Destination Host Unreachable

[Lemmingz]

Hi I just got a couple of old ne2000 compatable ethernet cards and tried to connect 2 of my linux boxes together to transfer files.
Both cards seem to be installed correctly and the modules load ok, but neither computer can connect to each other. When I hook them up to the hub I can see activity on the computer doing the pinging, but all I get with either is Destination Host Unreachable. Even with a direct crossover connector I still get the same results.
One comp is running RH 7.1 and the other is running RH 8.0
Any troubleshooting ideas?

[bmccarthy]

Start with ifconfig... whats your ips?

do a route and see what your routing table looks like.

Post those and lets see what you have.

[Lemmingz]

Ok here it is for the first computer with Red Hat 8.0

eth0 Link encap:ethernet WHaddr 00:80:29:62:92:10
inet addr:192.168.1.1 Bcast:206.0.0.255 Mask:255.255.255.0
UPBROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
RX bytes:0 (0.0 b) TX bytes:240 (240.0 b)
Interrupt:13 Base address:0x300

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:64 errors:0 dropped:0 overruns:0 frame:0
TX packets:64 errors:o dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:3812 (3.7 kb) TX bytes:3812 (3.7 kb)

BTW, How do you cut and paste from a termilal anyway? I had to type all of this in!
Thanks

[bmccarthy]

Sorry, I didn't mean to make you type all that out. I just really wanted the ip address and subnet mask. Can you do the same with the other machine? Also, get the output from typing "route" at the command prompt on both machines, should only be two or three lines.

[Lemmingz]

OK here it is. This is all from the first computer.

Destination Gateway Genmask Flags Metric Ref Use Iface
nas30.bellevuel * 255.255.255.255 UH 0 0 0 ppp0
206.0.0.0 * 255.255.255.0 U 0 0 0 eth0
192.168.1.0 * 255.255.255.0 U 0 0 0 eth0
127.0.0.0 * 255.0.0.0 U 0 0 0 lo
default nas30.bellevue1 0.0.0.0 UG 0 0 0 ppp0+

[Lemmingz]

second computer results for route
Kernel IP Routing Table
Destination Gateway Genmask Flags Metric Ref Use Iface
192.168.1.0 * 255.255.255.0 U 0 0 0 eth0
127.0.0.0 * 255.0.0.0 U 0 0 0 lo

Results for ifconfig

eth0 link encap:Ethernet HWaddr 00:40:33:2D:EC.53
inet addr:192.168.1.5 Bcast:192.168.1.255 Mask:255.255.255.0

[jumpedintothefire]

----quote----
206.0.0.0 * 255.255.255.0 U 0 0 0 eth0
192.168.1.0 * 255.255.255.0 U 0 0 0 eth0
------------

Looks like the second card is not configured, or you gave eth0 2 ip's what do you have in

/etc/modules.conf??
and
/etc/sysconfig/network-scripts/ifcfg-eth1??
/etc/sysconfig/network-scripts/ifcfg-eth0??
/etc/sysconfig/network-scripts/ifcfg-eth0:0??

while your at it, the output of
cat /proc/interrupts
cat /proc/ioports


I cheat, I get the output in to a file and the cut and paste from the file.....

route -n > /root/dump.txt
ifconfig > /root/dump2.txt

[pbratkowski]

copy paste:

select the text you want to copy, then press the middle button on the mouse... if you only have 2 buttons, you can press both at the same time if u have the emulate 3 buttons option turned on...

doesn't always work in different windows though

[bmccarthy]

You also have on the second machine a route of 192.168.1.0 with a mask of 255.255.255.255. Is that real or a mistype?

