Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
SOLVED: Help!! What can cause all TX packets to error?
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Networking & Security
View previous topic :: View next topic  
Author Message
terrycolligan
n00b
n00b


Joined: 24 Oct 2007
Posts: 17

PostPosted: Fri Apr 10, 2009 4:23 pm    Post subject: SOLVED: Help!! What can cause all TX packets to error? Reply with quote

I rebooted my gentoo box last Sunday and lost my internet connection.

I thought at first it was a dhcp problem, but now I think it's something else,
but I don't know what.

I have the problem on two different ethernet cards, with two different drivers,
so I think it must be a configuration problem, not a hardware or driver.
The internet connection worked fine for months before Sunday's reboot.

ifconfig eth1 shows lots of RX packets and bytes, with no errors, but
*every* TX packet apparently results in an error, and there are the TX byte
count is zero.

If I plug the internet into eth0 (and reconfigure), I get exactly the same
situation on eth0. (The problem moves with the cable modem cable.)

Please help! We've been without the internet for 5 days now and are getting
desperate!


Last edited by terrycolligan on Sun Apr 12, 2009 12:51 am; edited 1 time in total
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54808
Location: 56N 3W

PostPosted: Fri Apr 10, 2009 4:39 pm    Post subject: Reply with quote

terrycolligan,

It can be your Ethernet cable, as Tx and Rx data have their own paths, or it can be the router at the other end of the Ethernet wire.

Its unlikely to be your interface cards both failed in the same way, so look at other common elements of the data path.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
terrycolligan
n00b
n00b


Joined: 24 Oct 2007
Posts: 17

PostPosted: Fri Apr 10, 2009 8:24 pm    Post subject: Reply with quote

Thanks for your suggestions -- they are things I hadn't thought of.

Quote:
It can be your Ethernet cable, as Tx and Rx data have their own paths, or it can be the router at the other end of the Ethernet wire.


Okay, I'll try another cable, but it seems unlikely to me to be the problem.

Quote:

Its unlikely to be your interface cards both failed in the same way, so look at other common elements of the data path.


I am using a borrowed Mac PowerBook G4, which is able to connect to the internet
using exactly the same cable.

Also, the same cable has worked for 5 months until the reboot. How can rebooting
damage the cable?

Any ideas on software configuration issues that can cause TX packets to error?
(I have a vague memory of MTU problems, but I can't find any reference in recent
forum posts.)

Thanks again for your suggestions!
Back to top
View user's profile Send private message
terrycolligan
n00b
n00b


Joined: 24 Oct 2007
Posts: 17

PostPosted: Sat Apr 11, 2009 3:45 pm    Post subject: TX errors solved, but still no network connection Reply with quote

Progress update:

I got a replacement cable, one long enough to bypass the power-line extenders,
and the TX errors have gone away.

However, dhcpcd still fails with a timeout, and when I manually configure eth1
and routing, I still have no success -- pinging the gateway still fails.

I don't have a non-ethernet way to transfer files from the Gentoo box, so I am
typing the command output below (i.e., there may be typo's in it).

ifconfig eth1 output:
Code:

eth1      Link encap:Ethernet  HWaddr xx:xx:xx:xx:xx:xx
          inet addr:70.176.212.234  Bcast:70.176.215.255  Mask:255.255.248.0
          inet6 addr: xxxx:xxx ... xxxx/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:13144 errors:0 dropped:0 overruns:0 frame:0
          TX packets:647 errors:67 dropped:0 overruns:0 carrier:134
          collisions:0 txqueuelen:1000
          RX Bytes:875252 (854.7 KiB)  TX bytes:56966 (55.6 KiB)
          Interrupt:18 Base address:0x6000


The TX errors and carrier happened while I was installing the new cable. Once
the new cable was connected, there have been no more TX errors or TX carrier
counts.

So, many thanks! This is a big improvement. However, I still can't ping the
default gateway -- I get '100% packet loss'.

The route output:
Code:

route -n
Kernel IP routing table
Destination       Gateway        Genmask          Flags Metric Ref      Use Iface
192.168.100.0     0.0.0.0        255.255.255.0    U     0      0        0 eth0
70.176.208.0      0.0.0.0        255.255.248.0    U     0      0        0 eth1
127.0.0.0         127.0.0.1      255.0.0.0        UG    0      0        0 lo
0.0.0.0           70.176.201.1   0.0.0.0          UG    0      0        0 eth1


Again, all this worked fine for 5 months until I rebooted last Sunday.

Does anyone have suggestions about what I should try next?
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54808
Location: 56N 3W

PostPosted: Sat Apr 11, 2009 11:38 pm    Post subject: Reply with quote

terrycolligan,

Code:
eth1      Link encap:Ethernet  HWaddr xx:xx:xx:xx:xx:xx
          inet addr:70.176.212.234  Bcast:70.176.215.255  Mask:255.255.248.0


You must not allocate yourself a public IP like that. You may only enter it manually if it was statically assigned to you by your ISP.
As you normally use dhcpcd, you don't have a static IP.
That IP could be already in use. Then not only does your internet not work, nor does the other users.

