Linux: Network probs

[vass0922]

Ok I got my linux box up and running on fedora linux the other day and... TADA no network .. yay!

Well first of all I dont' blame fedora, cause I couldnt' get it to work on slack either

I thought I was just being a dolt and couldnt' get it to work on slack.. I Feel better now it doesn't work on the easy distro either

Anyways...

I have the linux box connected directly to my win2k PC (cross over cable)

I have a link light on both boxes, and the windows box shows there IS a connection.

However, on the linux box I can only ping the loopback address

ping 127.0.0.1 -- GOOD
ping 192.168.0.1 -- DEAD

.0.1 is windows PC
.0.2 is linux PC

.1.100 is my windows wireless NIC to connect to the router in the basement.

I've tried using windows ICS to work as DHCP for the linux box to pull an IP and I've tried setting the IP manually... no workie

I even got a packet sniffer on the windows box (set to correct interface) and it doesn't seem to get the packets from the linux box at all
I'm pretty sure the sniffer is working because I ran it for an hour and it did pick up broadcast packets from teh correct interface.

Unfortunately I don't have anything to help me test the cable, but I just received it from NewEgg the other day so assuming its good thus far.

[Scott Tiger]

This is a bit over my head but what does

/sbin/ifconfig eth0

(ass-uming you only have on network adapter) give you?

This is my output:

eth0 Link encap:Ethernet HWaddr 00:A0:CC5:AC:58
inet addr:192.168.2.100 Bcast:192.168.2.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:17169276 errors:1 dropped:0 overruns:0 frame:0
TX packets:16451954 errors:6 dropped:0 overruns:5 carrier:1
collisions:0 txqueuelen:100
RX bytes:4240952064 (4044.4 Mb) TX bytes:3089529835 (2946.4 Mb)
Interrupt:11 Base address:0xbf00

[krohnjw]

Look at your routing table, my guess if you have no default route. (use the route command)

[vass0922]

ya but I should be able to get that from DHCP
When my box broadcasts for DHCP my winders box doesn't even get it.

I noticed in my /var/log/messages a ton of messages

eth0: Too much work to do at interrupt, IntrStatus=0x0040 error

So I've been looking into that
I see that I can bump up the queue length but not sure how I see a place that does this, but it says it may happen under heavy loads.. this thing is just sitting there, no DC on it even.

I'm wondering if maybe the driver is arguing with it a bit (the driver installed with the OS)

Or considering swapping to a different PCI slot, even though I dont' see anything else on the same interrupt.

[lynchmob]

Or considering swapping to a different PCI slot, even though I dont' see anything else on the same interrupt.

I fixed one of my SuSE 8.1 boxes by doing that
lynch

[pbharris]

hi vass - have you tried assigning it a static IP to see if the hw is okay?

[nukes]

Does it show up in dmesg and /proc/pci?

[vass0922]

ya I tried static IP, same results

Haven't tried dmesg or /proc/pci as I'm not sure what the module name would be

I think its like r8203 or something like that, but couldn't find anything in lsmod

I can see it when I look for interrupts in the RH contral panel, but it just names it as eth0

[vass0922]

WOohoo!!

Comin' in from Fedora Linux

Had to swap the PCI slot

Damn board is known for that (Abit BE6-II)

Workin' smooth now

Thanks folks

wow even fonts look nice on here for a change

now just gotta get a REAL browser... not this blech mozilla thing

[Scott Tiger]

The reason why your fonts look so nice is because you're using Mozilla.

j/k

Just out of curiousity, have the fonts changed at all since RH 8/9 in Fedora? I've always thought that RH9 with Vera family of fonts is easier on the eyes than anything else. What's so different now?