Red Hat 9 eth0 not accepting incoming packets.

BinkyTheOracle · 01-16-2004, 01:16 PM

I've found several threads that were close to my problem but none that fit exactly.

First a quick system overview: P3/450, 512m Ram, RH9, KDE.

The machine has two Intel EtherExpress/100 cards. One built in (eth1) and one PCI (eth0). I'm connecting to the eth0 device with a known working cable connected to a known working router port. After reading some of the threads I disabled eth1 to avoid any conflicts.

I'm mostly just wanting to use this as a toybox/browsing machine at this point and don't need to do anything fancy. I have a working LAN with a mixture of Mac & PCs connecting to a DSL line via a LinkSys 8 port firewall/router.

Note that I've tried everything I'm about to describe about eth0 on eth1 and had the exact same results, so let's just pretend eth1 doesn't exist.

Initially I set up eth0 with a static ip pointing to 192.168.1.1 as the gateway and provided it with my ISP's name servers. I was unable to send or receive pings from other LAN machines, but I did get successful pings to localhost and the eth0 device itself.

ifconfig showd the interface as being up and happy, and ifconfig -a shows the eth1 device as well, so I don't think it's a driver issue.

I watched the LED on the interface and it blinks when I try to ping, as does the light on the router, indicating that ping traffic is getting out of the box, but not coming back in.

I then pinged the red hat box from another machine on the LAN and watched the eth0 LED. Again, it was blinking, but the LAN machine reported that the red hat box was unreachable.

On a hunch, I switched the red hat box from a static IP to DHCP and told the Linksys router to provide addresses. While the machine rebooted, I refreshed the DHCP clients listing on the router and saw that the linux box had requested and been assigned an IP, but the eth0 failed to come up, further indicating to me that traffic isn't getting into the interface but is successfully going out.

Finally, with DHCP enabled I tried to set a default route to the gateway (router) and got the following:

SIOCADDRT: Network is unreachable.

I've tried these various things with the Network Security setting at High, Medium, and Off. I'm a linux newbie and pretty rusty on my other Unix OSes, so any help would be appreciated.

As I mentioned above, ifconfig is seeing the interface and reports it as being up when I use a static IP. Any help on what I should do now would be greatly appreciated!

david_ross · 01-16-2004, 02:02 PM

Welcome to LQ.

Is the netmask set right. Can you post the output of:
route -n
ifconfig eth0

And the same on another client. If it is a windows machine then:
ipconfig /all
route print

BinkyTheOracle · 01-16-2004, 02:47 PM

Thanks for the response David. Here's the requested output from the Red Hat 9 box. Note that this is with the DHCP settings in effect and eth0 fails to come up. Also, I'm typing this in since I can't cut & paste and zap it over the network. =-)

Linux box SET FOR DHCP:

ifconfig eth0:

Link encap: Ethernet HWaddr 00:90:27:BB:BD:58
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:128 errors:0 dropped:0 overruns:0 frame:0
TX packets:13 errors:0 dropped:0 ovverruns:0 carrier:0
collisions:0 txqueuelen:100
RX bytes:0 (0.0 b) TX bytes: 4446 (4.3 Kb)
Interrupt:11 Base address:0x1080 Memory:fa200000=fa200038

route -n

Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 lo
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo

Linux box SET FOR STATIC IP

ifconfig eth0:

Link encap: Ethernet HWaddr 00:90:27:BB:BD:58
inet addr:192.168.1.80 Bcast 192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:129 errors:0 dropped:0 overruns:0 frame:0
TX packets:19 errors:0 dropped:0 ovverruns:0 carrier:0
collisions:0 txqueuelen:100
RX bytes:0 (0.0 b) TX bytes: 4782 (4.6 Kb)
Interrupt:11 Base address:0x1080 Memory:fa200000=fa200038

route -n

Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo
0.0.0.0 192.168.1.1 0.0.0.0 UG 0 0 0 eth0

