answered my question!"
Mark.
--
ommnes exeunt in mysterium
All Things Lead into Mystery