[Lemmingz]

the actual ip should have been
255.255.255.0
I have corrected it on the origional post

[Lemmingz]

/etc/modules.conf
alias paraport _lowlevel paraport_pc
alias eth0 ne
options ne irq=5 io=0x300

/etc/sysconfig/network-scripts/ifcfg-eth0
DEVICE=eth0
USERCTL=yes
ONBOOT=yes
BOOTPROTO=none
BROADCAST=192.168.1.255
NETWORK=192.168.1.0
NETMASK=255.255.255.0
IPADDR=192.168.1.5

interrupts are
CPU0
0=TIMER
1=KEYBOARD
2=CASCADE
4=SERIAL
5=NE2000
8-RTC
14=IDE0
15=IDE1

[jumpedintothefire]

Just to keep thing clear, the second machine has just one nic card? and the first has just one?? or two?? Where does the 206. network and ppp0 fit into the scheme of things? Just to have it clear in my head... Are you connecting both machines to a hub along with dsl, but only one machine (first) has the connection to the inet??

The second machine looks fine to me, if that is what the nic is set to... What about the first machine??

Just caught this: typo??
> Interrupt:13 Base address:0x300 <

Isn't irq 13 reserved for something....
math co-processor??

Time for some shut eye... I'll check back later....

[jrbush82]

Well lets get things a bit more simple. Have you tried pinging your local loopback (127.0.0.1)? If not, give that a shot on each system. That would give you a good idea that the ethernet interface is up and running. Have you checked the cables? Making sure your not using a crossover going to a hub, etc... ?

Have you ran netconfig? What distrobution are you running?

Good luck.

[Lemmingz]

I'm just hooking up one linux box up to another with a crossover cable. I have also a hub with 16 ports that I plan to use for a LAN party.
I have used the hub to connect them too, using the uplink port for the crossover cable. This is how i can see activity when I ping with either box. The RedHat 8.0 computer is connected to the internet via 56k dialup.
both machines have only one ethernet card.
At the beginning of my replys I state if it is the first or second computer I am referring to
First somputer is running Red Hat 8.0
Second is running Red Hat 7.1 (it won't boot from version 8.0, don't know why)

I'll check the irq's for the first computer now. that might be the problem if irq:13 is reserved for something else. I'll post the results tomorrow, it's getting late here.
Thanks for the help

[bmccarthy]

Well, after the correction for the typo, machine 2 looks ok to me. jumpedintothefire is right, though, lets get the cfg for machine 1. You shouldn't have two ip neworks, 206.0.0.0/24 and 192.168.1.0/24 in your routing table. That would show that both of those networks are local to eth0.

As for jrbush82's comment, that is where I was going next, just wanted to iron out ip and route information first. After we get that, we'll do a ping local, ping machine on network, etc.

[Lemmingz]

here are the interrupts for machine 1
CPU0
0: 14474735 XT-PIC timer
1: 82639 XT-PIC keyboard
2: 0 XT-PIC cascade
5: 352426 XT-PIC soundblaster
8: 1 XT-PIC rtc
11: 12323684 XT-PIC nvidia
12: 2413157 XT-PIC PS/2 Mouse
13: 0 XT-PIC NE2000
14: 1059308 XT-PIC ide0
NMI: 0
ERR: 0

Here is /etc/sysconfig/network-scripts/eth0
DEVICE=eth0
ONBOOT=no
BOOTPROTO=none
IPADDR=192.168.1.1
TYPE=Ethernet
USERCTL=yes
PEERDNS=no
NETWORK=206.0.0.0
BROADCAST=206.0.0.255

Here is /etc/modules.conf

alias parport_lowlevel parport_pc
alias char-major-195 NVdriver
alias sound-slot-0 sb
post-install sound-slot-0 /bin/aumix-minimal -f /etc/.aumixrc -L >/dev/null 2>&1 || :
pre-remove sound-slot-0 /bin/aumix-minimal -f /etc/.aumixrc -S >/dev/null 2>&1 || :
options sound dmabuf=1
alias synth0 opl3
options opl3 io=0x388
options sb isapnp=1
alias eth0 ne
options ne irq=13 io=0x300

