On 10/03/2011 04:18 PM, Brian Cuttler wrote:
Jean-Louis,
thank you for your help, I don't quite have it though.
The default changed, so you must add '-auth=bsd' if you want to use the
bsd auth.
For bsd auth:
amanda dgram udp wait amanda /usr/local/libexec/amanda/amandad
amandad -auth=bsd amdump
for bsdtcp auth:
amanda stream tcp nowait amanda /usr/local/libexec/amanda/amandad
amandad -auth=bsdtcp amdump
Also add the auth in the dumptype.
Increase 'debug_auth' to 9 in amanda.conf and amanda-client.con, then
look in the debug files, check to be sure the server and client use the
same auth
Jean-Louis
On the client.
amanda stream tcp nowait amanda /usr/local/libexec/amanda/amandad amandad
-auth=bsdtcp amdump
On the Server.
grifserv /gmultip/climsgl zfs-snapshot-bsd
grifserv /gmultip/csssoft zfs-snapshot2
grifserv /gmultip/dew zfs-snapshot-bsdtcp
grifserv /gmultip/encphrev zfs-snapshot
Do not do that, amanda do not work with different auth for the same
client, it use the auth randomly from one of the dle.
Since you want to use bsdtcp, set all dle to use zfs-snapshot-bsdtcp
With the following amanda.conf dumptypes
define dumptype zfs-snapshot {
index
program "APPLICATION"
application "app_amgtar"
script "script_zfs_snapshot"
# auth "bsdtcp"
property "DF-PATH" "/usr/sbin/df"
property "ZFS-PATH" "/usr/sbin/zfs"
property "PFEXEC-PATH" "/usr/bin/pfexec"
property "PFEXEC" "YES"
estimate server
}
define dumptype zfs-snapshot-bsd {
index
program "APPLICATION"
application "app_amgtar"
script "script_zfs_snapshot"
auth "bsd"
property "DF-PATH" "/usr/sbin/df"
property "ZFS-PATH" "/usr/sbin/zfs"
property "PFEXEC-PATH" "/usr/bin/pfexec"
property "PFEXEC" "YES"
estimate server
}
define dumptype zfs-snapshot-bsdtcp {
index
program "APPLICATION"
application "app_amgtar"
script "script_zfs_snapshot"
auth "bsdtcp"
property "DF-PATH" "/usr/sbin/df"
property "ZFS-PATH" "/usr/sbin/zfs"
property "PFEXEC-PATH" "/usr/bin/pfexec"
property "PFEXEC" "YES"
estimate server
}
The property should go in script section for script_zfs_snapshot
On the server I executed
# amcheck -c curie grifserv
- and then one of the DLEs listed above, ran the test with
only the single DLE.
All produced the same output on the client # snoop curie
curie.wadsworth.org -> grifserv.wadsworth.org UDP D=10080 S=941 LEN=137
grifserv.wadsworth.org -> curie.wadsworth.org ICMP Destination unreachable
(UDP port 10080 unreachable)
I'd have thought one of those combinations was correct.
Apparently I am not activating the service on the client, I am
NOT producing any files in /tmp/amanda/
The SMF service remains online, I am not erroring it out, nor
am I producing smf errors in the system messages file.
I don't know how solaris works.
Jean-Louis
