networking 2 Solaris 8 x86 PCs

[geoelder]

I have two Intel PCs running Solaris 8 and am trying to network them together for training purposes. During installation on each system, I selected "non-networked" system because I didn't know what to enter for IP address and subnet (and still don't). I had read that the files could be manually edited after installation. I am not sure which files need to be created or modified for this simple network. Did some reading and this is what I have done so far. Thanks, I know I am asking alot.

intel166/
/etc/hostname.elxl

/etc/nodename
intel166

/etc/defaultdomain (no entries)

/etc/inet/hosts
127.0.0.1 localhost loghost intel166
( ?IP address? intel120)
( ?IP address? intel166

/etc/defaultrouter (no entries)

/etc/hosts.equiv
intel120
intel166

the intel120 machine is set up the same with the exception of the /etc/nodename

Thanks again

 

[jerryshao]

The /etc/hostname.elxl on intel120 has to be intel120.

What message can you get if you type "ifconfig -a"?

 

[geoelder]

Thanks for the quick response. I have corrected the /etc/hostname.elxl. I am not sure if I renamed the /etc/hostname.interfaces file properly either. There were two different entries under /devices. elxl and elxl0.
I dont belive I want the pseudo entry of elx. But, really dont know.

Now, the ifconfig -a output for intel120and intel166:
lo0: flags=1000849<up,loopback,running,multicast,IPV4> mtu 8232 index 1 inet 127.0.0.1 netmask ff000000

 

[jad]

have you rebooted?

if you don't want to reboot you can type:
on intel120

ifconfig elxl0 plumb
ifconfig elxl0 intel120 up

on intel166

ifconfig elxl0 plumb
ifconfig elxl0 intel166 up

if you type 'ifconfig -a' and 'netstat -r' you should notice the interface started and the route added.

 

[geoelder]

Yes, I have been rebooting each time I make a change. These are the results of running ifconfig and netstat -r on intel120 and intel166.

intel120:

ifconfig elxl0 plumb
(no response)

ifconfig elxl0 intel120 up
ifconfig: setifflags: SIOCSLIFFLAGS: elxl0: cannot assign requested address

netstat -r

Routing Table: IPv4
Destination Gateway Flags Ref Use Interface
------------------------------------------------------------
224.0.0.0 localhost U 1 0 lo0
localhost localhost UH 17 2637 lo0

ifconfig -a

lo0: flags=1000849<UP,loopback,running,multicast IPv4,>mtu 8232 index 1 inet 127.0.0.1 netmaskff000000

elxl0: flags=1000842<Broadcast,running,multicast,IPv4>mtu 1500 index 2 inet 127.0.0.1 netmask ff000000 broadcast 127.255.255.255 ether 0:60:97:bf:d6:eb


intel166:

ifconfig elxl0 plumb
(no response)

ifconfig elxl0 intel120 up
ifconfig: intel166: bad address

netstat -r

Routing Table: IPv4
Destination Gateway Flags Ref Use Interface
------------------------------------------------------------
224.0.0.0 localhost U 1 0 lo0
localhost localhost UH 16 792 lo0

ifconfig -a

lo0: flags=1000849<UP,loopback,running,multicast IPv4,>mtu 8232 index 1 inet 127.0.0.1 netmask ff000000

elxl0: flags=1000842<Broadcast,running,multicast,IPv4>mtu 1500 index 2 inet 0.0.0.0. netmask 0
ether 0:60:97:b7:8a:8d

Thanks,
George

 

[jad]

ls /dev/el*
ls /dev/ip*
ls /dev/sm*

please

 

[Annihilannic]

You need to assign the IP, and netmask before &quot;up&quot;ing the interface.

[tt]ifconfig elxl0 1.2.3.4 netmask 0xfffff00 up[/tt]

1.2.3.4 and 0xffffff00 are of course example values.

Annihilannic.

 

[jad]

well ... that is what the intel120 and intel166 should have done ...

if no netmask is specified it works in out according to the class of the address > 126.0.0.0 and it is a class c address (255.255.255.0) by default

can't remember the other breaks though

 

[geoelder]

intel120:
ls /dev/el*
/dev/elx /dev/elxl /dev/elxl0

ls /dev/ip*
/dev/ip /dev/ip6 /dev/ipd /dev/ipdcm /dev/ipdppt
/dev/ipsecah /dev/ipsecesp

ls /dev/sm*
/dev/sm*: No such file or directory

intel166:
ls /dev/el*
/dev/elx /dev/elxl /dev/elxl0

ls /dev/ip*
/dev/ip /dev/ip6 /dev/ipd /dev/ipdcm /dev/ipdppt
/dev/ipsecah /dev/ipsecesp

ls /dev/sm*
/dev/sm*: No such file or directory

I am guessing that an ip address of 1.2.3.4 for intel120 and 1.2.3.5 for intel166 will be ok. But, I am uncertain about the exactly what to enter in /etc/netmasks

 

[jad]

umm ... remove intel166 from 127.0.0.1 in /etc/hosts

and try again.

 

[geoelder]

Removed the intel166 and intel120 entries from the /etc/hosts.

Modified etc/inet/hosts on intel120 to include intel166 and 1.2.3.5

Modified etc/inet/hosts on intel166 to include 1.2.3.4 on intel120

Tried upping the ip addresses using:
ifconfig elxl0 1.2.3.4 netmask fff00000 ( I created the netmask)
ifconfig fff00000: bad address

Didn't know what to use for the netmask, so i created my own. Bad idea. What is an acceptable netmask? Do I use the same one on both machines? And, are my made up ip addresses ok?

Thanks,
George

 

[Annihilannic]

You need a 0x on the beginning if you're going to use hexadecimal format for the netmask. Alternatively try 255.240.0.0. Normally they are 255.255.255.0 unless you are going to have a lot of machines (i.e. more than 250) in the subnet.

Annihilannic.

 

[geoelder]

Ok, that did the trick. I changed the netmask to 255.255.255.0. I can now telnet and rlogin between the two machines. Just one problem. Each day when I power up the two machines, I have to do the ifconfig elxl0 intel166/intel120 up on both machines. Any ideas? I can deal with this if I have to. It's not that big a problem.

Thanks for the help,
George

 

[jad]

you power the machines down every day?

do you have the file '/etc/hostname.elxl0' ?

if you don't then create it and put the hostname in it, with the appropriate hostname for each host in it? (on intel166 the file has 'intel166' in it)

reboot and it should work ...

 

[geoelder]

I power the two intel pcs down to save energy and wear and tear. They are old units with p-120 and p-166mx processors.

my /etc/hostname.elxl0 was correct, but your suggestion prompted me to go back and check my other files. And sure enough, my /etc/inet/hosts files on both machines were wrong. The intel120 and intel166 didnt have entries for themselves. Corrected that, rebooted and I was able to connect without any trouble. These systems will help me train for the second Solaris test.

Thanks to jerryshao, jad and Annihilannic

 

[jad]

the most wear and tear on PC's generally happens when you power them on and off ... you get power spikes and other things ... sometimes it is false economy to turn them off

Just saying