Re: My two boxes will not talk to each other - now they do.

2007-01-20 Thread Andrei Popescu
On Sat, 20 Jan 2007 08:14:50 -0500
Ken Heard <[EMAIL PROTECTED]> wrote:

> However, even though the DHCP server is enabled, it appears that once 
> the gateway -- or at least mine -- initially assigns an IP address to
> a computer at random it is never changed thereafter.  This
> arrangement consequently obviates the need to assign them manually, a
> complicated procedure.

If you ever get in trouble this should be the first place to look. And
setting static is not so difficult. This is from
my /etc/network/interfaces

--
# The primary network interface
allow-hotplug eth0
#iface eth0 inet dhcp
iface eth0 inet static
address 192.168.77.77
netmask 255.255.255.0
gateway 192.168.77.1

# this is needed if you have resolvconf package installed
# otherwise comment it out and use /etc/resolv.conf
dns-nameservers 192.168.77.1
--

I commented out the line with dhcp and added the rest. Please note
that my gateway runs a caching DNS proxy (dnsmasq) so I don't need to
use the real DNS server of my ISP except on the gateway.

Regards,
Andrei
-- 
If you can't explain it simply, you don't understand it well enough.
(Albert Einstein)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: My two boxes will not talk to each other - now they do.

2007-01-20 Thread Ken Heard
Thanks to the three people who responded to my query on the subject, I 
was able to get my two boxes talking to each other.  Both Patrick 
Wiseman and Douglas Allan Tutty suggested that I not use aliases for 
each box tied to a local domain name, which I had not properly 
configured in any event.  They both suggested that I use static IP 
addresses for each box instead assigned by me, rather than dynamic ones 
assigned by the gateway through DHCP.  Mr. Tutty went further, pointing 
out errors in the /etc/hosts file in both boxes.


On the strength of this information I added to both /etc/hosts files 
arbitrarily assigned IP addresses with aliases:


192.168.0.2 SOL
192.168.0.4 LAP

(I left untouched all the other information in both those files.)

Then on the server box (SOL) I ran "/etc/init.d/nfs-server start", which 
was accepted.  I then pinged each box from the other.  The response was 
"interesting".  The SOL box returned:


From 192.168.0.114 imcp_seq=1 Destination Host Unreachable

The LAP box returned

From 192.168.0.142 imcp_seq=1 Destination Host Unreachable

   	I consequently concluded that 192.168.0.114 was the current address 
of the SOL computer and that 192.168.0.142 was that of the LAP computer. 
 I changed the two /etc/hosts files to conform to these addresses and 
again pinged each box from the other.  This time the returns included 
the current IP addresses of both, together with their aliases: LAP 
(192.168.0.142) and SOL (192.168.0.114).


   	Still using the aliases -- now linked to IP addresses -- I once 
more ran on the SOL computer as root "/etc/init.d/nfs-kernel-server 
start", which was accepted.  Finally, as root on LAP I ran as root 
"mount -t nfs SOL:/home /home", which worked.  The /home directory on 
SOL thereby became /home directory on the laptop -- a temporary 
arrangement set up for testing -- there is nothing yet in the /home 
directory on the LAP.


	I then rebooted both boxes to see if these IP addresses were static. 
As the same ones reappeared after rebooting I assumed that they were 
indeed static.


   	After I had done all the foregoing I received a third answer to my 
query, from Ivan Tiliatnikov.  As this one was off list I quote it in full:

-
To find a real IP address of the host use ifconfig command.

For example it shows that my laptop has an ip address

eth0  Link encap:Ethernet  HWaddr 00:12:3F:EA:0B:07

  inet addr:192.168.1.10  Bcast:129.78.236.255 Mask:255.255.255.0
  =

  inet6 addr: fe80::212:3fff:feea:b07/64 Scope:Link
  UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
  RX packets:72991 errors:0 dropped:0 overruns:0 frame:0
  TX packets:35417 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:1000
  RX bytes:80972369 (77.2 MiB)  TX bytes:4090925 (3.9 MiB)
  Interrupt:169

The IP address is assigned either statically or dynamically via DHCP
( from your router ). In my case it is static so that I do not have to
change /etc/hosts all the time. Some of the routers can be configured
give fixed IP addresses to known hosts. Check if yours can do it. If
answer is yes, configure it to give your machine a static address e.g.
192.168.1.10. You will need MAC address of you network card to do that.
"ifconfig" command will help you again, see HWaddr feild ( HWaddr
00:12:3F:EA:0B:07 ).

Here is my setup I have two machine a desktop and a laptop.

laptop: 192.168.1.10
desktop: 192.168.1.11

In my LAPTOP  /etc/hosts a have an entry

127.0.0.1 localhost.localdomain localhost
192.168.1.10 mylaptop

On my desktop /etc/hosts is
127.0.0.1 localhost.localdomain localhost
192.168.1.11 mydesktop

If both machines are on and ifconfig shows that IP addresses are
correctly assigned, try

>From 192.168.1.10 try diagnostic command:

ping 192.168.1.11
or mydesktop

It will show you that traffic is reaching the second machine.

Press Ctrl-C to stop.

