High error count on a port

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

Patrick

Administrator
Staff member
Dec 21, 2010
12,512
5,800
113
For some reason I was looking at the Netgear XS712T's stats page. Check out xg8!
upload_2015-11-25_6-19-26.png

Very interesting since there are at most 3m cable runs. I did a bad job and forgot to label the ports in the UI (will do it this weekend.) I am however a bit alarmed at the error rate.
 

Mike Bailey

Member
Sep 24, 2015
34
24
8
34
New York
Do you have a cable tester on hand? I'd pull the cable and test it.


If you have one of the fancier ones, you can also check to see the loss per meter on the cable.

Also, is the cable routed near any power cabling? If it is, try to separate them asap. Power cable near any kind of data cabling can lead to bit errors and, in turn, packet errors.

That's a solid .045% error rate which is way above what Ethernet considers acceptable BER. You've got a nearly 45 *billion* times higher error rate.
 
Last edited:

Lost-Benji

Member
Jan 21, 2013
424
23
18
The arse end of the planet
The fact they are on the Rx packets, the device Tx-ing may be suspect. Don't bother with testing Patch leads, golden rule applies: If in doubt, THROW it out. Swap the cable, reset stats and see what happens. Make sure cable is not run near power or any other electrically noisy devices. Take photo of cable run/setup if you are not sure what might be electrically noisy.
I know in the past, you raved about a bloody PDU that also had a network Switch in it....
 

Patrick

Administrator
Staff member
Dec 21, 2010
12,512
5,800
113
I know in the past, you raved about a bloody PDU that also had a network Switch in it....
It works well as a dumb IPMI NIC switch actually, but demands are much lower there. It is actually generating fewer errors than the XS712T.

Need to get to the datacenter to see what is up with this. Seems like they are FCS errors.
 

mstone

Active Member
Mar 11, 2015
505
118
43
46
For some reason I was looking at the Netgear XS712T's stats page. Check out xg8!
View attachment 1246

Very interesting since there are at most 3m cable runs. I did a bad job and forgot to label the ports in the UI (will do it this weekend.) I am however a bit alarmed at the error rate.
1) I suggest using LLDP to keep track of what's attached to each port. In the meantime you should be able to figure it out from the MAC table on the switch, no?
2) FWIW, I've also had problems with cables that were too short (<<1M)
 

maze

Active Member
Apr 27, 2013
576
100
43
Found out what was wrong?

I'v seen quite a few patch cables being bent and twisted throwing out stuff like that. If you get a cable that might not have been that great from the beginning, it might not take much to have it pop out with errors like that.

But yeah, as @Lost-Benji said.. if it doubt, throw it out. The time you spend troubleshooting isnt worth the cost of a new cable really.. is it?
 

Patrick

Administrator
Staff member
Dec 21, 2010
12,512
5,800
113
Found out what was wrong?
It has not incremented in terms of error count after 3x more traffic. It may have been from an older config as I did forget to reset the counters when it moved to the rack.

I was going to head to the datacenter today, but there is a 49ers game so I might pass on that and wait for less traffic.
 

Biren78

Active Member
Jan 16, 2013
550
94
28
I couldn't figure out why a NFL game would matter to you, then I re-read. Good luck. Not like the 49'ers are gonna win anyway.
 

Marsh

Moderator
May 12, 2013
2,645
1,496
113
Hard to figure which great mind would build a stadium in the heart of Silicon Valley.
Traffic is already a nightmare without help from football event.

On paper, the idea sounds great and profit for the franchise owner. I heard on the news, there are many empty seats.