"Jordan K. Hubbard" wrote:
> > Why is it clearly broken? proto=tcp,vers=3 is what is in 3.0-RELEASE,
> > Amd in 3.0 works for many. I won't defend that the new Amd works the
> > best with us, but then neither did the old Amd.
>
> Erm, I haven't tried it between 3.0 and 3.0 boxes because all my test
> environments currently involve one of each (4.0 and 3.0), but I can
> certainly say that in none of these test environments does amd work at
> all. On freefall, for example, it's really simple to demonstrate the
> error. First, we start amd:
>
> # amd -a /net -c 1800 -k i386 -d freebsd.org -l syslog /host /etc/amd.map
> # ps ax | grep amd
> 9127 ?? Is 0:00.01 amd -a /net -c 1800 -k i386 -d freebsd.org -l syslo
g
> <ok, it's running>
> # cat /etc/amd.map
> /defaults type:=host;fs:=${autodir}/${rhost};rhost:=${key}
> * opts:=rw,grpid,resvport,vers=2,proto=udp,nosuid,nodev,intr
> <looks reasonably stock>
I added the vers=2 and proto=udp, "just in case", but it wouldn't work
either way. At one point there was: fs:=${autodir}/${rhost}/host, but
that didn't work either (as /host/hub or /host/hub/host).
> # ls /host/bento
> ls: /host/bento: No such file or directory
>
> But no workee at all. Any ideas? Using only FreeBSD boxes and
> udp/ver2 mounts, amd fails to function as far as I can tell.
It would appear the the NFS interface between the kernel and amd is
working otherwise the accesses to /host would hang. But beyond that I
don't know.
Cheers,
-Peter
To Unsubscribe: send mail to majord...@freebsd.org
with "unsubscribe freebsd-current" in the body of the message