From Mac OS X 10.3.2 using static IP 192.168.1.50

ifconfig en0

en0: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu 1500
inet6 fe80::230:65ff:fe83:e11c prefixlen 64 scopeid 0x4
inet 192.168.1.50 netmask 0xffffff00 broadcast 192.168.1.255
ether 00:30:65:83:e1:1c
media: autoselect (100baseTX <full-duplex>) status: active
supported media: none autoselect 10baseT/UTP <half-duplex> 10baseT/UTP <full-duplex> 10baseT/UTP <full-duplex,hw-loopback> 100baseTX <half-duplex> 100baseTX <full-duplex> 100baseTX <full-duplex,hw-loopback>

route -n (note - the command is a bit different in darwin... I specified the route to the router for lack of a better option.)

Code:

route -n get 192.168.1.1
   route to: 192.168.1.1
destination: 192.168.1.1
  interface: en0
      flags: <UP,HOST,DONE,LLINFO,WASCLONED>
 recvpipe  sendpipe  ssthresh  rtt,msec    rttvar  hopcount      mtu     expire
       0         0         0         0         0         0      1500      1048

From Windows XP Pro using static IP 192.168.1.70

IPCONFIG /ALL

Windows IP Configuration

Host Name....................: RHETT
Primary DNS Suffix........:
Node Type.....................: Unknown
IP Routing Enabled........: No
WINS Proxy Enabled......: No

Ethernet adapter Local Area Connection

Connection-specific DNS Suffix.......:
Description....................................: 3Com Gigabit LOM (3C940)
Physical Address............................: 00-0c-6E-C0-BF-7A
Dhcp Enabled................................: No
IP Address.....................................: 192.168.1.70
Subnet Mask.................................: 255.255.255.0
Default Gateway...........................: 192.168.1.1
DNS Servers.................................: 204.117.214.10
199.2.252.10

ROUTE PRINT

Interface List
0x1....................................................MS TCP Loopback interface
0x2 ...00 0c 6e c0 bf 7a......................3Com Gigabit LOM (3C940) - Packet Scheduler Miniport

Active Routes:
Network Destination Netmask Gateway Interface Metric
0.0.0.0 0.0.0.0 192.168.1.1 192.168.1.70 20
127.0.0.0 255.0.0.0 127.0.0.1 127.0.0.1 1
192.168.1.0 255.255.255.0 192.168.1.70 192.168.1.70 20
192.168.1.70 255.255.255.255 127.0.0.1 127.0.0.1 20
192.168.1.255 255.255.255.255 192.168.1.70 192.168.1.70 20
224.0.0.0 240.0.0.0 192.168.1.70 192.168.1.70 20
255.255.255.255 255.255.255.255 192.168.1.70 192.168.1.70 1

Default Gateway: 192.168.1.1

Persistent Routes: None

Thanks again for the help. Let me know if there's any other information that would be helpful.

david_ross · 01-16-2004, 03:03 PM

Well all of your configuration when set to a static address looks perfect.

The next thing I would check is firewall rules although I think it is an unlikely problem. Just disable iptables with:
service iptables stop

Then try again with your static IP.

If it still fails then my guess is that your network cards aren't being recognised properly or the drivers aren't being loaded correctly. Do you see the correct driver in:
lsmod

dubman · 01-16-2004, 03:05 PM

Quote:

Linux box SET FOR DHCP: ifconfig eth0: Link encap: Ethernet HWaddr 00:90:27:BB:BD:58 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:128 errors:0 dropped:0 overruns:0 frame:0 TX packets:13 errors:0 dropped:0 ovverruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:0 (0.0 b) TX bytes: 4446 (4.3 Kb) Interrupt:11 Base address:0x1080 Memory:fa200000=fa200038

This interface is not pulling a dhcp address. Try the command:

#ifup eth0

if this dosen't work, try editing this file:

#/etc/sysconfig/network-scripts/ifcfg-eth0

and make this file read like this:

DEVICE=eth0
BOOTPROTO=dhcp
ONBOOT=yes

then try the ifup command again.

