HI again ,
First at all I want to thanks all the sun guys who responded to my
question , some mentions is necessary before I post my summary :
-documentation for configuring Solstice BACKUP is very poor (Chapter
:Configuring SB to use 2 network interface contains just 50 words )
- after you try to set the second interface by default when you
start SB it use the first one even if in /etc/init.d/networker file is
specified to start
sever with the second interface name "nsrexecd -s second interface "
, when you looking with nwadmin to the sever name , surprise .... is the
first network name
- if you looking in configuration files /nsr/res/* all the IP
addressees you can see is just first network , so it doesn't meter if
you specified in installation script all interfaces name or
just the second .
Type the ip address of the second interface into the field.
Sr. Systems/Network Administrator
(425) 709-2900 Ext. 13
I don't know if you got an answer on this yet or not....
When setting up the client for the secondary interface, you
need to view "Details" under the client setup, then set the
Networker Server Interface to the name of the server's
the interface needs its own name
I use a different mount point,
works fine that way
Richard Bond (firstname.lastname@example.org (206) 605-3561
System Administrator K-351, Health Sciences
Department of Molecular Biotechnology Box 357730
University of Washington Seattle, WA 98195
You did not mention if you are putting in a second interface
on the same physical network - i.e., another connection
to the same switch or router.....and the IP address/netmask/network
The network number is the IP address - bitwise & with the netmask.
If you do nothing, the default netmask is used, depends on the IP
0.0.0.0 to 127.0.0.0 is class A (I think)
22.214.171.124 through 126.96.36.199 are 64 * 256 different class class B
192.0.0.0 through 254.255.255.0 are 64*256*256 different class C
I have 4 machines that are on multiple networks.
Host A 188.8.131.52 and 184.108.40.206....netmask on second
interface is 255.255.255.0
Host B 220.127.116.11 and 18.104.22.168....netmask on second
interface is 255.255.255.0
Therefore, I have 2 logical networks - even though they are on the same
22.214.171.124 is my primary network
126.96.36.199 is my secondary network.
I can control / force traffic that is between 188.8.131.52 and
184.108.40.206 to use
the second ethernet port. However, when I receive messages from
on that interface (220.127.116.11), the return packets go out of the
interface...and solaris knows that 18.104.22.168 network is on the primary
Caveat - this may be oversimplified...and some precise details of the
network classes may be off.
Note, having different netmasks on the public network might cause you to
calls from network engineering. In the future, we will probably NOT do
we will do things a bit differently...eg, have 2 different networks
10.20.4 and 10.20.5, for example,
with the same netmask on both - still not the default, but at least
Your clients must indicate which interface they talk to:
# nsrexecd -s <second_interface>
At my last work I do this by following:
not used direct Solstice Backup save command
write script which run save command with another arguments/
one of its ( first or second, i don't remember) is adress of server
the command in script look like:
save -s <your new adress> <all another save arguments from Backup>
You can try the following.
1) Make sure that both of your network cards ip addresses are in the
2) Edit the following 2 files
3) look for a line that says
(//usr/sbin/nsr/nsrexecd -s HostName) > /dev/console 2>&1
All this line does is startup Solstice Backup with HostName as the
Replace the HostName with the hostname of your 2nd interface card.
I've never done this but I suspect you need to do the following;
confgure your 2nd interface with a different hostname; e.g
Install Solstice Backup and when prompted for the server name use
servername.hme2. I'd be interested to know whether this works.
It's not so much a SBU configuration issue as it is a routing issue.
sure the routes to and from the backup server and clients are via your
backup net. I use DNS for all with the exception of host file entries
pre-empt DNS in the case of SBU server and client communication.
LDN . . .
In the finaly solution .... I modified in /etc/init.d/neworker
"/usr/sbin/nsr/nsrexecd -s HostName) > /dev/console 2>&1"and I replace
in /nsr/res/* files all references about
SERVER , IP address with the name and IP address of the second
Thanks to all.
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:14:05 CDT