cannot ping both ways 1

[mightyscotchpine]

Could anyone lend me a hand?
I have a lan with one sco openserver 5.05 machine and a bunch of pcs running windows and Facetwin. All of them have only one nic and all are on the same subnet. I am having problems with one of the pcs (station2). Station2 can ping the server and every other pc on the lan. Every pc on the lan can ping station2 and the server. The server can ping every other pc on the lan, but cannot ping station2.
To top it off, station2 can connect to the server just fine with Facetwin.

Has anyone had this problem before?
I was thinking that something was screwed up with ICMP on station2, but if that were the case it wouldn't respond to any other pings, right?
Could it have something to do with filtering on the server?
We have no filters set up that I know of.

Thanks in advance for any help out there.

 

[stanhubble]

mightyscotchpine

when you ping from the server to station 2,
what is the message you receive?....
is it no route or just packet loss?
are you pinging the ip or the hostname?
are you using dns or hosts?
what does an arp -na show?

lots of questions i know but i just had a similar situation in a remote site that ping would seem to hang and you would have to break out of it. it would then show that 1 packet was successfully received but all the others were lost.
it turned out to be a dns configuration problem in the reverse lookup. It took a lot of staring at the named.conf file to figure out that a leading zero on one of the quads ( 07 instead of 7 ) broke the reverse in-addr.arpa for the subnet.

hth
stan

 

[abovebrd]

I would check the gateway setting on workstation2 as compared to your other workstations.

It almost sounds like workstation2 either has no gateway setting or is set to some other gateway that the SCO box does not see. (I have seen this senerio many times)

This is of course a shot in the dark based on what I have read

Please post back if thats not it.





-Danny

 

[mightyscotchpine]

Stan,

In answer to your questions (in order):

I just get packet loss - 100% I have only one subnet, so I wouldn't expect a "no route".
I am pinging both ip and hostname - same result.
I am using /etc/hosts - no dns.
An arp -na gives:

? (ip address) at mac address (802.3)

This mac address matches that of the nic on the pc.

Incidently, ping -m craps out too.
And an arp -S gives:
arp statistics:
10 frames sent
774 frames received
0 had a bad hardware type
0 had a bad MAC address
0 had a bad IP address
0 had our address
0 updated an existing entry
1048 lookups
290 were generated locally
758 were received
401 were for us
357 were for proxies
357 failed
0 entries went into reject state
0 entries were revived
1336 entries were expired

Is this more clear?
Thanks for the help.

 

[mightyscotchpine]

.