You just remove eth0 since it is your Internet connection. Then just reconfigure your network as described as above

No luck yet...

I removed the eth0 connection, and re-created it. No change in behavior. I removed it and rebooted, only to find that it had been re-created upon boot. Re-created it anyway; no change.

At that point, I figured I had nothing to lose, so I dropped the Linux partitions, and re-installed so that I could start over. I let the Mandrake installation do whatever it thought it ought to in terms of a default configuration. After booting, I saw no change in behavior, and here are some diagnostics:

'/sbin/ifconfig' returns:
eth0 Link encap:Ethernet HWaddr 00:07:E9:F7:26:1A
inet addr:192.168.0.4 Bcast:192.168.0.255 Mask:255.255.255.0
inet6 addr: fe80::207:e9ff:fef7:261a/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:362 errors:0 dropped:0 overruns:0 frame:0
TX packets:552 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:30386 (29.6 Kb) TX bytes:41885 (40.9 Kb)
Interrupt:20 Base address:0x9000


lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:62 errors:0 dropped:0 overruns:0 frame:0
TX packets:62 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:3914 (3.8 Kb) TX bytes:3914 (3.8 Kb)

'/sbin/route' returns:
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
192.168.0.0 * 255.255.255.0 U 0 0 0 eth0
127.0.0.0 * 255.0.0.0 U 0 0 0 lo
default dslmodem.domain 0.0.0.0 UG 0 0 0 eth0

'dmesg | grep eth0' returns:
ip1394: eth0: IEEE-1394 IPv4 over 1394 Ethernet (fw-host0)
eth0: 0000:02:08.0, 00:07:E9:F7:26:1A, IRQ 20.
eth0: no IPv6 routers present


I thought that first entry from 'dmesg' looked interesting in terms of the presence of both eth0 and ip1394 -- that can't be right, but I don't know what to do about it.

I removed the eth0 connection and re-created it using MCC. No change. After rebooting, here is the output from the same set of diagnostics:

'/sbin/ifconfig' returns:
eth0 Link encap:Ethernet HWaddr 00:07:E9:F7:26:1A
inet addr:192.168.0.4 Bcast:192.168.0.255 Mask:255.255.255.0
inet6 addr: fe80::207:e9ff:fef7:261a/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:21 errors:0 dropped:0 overruns:0 frame:0
TX packets:31 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:2398 (2.3 Kb) TX bytes:2917 (2.8 Kb)
Interrupt:20

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:42 errors:0 dropped:0 overruns:0 frame:0
TX packets:42 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:2656 (2.5 Kb) TX bytes:2656 (2.5 Kb)


'/sbin/route' returns:
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
192.168.0.0 * 255.255.255.0 U 0 0 0 eth0
127.0.0.0 * 255.0.0.0 U 0 0 0 lo
default dslmodem.domain 0.0.0.0 UG 0 0 0 eth0

'dmesg | grep eth0' returns:
eth0: 0000:02:08.0, 00:07:E9:F7:26:1A, IRQ 20.
eth0: no IPv6 routers present

As nearly as I can tell, the only difference is the absense of the ip1394 message in the 'dmesg' output stream.

I have also tried adding the line 'alias ipv6 off' as both the first and then the last line of /etc/modprobe.conf based on several other network problem-related threads here within the forums, and then rebooting after each change. Still no luck...

Can any of you make heads or tails out of what is going wrong with the network configuration? I also have the complete 'dmesg' stream saved, too, if that would shed any additional light on this.

I may try installing a different distribution and/or booting under Knoppix to see if they handle the networking configuration better and hopefully to give me a clue as to what needs to be different...

Thanks again for all your help!

Could this be an IPv6/inet6 issue? If so, how do I disable IPv6/inet6 under Mdk 10.1?

I did quickly boot up under Knoppix this morning, and it recognized the network without any problems whatsover. The output from that same set of commands looked like this:

'/sbin/ifconfig' returns:
eth0 Link encap:Ethernet HWaddr 00:07:E9:F7:26:1A
inet addr:192.168.0.4 Bcast:192.168.0.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:880 errors:0 dropped:0 overruns:0 frame:0
TX packets:1107 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:391858 (382.6 KiB) TX bytes:156291 (152.6 KiB)
Interrupt:20 Base address:0xd800 Memory:feafd000-feafd038

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:8 errors:0 dropped:0 overruns:0 frame:0
TX packets:8 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:400 (400.0 b) TX bytes:400 (400.0 b)

'/sbin/route' returns:
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
192.168.0.0 * 255.255.255.0 U 0 0 0 eth0
default dslmodem.domain 0.0.0.0 UG 0 0 0 eth0

'dmesg | grep eth0' returns:
e100: eth0: Intel(R) PRO/100 Network Connection
e100: eth0 NIC Link is Up 100 Mbps Full duplex

There are some obvious differences (not least of which is that I booted Knoppix under its 2.4 kernel): no 'inet6' address on the interfaces under ifconfig, no entry for the lo interface in the routing tables, and Knoppix is using a different driver to talk to eth0 (eepro100 vs e100).

