Sorry, for the long post, but I need to give a clear picture of
the problem I am facing

I know you all have seen this message many times here. But still I
have this related problem : I am running on a Class A network
(actually our network is quite small but it is a long story) with
netmask 255.255.255.0. I used to have no problem accessing my machine (which is connected through a hub and then through a switch) to my servers.But recently, I have not been able to ping the servers. My configurations have remained the same. I am running Red Hat 7.2 and Win98. My NIC is a Linksys 10/100 Card. Testing on both of these platforms confirm that I cannot ping the servers.

To eliminate any LAN problems in the equation, I have tested out with a cross cable and link it to another machine. It shows that I can ping from my Linux to a Win98 machine. Cabling problem from my machine is also eliminated as I have tested it with another machine and it works fine.

Doing a dmesg:
Linux Tulip driver version 0.9.15-pre6 (July 2, 2001)
PCI: Found IRQ 10 for device 00:13.0
eth0: ADMtek Comet rev 17 at 0xc4867000, 00:04:5A:49:F4:F0, IRQ 10.

Doing a tcpdump that monitors my host server:
./tcpdump host 16.153.64.22
tcpdump: listening on eth0
17:03:23.212042 arp who-has 16.153.64.22 tell 16.153.64.118
17:03:24.212042 arp who-has 16.153.64.22 tell 16.153.64.118
17:03:25.212042 arp who-has 16.153.64.22 tell 16.153.64.118

I tried other protocols such as ftp and telnet and no success.
I can also confirm that there are no firewalls for these servers.

The question here is where exactly is the problem ?
a) My motherboard ?
b) The hub (Intel Express 330T) ?
c) The switch (Intel Express 460T)?
d) Or something else ?

Any help will be appreciated !

Well try this.

Make sure that the netmask on all the machines is 255.0.0.0
After that is what it is suppose to be if your not subnetting.
Aother thing what is the server operating system that you cannot ping? Is it running a firewall?

--tarballedtux

Are you sure tulip is the right drive? I thought Linksys used
a generic ne ? tulip is for card that use DEC chip sets?

Can you ping the interfaces fron a local login by ip and hostname?

We are using subnetting therefore the different
netmask value. The servers are running on Digital
AlphaServer and Linux. Nope no firewall present.

I remember reading from somewhere that Linksys
uses Amtek chipset. I think it should be correct.
Anyway this driver was autodected by Red Hat 7.2 ...

What do you mean "ping the interaces from a
local login by ip and hostname ?"

login on to Linux

ping localhost
ping ip_address
ping hosthame

ifconfig show's that the interface is up?
Have you tryed swapping out nic cards?

It is strange. My networking is now working after
rebooting my set. I am really puzzled by these.

Most linksys nics do use DEC chipsets and use the tulip driver to clear that up. If they don't, I would have no idea why all 6 computers on my network at home don't work using the tulip module along with linksys cards.

Did you ever try bringing down the ethernet card and back up with the ifconfig eth0 down then up ? I have found sometimes when one of my machines starts acting funny like that, this usually clears this up. Sometimes I have to re-enter the route command as well, route add default gw 192.168.1.1 would fix it as well. Usually if those didn't work, I would reboot to get it going again at times. I think its just the magic of a reboot that fixes problems.