here is /proc/ioports
0000-001f : dma1
0020-003f : pic1
0040-005f : timer
0060-006f : keyboard
0070-007f : rtc
0080-008f : dma page reg
00a0-00bf : pic2
00c0-00df : dma2
00f0-00ff : fpu
01f0-01f7 : ide0
0213-0213 : isapnp read
0220-022f : soundblaster
02f8-02ff : serial(auto)
0300-031f : eth0
0378-037a : parport0
037b-037f : parport0
03c0-03df : vga+
03f6-03f6 : ide0
03f8-03ff : serial(auto)
0a79-0a79 : isapnp write
0cf8-0cff : PCI conf1
1600-163f : Acer Laboratories Inc. [ALi] M7101 PMU
1c40-1c5f : Acer Laboratories Inc. [ALi] M7101 PMU
c000-cfff : PCI Bus #01
dfa0-dfa7 : US Robotics/3Com 56K FaxModem Model 5610
dfa0-dfa7 : serial(auto)
ffa0-ffaf : Acer Laboratories Inc. [ALi] M5229 IDE
ffa0-ffa7 : ide0
ffa8-ffaf : ide1

this is all for machine 1 with Red Hat 8.0

[jumpedintothefire]

This one is not the cause but you might want yes:
ONBOOT=no

These two:
NETWORK=206.0.0.0
BROADCAST=206.0.0.255

should be:
NETWORK=192.168.1.0
BROADCAST=192.168.1.255

Try that first, still not sure about that irq tho....

[Lemmingz]

I edited the network-scripts file like you show but still I get nothing.
I'm using the hub now so I can see that each machine doing the pinging is really sending something, but neither is receiving and returning the result.

[jumpedintothefire]

How is the nic card setup jumpers or a dos program?? I'd check to see what it is set to, and try moving the irq away from 13... then edit /etc/modules.conf to reflect the change...

[bmccarthy]

Couple of things...

I agree with jumped about the netmask and broadcast. Just to verify, you did restart your cards, or rebooted your machine, correct?

Still trying to stick to the basics before we go further, lets try pinging 127.0.0.1 and then 192.168.1.x (each machines eth0 ip) and finally ping the other machine. Lets see what we get each time around.

As for the IRQ 13, I don't particularly like the idea of using it either, but I hesitate to change IRQs now anytime. Computers in the past few years, with sharing and steering, don't follow all the rules they used to. We know the card is working on 13 since you are getting some indications on the hub, you may as well leave well enough alone.

Besides, if you are getting dest unreachable on both machines, most likely it is a package or setting you set on both machines that we just have to "unset".

[Lemmingz]

Each machine can ping it's own loopback and own card, but not each other.
I also found out that Red Hat 8.0 uses a file called "sndconfig-WHsVUI" and it is almost the same as the "modules.conf" file but it had a module listed for one of my old sound cards with an io conflicting with my eth0 card. so i deleted the old sound card entry.
I rebooted after this but still no results.

[Lemmingz]

Machine 1:
PING 192.168.1.5 (192.168.1.5) from 192.168.1.1 : 56(84) bytes of data.
From 192.168.1.1 icmp_seq=1 Destination Host Unreachable
From 192.168.1.1 icmp_seq=2 Destination Host Unreachable
From 192.168.1.1 icmp_seq=3 Destination Host Unreachable
From 192.168.1.1 icmp_seq=4 Destination Host Unreachable
From 192.168.1.1 icmp_seq=5 Destination Host Unreachable
From 192.168.1.1 icmp_seq=6 Destination Host Unreachable

--- 192.168.1.5 ping statistics ---
7 packets transmitted, 0 received, +6 errors, 100% loss, time 6028ms
, pipe 3

Machine 2:
PING 192.168.1.1 (192.168.1.1) from 192.168.1.5 :56(84) bytes of data.
From 192.168.1.5 Destination Host Unreachable
From 192.168.1.5 Destination Host Unreachable
From 192.168.1.5 Destination Host Unreachable
--- 192.168.1.1ping statistics---
5 packets transmitted 0packets received +3 errors 100% packet loss

[bmccarthy]

Running out of ideas here... if you can ping local, that means networking is up. Your routes look fine, ip configs look fine, if it was iptables or ipchains you shouldn't get dest unreachable. I don't think it's the interrupt since you are getting the exact same error on the machine using irq5.