If I get a chance later, I will likely try to boot Knoppix under its 2.6 kernel and see what happens.

Does this give any of the resident gurus anything more to go on?

Thanks!

First things first.......

does Mandrake detect the precense of your NIC? if that is so, then proceed to create a new network connection in mandrake control center. then you can delete the "firewire" network conneciton. If mandrake cannot detect your ethernet nic, then you have bigger issues and you need to make sure that it is detected first, prior to create the connection

The result of the dmesg is ok. It is telling you that Eth0 is definitely the firewire port and it is trating it as an ethernet device. Question, when you go to the mandrake control panel and select hardware (to see the detected hardware devices) do you see an entry for the actual network card?

Lets take it from the top, before we proceed any further…….

You have a DSL connection, therefore you have a modem with a connection to your wall-jack and another connection to your computer (network card, not the firewire port). Please bear with me on this, ok? The other way could be that you have the modem connected to a broadband router and then your computer is connected to the router. If your setup is differet ofrm the one tha tI have explained or you have any comments regarding it, please let us know.

Having said that, lets see what we can do…..

Mandrake is detecting the firewire port as a capable ethernet device, but not the actual nic. First, change the configuration of the "firewire" port (eth0) to use DHCP. If you were giving it an static IP address makes sure that it is configured to get the ipaddress automatically. Unless I'm smoking something, there is no way that it will get an IP address.

Next, create a new connection, making sure that you select the actual Network card for it, and go through the configuration steps. It is important to make sure that the gateway intefrace is also the actual NIC. After you do that chec if you can get on the web. If everything works ok you should not have any problems.

It honestly do not think it is an issue with IPv6. You can disable it in /etc/modules if you want, but do it after you get this working.

As I said, I have experieced this since I started using mandrake 10.0, na dunless you have a wierd configuration, We should be able to help you fix this...

Thanks for the reply, xavierh!

I will try to answer as many questions as I can, given that I am not at my computer at home...

Under MCC > Hardware, I do see separate, different, distinct entries for both the NIC and the FireWire port in the "Ethernetcard" section of the display. If I select each of them, it shows me different information on each. Let me know if I need to post the information shown for each of them to help troubleshoot.

I have DSL service through Qwest on an ActionTec GT701-wg, which Qwest describes as a "DSL modem with wireless gateway". It is connected to the phone wall jack via cable, and to the NIC's RJ45 port (not the FireWire port) on my computer. Internet access via Windows XP is fine and was fine under Mdk9.2.

Within the output from ifconfig, above, the HWaddr listed on eth0 is the same MAC address as my network connection under Windows XP (I have checked a couple of times and I just had my son double-check); wouldn't that mean that eth0 does actually correspond to the NIC, rather than the FireWire port?

eth0, under MCC, is listed as corresponding to the "Intel Corp. | EtherExpress PRO/100" and uses driver eepro100; it is also set to use DHCP. eth1 under MCC is listed as corresponding to the "IEEE1394 IPv4 Driver (IPv4-over-1394 as per RFC 2734)". I /think/ it is listed as using an "ip1394" driver, but I can't be certain about that until I get home and check this evening.

Given the above, am I understanding your directions correctly to try to configure the FireWire connection to use DHCP, even though there is nothing connected to it? Then reconfigure the connection on the NIC (which I believe is eth0)?

One question that I have in reading your response: you say "It is important to make sure that the gateway interface is also the actual NIC." I don't know what that means. I can't remember ever specifying a "gateway interface" in any of the network configuration experimenting I have been through in trying to get this to work...

Thanks again for your patience and your willingness to help!

Don't worry amigo, this is a slow week…..

Reviewing your information, it seems that everything is where it should be. Open a terminal window and do a ping to your router (ping [router_IP_ADDRESS] and see if you get a response. If you can ping your router but can't access the web using a name, the problem may be either DNS or default gateway configuration. See if getting rid of the eth1 interface solves the problem.

I remember seeing on mine (I manually configured the ip address of my computers at home), that you may need to specify the gateway interface. And I know I had to do that before to make sure that the gateway interface was eth0, not eth1. Yours, since it is using DHCP, may not be the case.

I have tried to replicate that issue with my VMWare MDK 10.1 session at work but I have been unable to do that. Let me check that as soon as I get home and I'll get back to you tonight or tomorrow morning…

OK, xavierh...

I re-installed once more to get a clean system in its default state. Diagnostics are still as shown above...

