Very Computer

Just a question, why can't I ping outside my segment from my SCO-unix
with os openserver 5.x.x, have I forgotten something in my config-files,
domain,nameserver,gateway or any proxy-server stopping me ?????
mvh Rolf
:-) :-)

: Just a question, why can't I ping outside my segment from my SCO-unix
: with os openserver 5.x.x, have I forgotten something in my config-files,
: domain,nameserver,gateway or any proxy-server stopping me ?????

I don't know your exact configuration, but two things come to mind:

  1.  You haven't set a default route to get outside your subnet.

  2.  Your provider doesn't have their routing correct to route back
      to you.

Bill
--

UUCP:              camco!bill   2835 82nd Avenue S.E. S-100
FAX:           (206) 232-9186   Mercer Island, WA 98040; (206) 947-5591
http://www.celestial.com/
SPEED COSTS MONEY -- HOW FAST DO YOU WANT TO GO?

: : Just a question, why can't I ping outside my segment from my SCO-unix
: : with os openserver 5.x.x, have I forgotten something in my config-files,
: : domain,nameserver,gateway or any proxy-server stopping me ?????

: I don't know your exact configuration, but two things come to mind:

:   1.  You haven't set a default route to get outside your subnet.

:   2.  Your provider doesn't have their routing correct to route back
:       to you.

I have a similar problem:

We are running 8 machines on a token ring
network. Two of them with OSR5, the rest ODT3.2v4.2.

The gateway to the rest of the net is via a ACC router??? and for the
ODT machines this has been working fine for over a year.

The OS5 systems seem to have trouble accessing this gateway, but it
is on a seemingly random basis. At times this route will ping and at
other times not. There is no correlation between the two machines - one
may ping while the other one doesn't.

The only surefire way to correct the problem for a while is to reboot.

I have tried swapping cards with the ODT machines, but it makes no difference.

Can anybody suggest what I should be looking for or what steps I should
take to find out whats going on?

Any help would be much appreciated.

Cheers
Terry Purdon

....
: The OS5 systems seem to have trouble accessing this gateway, but it
: is on a seemingly random basis. At times this route will ping and at
: other times not. There is no correlation between the two machines - one
: may ping while the other one doesn't.

The last time I ran into a problem like this, somebody was testing PCs
on the network and had set the IP address the same as the Unix server.
I finally figured out what was happening when I noticed that the arp
overwritten messages on the screen of my luggable in the corner.  It
seems that whenever they booted the PC with the duplicate IP address
it would make the server invisible to anything on the network.  If I
did something like an 'arp -d' to get rid of the arp entry for the
router or whatever it was that I couldn't talk to it would apparently
send out the correct arp info for the server so everything would work
again until the PC got rebooted.

Bill
--

UUCP:              camco!bill   2835 82nd Avenue S.E. S-100
FAX:           (206) 232-9186   Mercer Island, WA 98040; (206) 947-5591
http://www.celestial.com/
SPEED COSTS MONEY -- HOW FAST DO YOU WANT TO GO?

Quote:

>I have a similar problem:

>We are running 8 machines on a token ring
>network. Two of them with OSR5, the rest ODT3.2v4.2.

>The gateway to the rest of the net is via a ACC router??? and for the
>ODT machines this has been working fine for over a year.

>The OS5 systems seem to have trouble accessing this gateway, but it
>is on a seemingly random basis. At times this route will ping and at
>other times not. There is no correlation between the two machines - one
>may ping while the other one doesn't.

>The only surefire way to correct the problem for a while is to reboot.

I had the same problem. I think, there is a bug in SCO with the
arp-table. The arp-table is neccessary to know the corresponding
hardware adresses to the ip adresses. The arp-entries expire from time
to time.  If you have a machine with a shorter expiry time than your
SCO system, then the SCO bug shows up: your SCO machine doesn't
respond to the arp-request (tell me your hardware adddress, please!)
of the other machine (your router) because the other machine's
hardware address ist still in the SCO's arp table. SCO thinks "I know
you so you should know me!".

If the problem shows up then do "arp -a" on the other machine. This
shows the arp table. Probably there's no entry or a 0.0.0.0. entry for
your SCO machine, so your router doesn't know where to send the
packets back.

My workaround is the following script started by root every minute via
crontab:

----------------------------------------------------------------
#!/bin/sh

# Loeschen aller arp-Eintraege, damit Anfragen von anderen
# Rechnern ordentlich beantwortet werden.

for i in  `/etc/arp -an |\
           /usr/bin/awk '{print(array[split($0,array,"(")])}' |\
           /usr/bin/awk '{print(array[split($0,array,")")-1])}'`
do
  /etc/arp -dn $i
done

----------------------------------------------------------------

It's a brute force method. It clears SCO's arp table every minute. It
works for me.

--
                 .o.
                /   \

               |     |
              /_______\
                   `o