Hope this help.
--
I had not known about the ifconfig command.  I ran it on both boxes.  It 
confirmed those IP addresses and aliases, and much else besides.


Mr. Tiliatnikov suggested that I set up the gateway to assign to each 
box a static IP address, rather than allow DHCP on the gateway to assign 
a dynamic address, which presumably would be different each time each 
box is booted.


My gateway is a D-Link DI-704. Looking at its configuration I found that 
the connection to the ISP was PPPoE, and that the DHCP server was 
enabled, with the starting and ending addresses for the IP pool set at 
192.168.0.100 to 192.168.0.199 respectively.


However, even though the DHCP server is enabled, it appears that once 
the gateway -- or at least mine -- initially assigns an IP address to a 
computer at random it is never changed thereaft

Re: My two boxes will not talk to each other

2007-01-16 Thread Douglas Allan Tutty
On Tue, Jan 16, 2007 at 06:09:18PM -0500, Ken Heard wrote:
> I have a P4 desktop, on which is Sarge installed, host name SOL; and a 
> P2 laptop (Toshiba Tecra 8000), host name LAP, on which yesterday I 
> successfully installed Etch RC1.  Both can communicate with the outside 
> world through a gateway and both can communicate with the printer 
> through the print server.  I cannot however get them to communicate with 
> each other.
> 
> I am trying to use the host names to identify each box to the other.  I 
> set SOL up as the server, but when I start nfs-kernel-server, it 
> returns: "exportfs: LAP has no inet address."  LAP is the client, but 
> when I try to mount on it the directory I want to export from SOL, the 
> mount command returns "mount: can't get address for SOL."
> 
> I assume that neither computer can resolve the host name of the other. 
> Running cat /etc/hosts on SOL returns the following:
> 
> 127.0.0.1   localhost.localdomain   localhost   SOL
> 
 
> Running the same command on LAP returns the following:
> 
> 127.0.1.1 localhost
> 127.0.0.1 LAP.localhost   LAP
> 
 
> Have I given enough information to indicate what is wrong?  Can anybody 
> advise me what to do to put these two boxes on friendly terms with each 
> other?
 

With this /etc/hosts, there's no way for one box to know about the
other.  They have to be on the same network.  How are they connected and
what is each ip address according to the gateway?  Eg: if the gateway is
192.168.1.1, SOL is 192.168.1.2, and LAP is 192.168.1.3 then this should
be reflected in the /etc/hosts file.  Note that having:

127.0.0.1   localhost.localdomain   localhost   SOL

is not correct.  It should be:

127.0.0.1   localhost.localdomain   localhost
192.168.1.2 SOL

The first step to seeing if there is a route is to try using ping

from SOL: ping LAP
from LAP: ping SOL

Until this works, don't worry about NFS.

If you still have trouble, then send more specific info on how things
are connected.

Good luck.

Doug.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: My two boxes will not talk to each other

2007-01-16 Thread Patrick Wiseman

On 1/16/07, Ken Heard <[EMAIL PROTECTED]> wrote:


I have a P4 desktop, on which is Sarge installed, host name SOL; and a
P2 laptop (Toshiba Tecra 8000), host name LAP, on which yesterday I
successfully installed Etch RC1.  Both can communicate with the outside
world through a gateway and both can communicate with the printer
through the print server.  I cannot however get them to communicate with
each other.

I am trying to use the host names to identify each box to the other.  I
set SOL up as the server, but when I start nfs-kernel-server, it
returns: "exportfs: LAP has no inet address."  LAP is the client, but
when I try to mount on it the directory I want to export from SOL, the
mount command returns "mount: can't get address for SOL."

I assume that neither computer can resolve the host name of the other.



Give each a fixed address (i.e. don't have them get it through DHCP) and
then put both in each /etc/hosts file.  Eg:

192.168.1.10SOL
192.168.1.11LAP

I think that should do it.  It works for me anyway!

Patrick


My two boxes will not talk to each other

2007-01-16 Thread Ken Heard
I have a P4 desktop, on which is Sarge installed, host name SOL; and a 
P2 laptop (Toshiba Tecra 8000), host name LAP, on which yesterday I 
successfully installed Etch RC1.  Both can communicate with the outside 
world through a gateway and both can communicate with the printer 
through the print server.  I cannot however get them to communicate with 
each other.


I am trying to use the host names to identify each box to the other.  I 
set SOL up as the server, but when I start nfs-kernel-server, it 
returns: "exportfs: LAP has no inet address."  LAP is the client, but 
when I try to mount on it the directory I want to export from SOL, the 
mount command returns "mount: can't get address for SOL."


I assume that neither computer can resolve the host name of the other. 
Running cat /etc/hosts on SOL returns the following:


127.0.0.1   localhost.localdomain   localhost   SOL

# The following lines are desirable for IPv6 capable hosts
::1 ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts

Running the same command on LAP returns the following:

127.0.1.1   localhost
127.0.0.1   LAP.localhost   LAP

# The following lines are desirable for IPv6 capable hosts
::1 ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts

Have I given enough information to indicate what is wrong?  Can anybody 
advise me what to do to put these two boxes on friendly terms with each 
other?


Ken Heard
Toronto, Canada


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]