I can ping my router:
PING 192.168.0.1 (192.168.0.1) 56(84) bytes of data.
64 bytes from 192.168.0.1: icmp_seq=1 ttl=255 time=0.718 ms
64 bytes from 192.168.0.1: icmp_seq=2 ttl=255 time=0.687 ms
64 bytes from 192.168.0.1: icmp_seq=3 ttl=255 time=0.700 ms
64 bytes from 192.168.0.1: icmp_seq=4 ttl=255 time=0.656 ms
64 bytes from 192.168.0.1: icmp_seq=5 ttl=255 time=0.721 ms
64 bytes from 192.168.0.1: icmp_seq=6 ttl=255 time=0.759 ms
64 bytes from 192.168.0.1: icmp_seq=7 ttl=255 time=0.718 ms
64 bytes from 192.168.0.1: icmp_seq=8 ttl=255 time=0.710 ms
64 bytes from 192.168.0.1: icmp_seq=9 ttl=255 time=0.645 ms

--- 192.168.0.1 ping statistics ---
9 packets transmitted, 9 received, 0% packet loss, time 7999ms
rtt min/avg/max/mdev = 0.645/0.701/0.759/0.043 ms


'nslookup 192.168.0.1' returns:
Server: 192.168.0.1
Address: 192.168.0.1#53

Non-authoritative answer:
1.0.168.192.in-addr.arpa name = dslmodem.domain.actdsltmp.

Authoritative answers can be found from:


'nslookup dslmodem.domain.actdsltmp' returns:
Server: 192.168.0.1
Address: 192.168.0.1#53

Non-authoritative answer:
Name: dslmodem.domain.actdsltmp
Address: 192.168.0.1


After the re-installation, I went into 'MCC > Network & Internet > Manage connections' and found the following:
eth0: Intel Corp.|EtherExpress PRO/100
TCP/IP
Configuration
Protocol DHCP
IP address: 192.168.0.4 (grayed)
Netmask: 255.255.255.0 (grayed)
Gateway: (blank)
DNS Servers
192.168.0.1
199.104.81.3
Search Domain
domain.actdsltmp
Options
[x] Start at boot
[ ] Track network card id (useful for laptops)
[ ] Network Hotplugging
Metric: 10
Information
Vendor: Intel Corp.
Description: EtherExpress PRO/100
Media class: NETWORK_ETHERNET
Module name: eepro100
Mac Address: 00:07:e9:f7:26:1a
Bus: PCI
Location on the bus: 2


eth1: IEEE 1394 IPv4 Driver (IPv4-over-1394 as per RFC 2734)
(nothing else)

Noticing that the Gateway field was blank and insensitive, I switch eth0 to static, entered 192.168.0.1 for the gateway, applied the change, switched it back to DHCP, applied the change and rebooted. MCC still shows the 192.168.0.1 gateway after rebooting, but I can't see the web.

Every time I remove the eth1 connection on the FireWire port, it just comes back when I reboot, so I can't really tell if removing it is doing anything...

I can run nslookup or host on an address like www.cnn.com or www.google.com and it will respond with a set of IP addresses for those domains. Can I assume, based on all of that, that I am not dealing with a DNS problem?

If I run a '/usr/sbin/traceroute -i eth0 www.mandrakelinux.com', I get this:
1 dslmodem.domain.actdsltmp (192.168.0.1) 0.614 ms 0.576 ms 4.028 ms
2 idflcr01.onewest.net (199.104.81.1) 130.680 ms 118.896 ms 115.927 ms
3 1533.atm6-1-0.bois-cust.bois.uswest.net (207.108.226.233) 132.554 ms 147.354 ms 137.364 ms
4 bse-core-01.inet.qwest.net (205.171.155.17) 142.385 ms 149.422 ms 145.740 ms
5 slk-core-01.inet.qwest.net (205.171.205.146) 168.798 ms 156.009 ms 168.919 ms
6 slk-core-02.inet.qwest.net (205.171.131.98) 172.216 ms 184.074 ms 187.157 ms
7 svl-core-02.inet.qwest.net (205.171.205.118) 213.330 ms 217.111 ms 218.741 ms
8 svl-core-03.inet.qwest.net (205.171.14.126) 223.451 ms 233.144 ms 233.263 ms
9 pax-brdr-02.inet.qwest.net (205.171.205.30) 256.659 ms 245.169 ms 255.193 ms
10 208.50.13.185 (208.50.13.185) 264.781 ms 81.857 ms 80.927 ms
11 so3-0-0-2488M.ar2.CDG2.gblx.net (67.17.65.114) 230.094 ms 217.027 ms 233.478 ms
12 Lost-Oasis.ge-1-2-0.403.ar2.AMS2.gblx.net (67.17.159.222) 200.279 ms 198.753 ms 200.197 ms
13 feth4-biales.rdb.fr.core.ielo.net (212.85.144.4) 203.866 ms 203.453 ms 205.117 ms
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * *

Every other site I try does basically the same thing: it makes a few hops and starts timing out...