Just for kicks, try to do a "service iptables stop" on both machines, and retry the ping.

You say you are seeing activity on the hub when you perform a ping, right?

And just to make sure, with all the changes, ifconfig shows
192.168.1.1 mask 255.255.255.0 for one machine and 192.168.1.5 mask 255.255.255.0 for the other...
and...
route shows
192.168.1.0 * 255.255.255.0 ... eth0
127.0.0.0 * 255.0.0.0 ... lo

Correct?

[Lemmingz]

yes that info is correct.
however idecided to look at my modules and tried to use the generic module that works with ne2000 cards.
the module loads but dosen't work.
when I try to load the be module with insmod I get the following error
using /lib/modules/2.4.2-2/kernel/drivers/net/ne.o
/lib/modules/2.4.2-2/kernel/drivers/net/ne.o: init_module: No suchdevice or address
Hint: insmod errors can be caused by incorrect module perameters, including invalid IO or IRQ perameters

and yes I did make sure that it was unloaded first.
this is for machine 2

[Lemmingz]

Well modprobe inserts the module ok.
weird.
also notice that the ping response from machine 1 has more info on it than the one from machine 2
I wonder why that is

[bmccarthy]

It still doesn't seem like that is the problem, since you are getting a reaction on the hub. Do you have another card you can try to attempt to isolate the issue? Perhaps that's worth a shot.

Did stopping iptables do anything?

[jumpedintothefire]

---quote----
Hint: insmod errors can be caused by incorrect module perameters, including invalid IO or IRQ perameters
------

So what kind if nic is it?? Does it have jumpers?
There should be a dos program to configure it if it's
jumperless. Run it and check what it is set to with it...
might be different... turn off the pnp it it is turned on, and then set it as you like. Then edit the /etc/modules.conf the the known irq and io-port.

May be the card is just plain hooped.......

[Lemmingz]

Both cards are jumperless isa pnp cards. machine 1 is a linux only comp and machine2 is a dual boot with win 95 and redhat 7.1. I used the hardware config tool in win95 to configure the card.

for machine 1 with redhat 8.0 only, I used ne2k-diag to probe the card.
i did try "service ip tables stop" but got a command not found error.

[jumpedintothefire]

The make and model would be nice.....
That is like saying "i drive a car" when you drive a dodge.... I duel booting would of been nice to know upfront, that can cause problems at times..The one that duel boots, pnp is turned off in the bios on that machine?
-----quote-----
I used the hardware config tool in win95 to configure the card.
-----
as in device manager? If you used a gui tool in win95 on that card, it is most likely in pnp mode and not jumperless mode. The driver disk should have the dos config program on it, with it check if the nic is set to pnp.

[Lemmingz]

machine 1 is an amd k6-2 400mhz on a alladin v 100mhz mainboard redhat 8.0
machine 2 is a pentium i200 on a 35-8258-03 mainboard (whatever that is). with win 95 on hard drive hda (from an old 486) and redhat 7.1 on hdb.
However I think I may have found the problem. one of the electrolytic caps on the card in machine 2 is bent over. this is usually normal in order to save space, but it looks like one of the legs is slightly pulled out of the body.
i'll switch the card out today to see if that works.

[Lemmingz]

well...
the new card didn't make any difference. :(
exactly the same problem

[Lemmingz]

well I did tcpdump on machine2 (192.186.1.5)
and pinged it from machine1 (`192.168.1.2) and this is what it says:

tcpdump: listening on eth0
13:54:35.923956 B arp who-has 192.168.1.1 tell 192.168.1.2

There are several lines like that. Any idea what this means?
When I tried to use tcpdump on machine 1 with redhat 8.0 it says command not found. When I tried to install tcpdump it says "already installed" I hate the package installer on redhat 8.0! it won't let you see any package info. It either installes it or doesn't, but that's another story.