Weird network problem in Slack 10.2
 

Hey people!

I made a big mistake on one of my servers. I installed slackpkg and forgot to use 10.2 sources. I ran an upgrade-all and of course, almost everything stopped to work. Enough about that. I reinstalled Slackware two times because of the same problem:

The network just doesn't work. I'm sure all is rightly configured. This server is very critical here, it runs backup of important files, so I need to get it up as fast as possible. I'm posting some output and info below:

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
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 U 1 0 0 eth0

resolv.conf:
search rebootlan.net
nameserver 192.168.1.10 217.13.4.21

ifconfig -a
eth0
Link encap: ethernet HWaddr: 00:10:b5:57:87:32
inet addr: 192.168.1.14 Bcast: 192.168.1.255 Mask: 255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU: 1500 Metric: 1
RX packets: 0 errors: 0 dropped: 0 overruns: 0 frame: 0
TX packets: 0 errors: 0 dropped: 0 overruns: 0 carrier: 0
collisions: 0 txqueuelen: 1000
RX bytes: 0 (0.0 b) TX bytes: 0 (0.0 b)
Interrupt: 12 base address: 0x9f00

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: 9 errors: 0 dropped: 0 overruns: 0 frame: 0
TX packets: 9 errors: 0 dropped: 0 overruns: 0 carrier: 0
collisions: 0 txqueuelen: 0
RX bytes: 1008 (1008.0 b) TX bytes: 1008 (1008.0 b)

found something in /var/log/messages that may help to solve the problem:

neptune kernel: NETDEV WATCHDOG: eth0: transmit timed out
neptune kernel: eth0: link up, 100mbps, full-duplex, lpa0x45E1

/var/log/syslog says some weird things about errors with insmod, something about shpchp.o.gz and pciehp.o.gz. It says: Hint: Insmod errors can be caused by incorrect module parameters, including invalid IO or IRQ parameters.

Can someone please help me with this? It doesn't work to reinstall.

make sure you know what kind of netork card you have.
run /sbin/netconfig
then check out your /etc/modules.conf and /etc/rc.d/rc.modules file for correct network modules setup.

yeah i've run netconfig. but why would slack detect my network card before i reinstalled and not now? I checked /etc/modules.conf. It was empty. /etc/rc.d/rc.modules contains everything it should contain, but no network card drivers is active. In the top of this file it is written that this file contains extra drivers. Slackware was able to detect my ethernet card last time i installed it with the exact same hardware, so why not now? I even reinstalled two times.

Your interface looks active and configured, your route table is OK. I would just re-write /etc/resolv.conf to
because yours is syntactically incorrect.

The network interface statistics show that no packet whatsoever is being transmitted - what does the output of these diagnostics commands tell you?
(the second one may yield "operation not supported" if your card is too new)

don't worry about these, you can blacklist those two modules if the errors annoy you. You don't have the hardware for them which causes the errors in the messagelog.

Eric

If it is not your network card; then check that you enable packet filters with firewall anywhere in your scripts
Ping is your friend;
ping localhost
ping google.com

then check your ifconfig; theose TX/RX number must be changed.

Hey!

ethtool eth0
settings for eth0
supported ports: [TP MII]
supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full

supports auto-negotiation: yes

advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
advertised auto-negotiation: yes
Speed: 100 MB/s
Duplex: Full
Port: MII
PHYAD: 32
Tranceiver: internal
Auto-negotiation: on
Supports wake-on: pumbg
Wake-on: d
Current message level: 0x00000007 (7)
Link detected: yes

mii-tool eth0
eth0: negotiated 100baseTx-FD flow control, link ok

ping localhost
All okey here.

ping google.com
ping: unknown host google.com

Try booting from a Knoppix cd or something similar to test the nic in question.

So, "link detected" means the network connections are OK. Not being able to resolve host names like google.com could mean a problem with your nameservers. Is "192.168.1.10" really running a working DNS server?
Try these commands:
The IP address is that for www.slackware.com, so if the first command returns results and the second doesn't, then your DNS is broken.
To see what your configured DNS servers say, try querying those two DNS servers (192.168.1.1 and 217.13.4.21) directly:
What does this return?

Eric

ping 64.57.102.34: destination host unreachable
ping www.slackware.com: unknown host

My local DNS server, 192.168.1.10 is working, but it is offline now. 217.13.4.21 is my ISP's dns server, set up as secondary dns server. this configuration works on all other computers in this house. I'm pretty sure my dns ain't broken.

i ran the last one, cause my dns is offline.

host www.slackware.com 217.13.4.21: ;; connection timed out; no serves could be reached

Something is not right in your network. You say that the rest of the computers in your LAN work fine? Can you ping their IP addresses from this SLackware computer?

Is 192.168.1.1 really the router for your network?

Also, you can use ping and the traceroute tool to see where the networking stops:
should return something.
should give an indication of how far down the routing breaks if the previous ping returned packets.

Eric

I cannot ping the other computers from the slack computer. 192.168.1.1 is the modem from my isp, 192.168.1.10 is my dns and other services-server.

I have tried to ping the modem, but same as the other pings, destination host unreachable.

traceroute -n 64.57.102.34 says it's stopping on the slack computer (the same computer that it is run on)

I would almost begin to believe that that NIC is fried... Do you have a Live-CD of any kind (Knoppix, SLAX) to test network functionality?

Eric

I hope that you check your network cable already.
is it patch cable or cross-over cable?

My networks contain those two different type of cable all the time.

Hey number22! The newer network cards and switches doesn't "care" wether it is crossover-cable or patch-cable you use. And it worked before reinstallation.

I have tested with an Ubuntu live-cd and it all works there, so the nic is not fried. Weird!

I think the problem is your routing tables - the default gateway entry in particular.

Let me explain.

Your posted routing table in your original post is as follows:
I compared that to my routing table, which looks as follows. Pay close attention to my default route entry's flags.
Notice mine has the 'G' gateway flag set, and yours does not.

This may or may not be the problem, but I think it should at least be considered. All my machines have the 'G' flag set and work properly.

Maybe you should try re-running netconfig. The reboot, and re-check your 'route -n' output to see if the G flag is now set as it should be.

Or, at lease delete the entry, add it back in, and re-check the routing table for the 'G' flag.
If anyone has a gateway entry in their routing table, in which the 'G' flag is not set, and your Internet is working properly, please let us know.

Good luck.