There is no need to hide your HWaddr. Its unique to your network interface but cannot be used to locate you.
Your public IP is much more useful for that.

Code:
          RX packets:13144 errors:0 dropped:0 overruns:0 frame:0
          TX packets:647 errors:67 dropped:0 overruns:0 carrier:134
still shows a lot of errors.

When you say " ... one long enough to bypass the power-line extenders" does that mean you are, or were using the electricity supply to carry Ethernet data?

As your Ethernet is not working, I'm not sure your assertion that its been ok since the cable was replaced indicates anything.
You can only have errors if you are transmitting data. With a non functional network, you will be transmitting very little or nothing. So the opportunity to generate errors may not arise.

Please check after you restart the interface or reboot. Both will kick off dhcpcd.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
terrycolligan
n00b
n00b


Joined: 24 Oct 2007
Posts: 17

PostPosted: Sun Apr 12, 2009 12:43 am    Post subject: SOLVED! Thanks Reply with quote

First, thanks for your help! Your suggestions were half of the solution.
I am happily composing this at my Gentoo box. Everything now works.

Quote:


Code:

eth1      Link encap:Ethernet  HWaddr xx:xx:xx:xx:xx:xx
          inet addr:70.176.212.234  Bcast:70.176.215.255  Mask:255.255.248.0


You must not allocate yourself a public IP like that. You may only enter it manually if it was statically assigned to you by your ISP.
As you normally use dhcpcd, you don't have a static IP.
That IP could be already in use. Then not only does your internet not work, nor does the other users.


I understand. I had been watching the IP assigned by the Mac's DHCP, and it hadn't changed in a week.
Similarly, the previously working Gentoo connection kept the same IP for weeks. I was trying to figure
out if DHCPCD was the problem -- it certainly appeared so at first.

Once I got DHCPCD to work, I switched to that.

Quote:

There is no need to hide your HWaddr. Its unique to your network interface but cannot be used to locate you.
Your public IP is much more useful for that.

I understand. I didn't 'hide' my HWaddr. I have no way (other than a then-not-working ethernet) to transfer files
or text from my Gentoo box to the borrowed Mac I was using. I had to type everything in by hand, and I used
xx:xx to speed up the process. Similarly, I shorted the IPv6 address, as I don't see how the specific address
was relevant. I used the actual IP address because I thought it might be relevant.

Quote:

Code:

          RX packets:13144 errors:0 dropped:0 overruns:0 frame:0
          TX packets:647 errors:67 dropped:0 overruns:0 carrier:134

still shows a lot of errors.

Yes, but I had been doing ifconfig eth1 between every experiment or change.
All of the errors occurred before the new cable was installed. The 67 errors and 134 no carrier
events all happened before I plugged in the new cable.

Quote:

When you say " ... one long enough to bypass the power-line extenders" does that mean you are, or were using the electricity supply to carry Ethernet data?


Yes, two NetGear XE102 Wall-Plugged Bridge (one on each end). They had been part of the previously
working (for 5 months) system, during which I had no problems and quite respectable download speeds
(300-500 KB/s).

Quote:

As your Ethernet is not working, I'm not sure your assertion that its been ok since the cable was replaced indicates anything.
You can only have errors if you are transmitting data. With a non functional network, you will be transmitting very little or nothing. So the opportunity to generate errors may not arise.


Point taken. What I meant was that, after the new cable, the RX count kept going up, presumably in response to
broadcast packets, the TX count kept going up, kind of as you would expect, as I ping'ed and tried other experiments.
The error count never increased once I started using the new cable.

The final solution happened when (at my wife's suggestion) cold booted both the cable modem and the Gentoo
box at the same time. Interestingly, during the week I've been struggling with this problem, I had rebooted
the Gentoo box dozens of times, including 5 or so cold boots, and I had cold-booted the cable modem about
10 times. None of the reboots solved the problem. Only cold-booting both boxes at the same time worked.
It's as though the Gentoo eth1 hardware/software and the cable modem were infecting each other with some
kind of bad-state virus.

Anyway, it's working again (using dhcpcd)! :D Thanks for your suggestions -- they gave me hope in my
darkest despair.

It's interesting that the borrowed Mac never showed any effects of the bad-state virus. Apparently, the Mac
DHCP implementation is immune to whatever problem was plaguing my Gentoo system.

I'm going to (carefully, very carefully) try various changes to the cabling, and will report if I learn something
else useful, like whether the XE102's were part of the problem, or just innocent bystanders.

In short, the problem didn't go away until I cold-booted both ends of the connection at the same time.
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Networking & Security All times are GMT
Page 1 of 1

 
Jump to:  
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