| View previous topic :: View next topic |
| Author |
Message |
HippoMan
n00b
Joined: 05 Apr 2006
Posts: 11
|
 Posted: Sun Apr 09, 2006 6:48 am Post subject: [SOLVED] Another "Destination Host Unreachable" pr |
 |
|
I am trying to bring up Gentoo on a dual-processor machine that I have used for a few years with the following OS's: RedHat 8, RedHat 9, Fedora Core 2, Fedora Core 3, Fedora Core 4, Debian Woody, Debian Sarge, and Ubuntu Breezy. In all cases, I have been able to use the network with no problem when SMP is enabled. My network card is a Linksys NC100 Network Everywhere Fast Ethernet 10/100 (rev 11). I believe that this is a tulip card.
When installing Gentoo with the Minimal Installation CD, the network functioned without any problem after I ran net-setup. As proof of this, I was able to download stage3 and portage from Gentoo mirrors during my installation. Also, the various emerge's that I did during installation all functioned perfectly.
I built my kernel with genkernel, but when I boot up into this kernel, I no longer am getting any network access. I get "Destination Host Unreachable" when I ping internet addresses by IP (in other words, not by domain names, which means that DNS configuration is not the culprit).
Here is all the rest of the pertinent data that I can think of:
- During boot-up, I the following information shows up:
| Code: |
* Starting eth0
* Bringing up eth0
* 216.27.138.216 [ ok ]
* Adding routes
* default gw 216.27.138.1 ... [ ok ]
|
- If after boot-up I stop the network and then manually bring it up again with ifconfig, that also returns with no error message.
- After starting up the network (either with /etc/init.d/net.eth0 or manually with ifconfig), issuing an ifconfig indeed shows the IP address, the broadcast address, and the netmask. All of these are correct. It shows "TX bytes" equal to the number of bytes that my pings attempt to send, but the "RX bytes" are zero.
- I have disabled APIC and APCI in the kernel.
- The tulip driver is compiled into the kernel. This is the default that I got with genkernel, and I didn't change anything about this setting. I don't know how to make the tulip driver into a module, because genkernel doesn't give me this option ... I can either put an asterisk or a space next to it, but not an "M".
- When I run "netstat -r", everything is correct.
- I indeed did "rc-update add net.eth0 default" before booting. And at any rate, running an ifconfig immediately after boot-up shows correct data.
- This is the 2006.0 build of Gentoo.
- My /etc/conf.d/net file looks like this:
| Code: |
config_eth0=( "216.27.138.216 netmask 255.255.255.0 broadcast 216.27.138.255" )
routes_eth0=( "default via 216.27.138.1" )
|
Given the fact that my machine had no network problems under all the other OS's, this implies that there are no hardware problems. This is confirmed by the fact that all the network accesses that I did during installation from tne Minimal Installation CD also worked.
I didn't change anything about the network when building the kernel via genkernel, which is supposed to be pretty much the same kernel I get on the Minimal Installation CD.
So what could be going wrong here? This has got to be a simple thing that I'm overlooking, but I can't for the life of me figure out what it might be.
Thanks in advance for any help you folks can give me with this.
P.S. -- I should add that I have read the other messages here where people discuss similar problems, and I think that I have tried everything that has been suggested. I also have been all over Google, seaching for information about this problem, but I haven't found anything helpful yet.
|
|
| Back to top |
|
 |
HippoMan
n00b
Joined: 05 Apr 2006
Posts: 11
|
| Posted: Sun Apr 09, 2006 8:08 am Post subject: |
|
|
I solved the problem, but I'm not sure why.
I needed to refer to my network interface as eth1 instead of eth0.
Duing my Minimal Installation CD installation, I set up my network via net-setup eth0, and that worked: when I ran ifconfig, it showed the interface as eth0. But for some reason, after booting into my kernel, the interface then started appearing as eth1, which I just now noticed in the dmesg listing.
Why the active interface changed between the installation procedure and the boot-up of my kernel is a mystery.
But all's well that ends well, and perhaps the explanation of this is best left to the philosophers.
|
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Networking & Security