On WinXP, tracert www.mandrakelinux.com returns:
Tracing route to www.mandrakelinux.com [212.85.147.164]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms dslmodem.domain.actdsltmp [192.168.0.1]
2 45 ms 40 ms 38 ms idflcr01.onewest.net [199.104.81.1]
3 47 ms 52 ms 49 ms 1533.atm6-1-0.bois-cust.bois.uswest.net [207.108.226.233]
4 47 ms 46 ms 47 ms bse-core-01.inet.qwest.net [205.171.155.17]
5 53 ms 62 ms 57 ms slk-core-01.inet.qwest.net [205.171.205.146]
6 58 ms 57 ms 59 ms slk-core-02.inet.qwest.net [205.171.131.98]
7 89 ms 86 ms 79 ms svl-core-02.inet.qwest.net [205.171.205.118]
8 70 ms 69 ms 72 ms svl-core-03.inet.qwest.net [205.171.14.126]
9 72 ms 72 ms 72 ms pax-brdr-02.inet.qwest.net [205.171.205.30]
10 70 ms 72 ms 70 ms 208.50.13.185
11 200 ms 201 ms 204 ms so3-0-0-2488M.ar2.CDG2.gblx.net [67.17.65.114]
12 206 ms 205 ms 206 ms Lost-Oasis.ge-1-2-0.403.ar2.AMS2.gblx.net [67.17.159.222]
13 207 ms 208 ms 208 ms feth4-biales.rdb.fr.core.ielo.net [212.85.144.4]
14 213 ms 216 ms 211 ms 212.85.147.164

Trace complete.

Does any of this tell you anything, or give you a lead to follow?

This is definitely mindboogling......

I also tried doing a traceroute to www.mandrakelinux.com and my request times out, but my internet is working. Try www.yahoo.com and see if it works for you....it works for me.

Again if you are able to to a traceroute on that (www.yahoo.com), it means that the internet is working for you. So the issue is not with dns. I also check my network settings and I also have a eth1 interface assigned to the firewire card but with no settings.

the traceroute test shows that you have connectivity to the internet beyond your modem, therefore what browser do you use to connect to the web? do you have set any type of proxy configuration in either konqueror, the kde control center, or the mcc?

See if the traceroute for www.yahoo.com works. If that works then the problme is elsewhere....
let me know how it goes....

Mindboogling -- good word...

traceroute www.yahoo.com completed successfully. When I point Konqueror at it, it retrieves most of the page, stalls, and fails to finish. It just sits there. Epiphany and Mozilla, on the other, just fail and throw up a message about being unable to connect.

I checked on the proxy idea. The proxy setting under Konqueror is set to "Connect directly to the Internet", the proxy in the KDE control center looks identical, and the both the HTTP proxy and FTP proxy on the MCC panel are empty.

As I indicated, I've tried Konqueror, Epiphany, Mozilla, and Firefox. It also seems to fail in a similar situation if I point Thunderbird at the mail servers for my ISP. Whatever the problem is, it isn't unique to the Web...

I'm toying with installing a different distro (Fedora Core 3, perhaps) and seeing it can see the network...

I think I know now what is going on: I installed Fedora Core 3 early this morning and it ran into exactly the same kind of problem. In looking through all of the diagnostic output I had accumulated over the past week from Mandrake 10.1, Knoppix 3.7 (2.4 kernel), Knoppix 3.7 (2.6 kernel), Fedora Core 3, and my son's Mandrake 9.2 box, the only thing that seemed to stand out to me was that when IPv6 was enabled (as shown in the 'inet6 addr' line in output from ifconfig and in the 'eth0: no IPv6 routers found' from dmesg), I couldn't reliably access the Internet.

Disabling IPv6 under FC3 solved the problem. A gentleman and scholar named David Ross pointed me toward how to do this in FC3: append the line 'alias net-pf-10 off' to the end of /etc/modprobe.conf. I don't know if this is generic to FC3 or is more generally applicable to other distros as well, but it worked!

xavierh: Thanks again for your help!

No problem amigo....I don't have that issue though and I have it enabled (IPv6), but it would to hurt to disable it...

merry christmas...

You can try to disable eth1 (ieee1394) by turning off the option "Start at boot" under MCC>Netw & Internet>Manage Conn>Options for eth1

Then reboot and see what happens.

If you can't get rid of eth1, take a look at /etc/sysconfig/network-scripts/ifcfg-eth1
and see if there's a line ONBOOT=no.