Problem reaching router and internet from fedora core 5

[mala@fedora]

Hi everyone!

I am trying to help a friend to set up up fedora core 5 on his computer, however I have reached a bizarre problem which I am not able to resolve. The problem is that we can't reach the router from fedora not even ping it, however if we use dhcp the router gives the computer an ip-number, but we can still not reach it. We have also tried with two different routers with the same problem so it probably isn't the router. We have also tried connecting the computer to another computer and then we could ping each other without any problem. I don't believe it is the router since we have tried two different ones, anyway here are our settings, the routers ip adress is 192.168.1.254 and the rest is

/sbin/ifconfig

eth0 Link encap:Ethernet HWaddrs 00:08:A1:02:90:6E
inet addr:192.168.1.50 Bcast 192.168.1.255 Mask:255.255.255.0
and so on.............

/sbin/route -n

Destination Gateway Genmask Flags Iface
192.168.1.0 0.0.0.0 255.255.255.0 U eth0
169.254.0.0 0.0.0.0 255.255.0.0 U eth0
0.0.0.0 192.168.1.254 0.0.0.0 UG eth0

Please help us before I go insane and go back to windows...... :S

Thx in advance!!!!!!!!!!!

/mala

[jbannon]

Try turning the firewall off and then pinging by IP address.

[mala@fedora]

The firewall has never been on, and I have tried with SElinux both on and off

[jbannon]

OK, start from the basics and go from there.
a. Can you ping the loopback?
b. Can you ping your own IP address?
c. Check that the netmask of the machine and the gateway are the same.
d. Can you ping the machine from the router?
e. Can you telnet into the router from any machine?
f. Is there a wireless interface on the router? Disable it if it has.
g. Are there any other protocols running? Disable them if there are.

How are you connecting to the router? You say you can ping between two machines. Is this connection point-to-point or via a switch? If it's point-to-point, try replacing the cable with a straight-through. If not and you're connecting directly to the router gateway interface use a cross-over.

Try forcing it by adding a route to the table direct to the gateway or try clearing the *nix routing table altogether (strictly speaking it shouldn't be needed as any message not destined for the local network goes via the gateway anyway).

The rest of the stuff looks right. I take that eth0 is UP & RUNNING?

[mala@fedora]

First of all the computer is a dual-boot computer and the network works fine under windows XP :S

a) Yes I can ping the loopback
b) Yes I can ping my own ip
c) The netmask is the same 255.255.255.0
d) The router doesn't the possibilty to ping machines on my network as far as I know, however I can't ping the computer from another computer on the network
e) Yes I can telnet in to the router from another machine on the network
f) Yes there is and we have disabled it, but it still doesn't work
g) No there isn't

I am connecting to the router through a straight cable.
When I can ping between two computers the connection is point-to-point with a straight cable


How do you mean adding a route to the gateway?

Thx very much for all the help!!!!!

[ccrvic]

eth0 Link encap:Ethernet HWaddrs 00:08:A1:02:90:6E
inet addr:192.168.1.50 Bcast 192.168.1.255 Mask:255.255.255.0
and so on.............

Run ifconfig again and post the rest of the results for eth0 - I'm particularly interested in the numbers of packets sent and received.

Also run "mii-tool eth0"

Have you got Ethereal installed? if not, install it & run a capture on eth0 (tethereal might be better for this). Show us what activity you get (if any).

Vic.

[jbannon]

How do you mean adding a route to the gateway?

Use the route command as superuser like this:

route -v add -host <gateway ip address> netmask <gateway netmask> dev eth0

This will force the set-up of a static route between eth0 and the gateway.

I'd also be interested in seeing the output of something like ethereal or tcpdump to see if any traffic is actually getting out of the interface.

[mala@fedora]

Here is the information you wanted

eth0 Link encap:Ethernet HWaddr 00:08:A1:02:90:6E
inet addr:192.168.1.50 Bcast:192.168.1.255 Mask:255.255.255.0
inet6 addr: fe80::208:a1ff:fe02:906e/64 Scope:Link
UP BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:9 errors:660 dropped:0 overruns:0 frame:0
TX packets:27 errors:7 dropped:0 overruns:0 carrier:7
collisions:0 txqueuelen:1000
RX bytes:1429 (1.3 KiB) TX bytes:3160 (3.0 KiB)
Interrupt:17 Base address:0xac00

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:1577 errors:0 dropped:0 overruns:0 frame:0
TX packets:1577 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:2198201 (2.0 MiB) TX bytes:2198201 (2.0 MiB)

from mii-tool eth0 I get

eth0: negotiated 100baseTx-FD, link ok


I installed ethereal and I get nothing if I ping the router or myself however if I ping the computer from another computer I get a ARP package with destination broadcast and the info Who has 192.168.1.50? Tell 192.168.1.91 (50 is the computer that I am trying to repair and 91 is my laptop connected to the router)

Thx once again for your time!

[ccrvic]

RX packets:9 errors:660
TX packets:27 errors:7

That's a lot of errors. Are you sure the system is mechanically OK? This sort of thing often means a cable fault.

Vic.

[mala@fedora]

The system works fine with windows XP so I doubt it is a mechanically problem

[mala@fedora]

ok someting weird just happend I pulled the plug and inserted it again and then I could ping the router 115 times and then it shut down again. If I then pull the plug and enter it again the same thing happens but I could ping 118 times....... I seriously don't have a clue what it could be :-( it is like a timer that shuts down

[jbannon]

Looks like you have an interface fault.

Check the cable first. Make sure it's the correct type of cable. It's difficult for us to ascertain what it should be as we don't have a diagram. Here's a simplified summary:

Host - Host :- Crossover (1 & 3, 2 & 6 crossed).
Host - Switch :- Straight through.

(Look it up under How Stuff Works or Cisco web sites).

To check this, turn both ends of the cable face up (lugs facing away from you) and check - pin 1 will be on the left as you look at the cable and pin 8 will be on the right.

The next thing it could be is the NIC itself. If, as you say, this is working under XP then it should be OK.

The next thing it could be is the driver. Check what your card is recognised as under XP (I assume it wil be plug'n play) and make sure you have a similar type of driver installed for Linux. There is a problem with early FC5 kernels recognising Tulip cards if I recall correctly.

[mickwombat]

Hi,

I seem to be having the same problem, although it was actually working a while ago. When I try to ping I get the message,

ping: sendmsg: Operation not permitted

The computer has an IP address and everything but somehow need to 'enable' ping. Have tried to disable Firewall and now just rebooting with SELinux turned off.

[mala@fedora]

Hi everyone!

I have finally fixed it :D

It was probably a driver problem, or something like that and since network cards are so cheap I went down and bought a new one.......

The old one that I had issues with was a Davicom card and I bought a Realtek instead.

I would like to thank everyone for all the great help!!!