If you fail to recieve a dhcp license this time, I would think you have a physical layer problem. Perhaps check your cable and/or the port in your router...try a different one.

One more thinkg you can do is to set a static ip like you did before, and use the command:

#tcpdump eth0

and let it run, then ping this system from a second system and see if the echo requests are logged in tcpdump. If you can see the packets coming through, then you are having gateway/routing problems. If you see nothing then you have a physical connection issue.

BinkyTheOracle · 01-16-2004, 03:25 PM

Well, stopping the iptables service didn't help. I have absolutely no idea what to look for in the lsmod output, so here it is:

Code:

Module              Size        Used by     Not tainted
ipt_REJECT          3928        0 (autoclean)
iptable_filter      2412        0 (autoclean)
ip_tables          15096        2 (ipt_REJECT iptable_filter)
soundcore           6404        0 (autoclean)
parport_pc         19076        1 (autoclean)
lp                  8996        0 (autoclean)
parport            37056        1 (autoclean) [parport_pc lp]
autofs             13268        0 (autoclean) (unused
e100               80644        1
sr_mod             18136        0 (autoclean)
cdrom              33728        0 (autoclean) [sr_mod]
keybdev             2944        0 (unused)
mousedev            5492        1
hid                22148        0 (unused)
input               5856        0 [keybdev mousedev hid]
usb-uhci           26348        0 (unused)
usbcore            78784        1 [hid usb-uhci]
ext3               70784        2
jbd                51892        2 [ext3]
sym53c8xx          68112        3 
sd_mod             13452        6
scsi_mod          107128        3 [sr_mod sym53c8xx sd_mod]

(Edited to improve layout. Boy it sucks to not be able to cut and paste that stuff.

)

I'm assuming that the e100 line is the one I want to look at, but I don't know what to do with it or where to look for better drivers.

Thanks again for the help!!!

david_ross · 01-16-2004, 03:39 PM

Yes that is the module for your ethernet card. I have read a few posts elsewhere that seem to suggest there are problems with some versions of acpi and these cards. Perhaps you should try upgrading to the most recent stable kernel from:
http://www.kernel.org

BinkyTheOracle · 01-16-2004, 03:44 PM

Hi Dubman, thanks for the response.

Quote:

This interface is not pulling a dhcp address. Try the command:

#ifup eth0

if this dosen't work, try editing this file:

#/etc/sysconfig/network-scripts/ifcfg-eth0

and make this file read like this:

DEVICE=eth0
BOOTPROTO=dhcp
onBOOT=yes

then try the ifup command again.

If you fail to recieve a dhcp license this time, I would think you have a physical layer problem. Perhaps check your cable and/or the port in your router...try a different one.

The thing that's strange to me is that it is requesting the DHCP address because the router (LinkSys 8 port) is assigning it one. I get a DHCP entry (on the router's DHCP client listing) with the correct hostname and MAC address. The problem is that the info isn't getting back into the linux box. I see blinky lights all the way back to the linux box's interface, so I know traffic is going through (not to mention that it works fine with the computer that's normally on that cable).

Quote:

One more thinkg you can do is to set a static ip like you did before, and use the command:

#tcpdump eth0

and let it run, then ping this system from a second system and see if the echo requests are logged in tcpdump. If you can see the packets coming through, then you are having gateway/routing problems. If you see nothing then you have a physical connection issue.

When I typed that command in, I got the following error:

Code:

tcpdump: parse error

I then used tcpdump -c eth0 and got:

Code:

tcpdump: invalid packet count eth0

Does that shed any light on the problem? I'm confident that it's not a hardware problem up until I get to the eth0 interface. I'm fairly sure that it's not the interface, though, because I get the exact same behavior when using eth1 instead of eth0 and the likelihood of both interfaces (one built-in, one a PCI card) having the same hardware defect seems kind of low.

