Like most that post in this forum, I'm a complete newbie. I've installed RH9 and have run the latest update to get to kernel 2.4.20-28.9. I'm getting an error on startup since I, like a moron, changed a router configuration. The error is "Could not look up internet address for localhost.localdomain. This will prevent GNOME from operating correctly. It may be possible to correct the problem by adding localhost.localadmin to the file /etc/hosts." Then I get the option to Try Again or Log In Anyway.

Back to the moron part of my story. I went into the router earlier to see what IP address was assigned to my new linux machine since I couldn't find it in the control panels. When I looked at the Virtual Server settings (I've got a SMC 7004AWBR router), I noticed that something had entered private and public port settings for UDP and TCP in the table for my IP xxx.xxx.xxx.7. I cleared it all out fearing a virus or some sort of malicious attack, because I knew that I hadn't set them up. Well sure enough, my linux box is the .7 box. Sure enough, that's when the problems started.

Is it possible that the box configured the router with those settings during the system install?

Any help on how to get this setup again without reinstalling everything?

When you say you "cleared it all out," what exactly did you do to it? How does your router assign your connection? Did you take down the xxx.7 info before nuking it? Can you ping the router from the *Nix box? Does eth0 display on boot?

How about posting an # ifconfig? Let's see what there is to see .

Thanks for taking a look at my question - when I run "ifconfig" from the terminal line, I get "bash: ifconfig: command not found". When I try to ping my router address I get "connect: Netowrk is unreachable"

However, when I run the ifconfig command as root, I get
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metrick:1
RX packets:475349 errors:0 dropped:0 overruns:0 frame:0
TX packets:475349 errors:0 dropped:0 overruns:0 frame:0
collisions:0 txqueuelen:0
RX bytes:32566136 (31.0 Mb) TX bytes:32566136 (31.0 Mb)

But the ping request yeilds the same result: "connect: Network is unreachable"

I also went in and looked at the /etc/hosts file and there are only 3 lines - 1 of which looks useful, I assume the other two are comments:

#Do not remove the following line, or various programs
#that require network functionality will fail.
localhost.localdomain 127.0.0.1 localhost

Thanks again.

Actually, josh, that's my bad. Although you said you were a noob, your post seemed experienced, so I couldn't tell if you were totally new to Linux.

Sorry about that.

To clarify, when someone types '#' followed by a command, that signifies 'root,' as you see when you pull up a shell (command line), such as:

[root@localhost root]#

as opposed to '$', which signifies user, as you see when you pull it up as a user, such as:

[joshnerl@localhost joshnerl]$

Thus, for purposes of this analysis, the Linux box doesn't 'have' an IP address right now.

You pinged the router with a 192.xxx.x.x? If so, ok, so the Linux box can't see the router either.

Now, try, as root

# netstat -rn

and

# cat /var/log/messages

There, I'm looking for something like:

Put even more easily...does eth0 boot up <<green/OK>>? Anything <<red/failed>>?

You also identified your router, but not your NIC.



You can let that ride for now, since, as you see, the entry is not inconsistent with the Gnome error message.

edit: changed indicator in boot up reference

Sorry for the book, hoping more info the better - thanks for the help.

On startup, I get all green - but I couldn't read them fast enough to see if eth0 was in the list. On shutdown, it fails on smb smbd and smb nmbd. (The original intent of the box is a file server for XP and Mac so that's why they're running if anyone wonders.)

The netstat command gave me a routing table with two Destination IPs: 169.254.0.0 and 127.0.0.0. The Gateway listing for both is 0.0.0.0. The Genmask listings are 255.255.0.0 and 255.0.0.0 respectively. The Flags are both U, the MSSs are both 0, the Window and irtt on both are 0 and the Iface for both is lo.

When I look at the messages file, I've got a couple of lines that look like what you're looking for, but nothing exact. The two localhost network lines are "Setting network parameters: succeeded" and "Bringing up loopback interface: succeeded" - there's no localhost network: bringing up eth0. I'm guessing that means that eth0 isn't getting fired up.

So I tried going into the network panel and sure enough eth0 is inactive. When I tried to click activate I got "Determining IP information for eth0:1...SIOCSIFFLAGS: Cannot assign requested address SIOCCSIFFLAGS: cannot assign requested address failed". I've got it setup to get IP via DHCP, and the hardware device tab shows both boxes checked - device alias = 1, and it's bound to the MAC address.

You also mentioned pinging the NIC - this will be a dumb question to a lot of you, but how do I ping the NIC, if I don't have an IP? As a guess I pinged 127.0.0.1 - I get a response with 0% packet loss.

Having just spent the last week messing with IP configuration of a recalcitrant Windows machine I can tell you this means, more or less, that the NIC is DHCP configured, but can't see the router, so it's assigning itself that IP address.

One question I would still like you to answer for me is whether you copied down the IP address assigned by the DHCP Server (the router) to the Client (your Linux box) before you cleared it out. This shouldn't matter, but it would help address whether what you thought you did to the network actually did what you thought it did to the network. If it was something else, then we need to look at more symptoms....

for example....

Did your hardware ever give you a network connection...with this router?... with this kernel?...with another kernel?...Could you ever browse the web on your 'broken' Linux box?



Put simply, your card's not being activated...the question is why, and can it?



This is why it says 169.254.0.0

I actually asked you to identify it . What I meant was tell me what brand of card you have...Linux can be touchy with some hardware, I'm trying to isolate your problem and evaluate it step by step. Originally, you had said you had updated the kernel. What I'm trying to do now is identify whether your card is supported under the latest kernel iteration, as you've currently configured your system. Also, knowing the type of card may help determine how to get it activated.

One way you can do this for me is to post the output of:

lspci -v

Yes, it was working. I managed to hit the internet, see other machines, etc. I was working on the previous kernel (2.4.20-8), then I ran an update with RHN right after installing the system. I never ran the new kernel before I cleared out the router. To answer a previous question, I didn't copy down the router settings.

Should I just reinstall everything? It obviously set something up on install that it should be able to do again. This is my first attempt at Linux so I don't have any data on the machine - it's just the install.

Thanks.

SiriusAB - I installed RH9 on another HD in the same box and it runs fine on both kernels. In fact, I'm typing this from the box right now. I don't have any idea what could have happened, but really thank you for taking the time to read my questions and help me out. I'm learning a lot from these forums and really appreciate all of your help.

Can anyone point me to a set of instructions to set up the print server on a 7004awbr for linux?
I'm presntly running mandrake kernel 2.6.3

Thanks a bunch.

Jim