I suspect a driver issue at this point, but don't know enough about linux drivers to know where to start looking for information or updates.

I appreciate the help!

BinkyTheOracle · 01-16-2004, 03:55 PM

David wrote:

Quote:

Yes that is the module for your ethernet card. I have read a few posts elsewhere that seem to suggest there are problems with some versions of acpi and these cards. Perhaps you should try upgrading to the most recent stable kernel from:
http://www.kernel.org

You're really trying to scare the hell out of me now.

Ok, I went to the site and poked around but didn't find a FAQ on changing out the kernel, so before I get too worried about it, I have the following questions:
[list=1][*]Will updating the kernel break the rest of the installation? In other words, am I going to need to update a lot of packages?[*]Will I need to compile the new kernel or are there binaries available? (I'm not worried about max performance at this point)[*]Would I be better off downloading a different distro like Mandrake or Gentoo? (I know Gentoo is more difficult to deal with, but since I don't have a lot of time invested in RH9, I'm not averse to working with something else.)[/list=1]

david_ross · 01-16-2004, 03:55 PM

Getting the kernel updated will update your driver. The 2.6 series is also generally faster than the 2.4 series you will be on with the default install.

david_ross · 01-16-2004, 04:01 PM

1) No but you should upgrade the rpm package itself as it is known not to work (or at least it's database isn't) with 2.6.*
2) I don't think there are rpms for 2.6.* yet but you could try upgrading to a later 2.4 kernel with an rpm.
3) There shouldn't be a reason to change distros, it is only a minor problem.

You could try the following rpm upgrade as a first step:
ftp://rpmfind.net/linux/redhat/updat...-28.9.i386.rpm

Mr.Kex · 01-16-2004, 04:06 PM

You have to install all the packages...

BinkyTheOracle · 01-16-2004, 04:11 PM

David, thanks so much for the info and pointers. I'm downloading the 2.4.20 kernel rpm, will burn it on a CD and attempt the upgrade.

One final question: are you saying that the most current RPM package doesn't work with the 2.6 kernel at all?

Or is it that the rpm package that's installed with RH9 won't work with 2.6 and I need to update RPM to the most current version if I update to the 2.6 kernel?

(Edit: Yes, that was two questions.)

BinkyTheOracle · 01-18-2004, 11:31 AM

I wanted to post an update of what I've learned so far. Essentially, it appears that the Intel EtherExpress / Pro/100 lines have some issues with the Linux 2.4 kernel.

After fighting with Red Hat 9 for a while, I decided to try my luck with the Gentoo distro. Despite the fact that Gentoo has both the e100.o and e100pro.o drivers already available and compiled, I've been unable to get any inbound traffic to get past the interface. The lights are blinken but nobody's answering the door.

I'm still a newbie with linux and very rusty on my linux skills, so I haven't had much success with updating the kernel on either distro. I'll continue to look for info and will post my results to this thread when I get some success in the hopes that it will help others in the future.

Alternately, I may just go buy a $20 card that works with the 2.4 kernel...

Thanks again for all the help!

david_ross · 01-18-2004, 01:05 PM

Sorry I missed this post:

Quote:

Originally posted by BinkyTheOracle
David, thanks so much for the info and pointers. I'm downloading the 2.4.20 kernel rpm, will burn it on a CD and attempt the upgrade.

One final question: are you saying that the most current RPM package doesn't work with the 2.6 kernel at all?

Or is it that the rpm package that's installed with RH9 won't work with 2.6 and I need to update RPM to the most current version if I update to the 2.6 kernel?

(Edit: Yes, that was two questions.)

It is the rpm program itself that does not work. Therefore it should be upgraded before you install 2.6. The latest version does work, I have done this myelf at work with 2.6.1 and RedHat 9. You can download it from here:
ftp://ftp.rpm.org/pub/rpm/test-4.2/r...4.2-1.i386.rpm

I would strongly reccomend installing the latest kernel to try getting your existing card to work.