How to Read Cable TDR Testing Result

[PolarPanda]

Hi network gurus, 

           I have a hard time to determine which part cable is broken at below. Do both short and open mean that we have cable issue from patch panel to the Ethernet port on the wall (the cable run inside the wall/ceiling), and the cable from switch port to patch panel is normal? Thank you. 

Swtich01#
Interface Speed Local pair Pair length        Remote pair Pair status
--------- ----- ---------- ------------------ ----------- --------------------
Gi2/0/3  100M  Pair A     0    +/- 1  meters Pair A      Normal
                Pair B     0    +/- 1  meters Pair B      Normal
                Pair C     47   +/- 1  meters N/A         Short
                Pair D     49   +/- 1  meters N/A         Short

Switch 02#

Interface Speed Local pair Pair length        Remote pair Pair status

--------- ----- ---------- ------------------ ----------- --------------------

Gi1/0/20  10M   Pair A     0    +/- 1  meters Pair B      Normal

                Pair B     0    +/- 1  meters Pair A      Normal

                Pair C     22   +/- 1  meters N/A         Open

                Pair D     23   +/- 1  meters N/A         Open

[Reza Sharifi · Accepted solution]

Hi,

Have a look at this post. Leo published a good document on the TRD testing.

https://community.cisco.com/t5/networking-documents/how-to-use-time-domain-reflectometer-tdr/ta-p/3119327

HTH

[Deepak Kumar]

Hi,

This will help you: https://community.cisco.com/t5/networking-documents/how-to-use-time-domain-reflectometer-tdr/ta-p/3119327

[Martin L]

Do both short and open mean that we have cable issue?

- Not for 100 Mbps but Yes for 1000 Mbps.

based on my experience, this depends on speed and type of device at the end;  For 100Mpbs it is normal to have short/short if the IP camera is at the other end.  If you replace camera with a PC, short/short pairs would change into Open/Open.  so, Gi2/0/3 appears to have IP camera or IP phone installed at about 48 meters

In case of 10 Mbps, Open/Open is OK too. I don't think Pairs C, D are used on those 10M.

Regards, ML
**Please Rate All Helpful Responses **

[Leo Laohoo · Accepted solution]

Swtich01#
Interface Speed Local pair Pair length        Remote pair Pair status
--------- ----- ---------- ------------------ ----------- --------------------
Gi2/0/3  100M  Pair A     0    +/- 1  meters Pair A      Normal
               Pair B     0    +/- 1  meters Pair B      Normal
               Pair C     47   +/- 1  meters N/A         Short
               Pair D     49   +/- 1  meters N/A         Short

The cable fault is located 47 or 49 metres away from the switch. 

Switch 02#

Interface Speed Local pair Pair length        Remote pair Pair status
--------- ----- ---------- ------------------ ----------- --------------------
Gi1/0/20  10M   Pair A     0    +/- 1  meters Pair B      Normal
                Pair B     0    +/- 1  meters Pair A      Normal
                Pair C     22   +/- 1  meters N/A         Open
                Pair D     23   +/- 1  meters N/A         Open

Cable fault is 23 metres away from the switch. 

NOTE:  

• If the devices are only 10/100BaseTx then the result matches, i. e. Pairs C & D are "open". 
• Starting with 3650/3850 (and newer), Cisco has "re-defined" the TDR functionality.  This feature will measure the distance of a BAD pair.  Good pair will "always" return with a distance of "0".  

[PolarPanda]

Hi Leo,

              THank you for the reply. Much appreciated. I looked at the article you wrote that the other two gurus mentioned in this post. I have a couple of follow up questions.

1. I don't understand this part. In my examples, the errors are C and D. Does it mean my cable issues relate to PoE and GigEtherent interface?

Pairs	Function
A	This pair controls whether or not the port should go up or down.
B	Protocol-level and controls FastEthernet.
C	Power over Ethernet (PoE)
D	GigabitEthernet

2. Basic on your article and reply, the broken cable should be either in horizontal line or Red. Am I correct? 

Thank you. 

[alessandro.s]

Hi PolarPanda,

as i know it means that pair C is the cable pair used to bring PoE to devices, it should be the white/blue-blue cable pair. Pair D should be the white/brown-brown pair of the cable, without this pair all must work properly except 1Gbps speed, you can connect only at a max speed of 100Mbps.

Regards,

[Leo Laohoo]

The question I'd like to ask is this: Are the downstream clients capable of 1 Gbps or not.
I have seen a lot of wired clients that can only support 10/100BaseTx and the TDR will show exactly the same result of having Pair C & Pair D as "short" or "open".

[Ciscorocks]

Hi Leo,

What if you have a 100M device but pairs C and D are showing as short with pairs A and B are Normal?

--------- ----- ---------- ------------------ ----------- --------------------

Gi1/0/38  100M  Pair A     0    +/- 1  meters Pair B      Normal

                Pair B     0    +/- 1  meters Pair A      Normal

                Pair C     40   +/- 1  meters N/A         Short

                Pair D     39   +/- 1  meters N/A         Short

[Leo Laohoo]

What platform is this?  Is this a switch that runs on IOS-XE (3650/3850, IE3300/IE3400, Catalyst 9200/9300)? 

If it is, the result is not to be "trusted".   There is a TDR "bug" affecting switches running IOS-XE versions 16.X.X and 17.X.X.

The bug is the switch will provide incorrect output.  See examples below: 

(Above)  Pair A & B distance is "0".

(Above)  Pair "A" distance is (significantly) more than Pairs B, C and D.

  

[Ciscorocks]

This is a 9300 running IOS-XE 16.6.5. So we should not trust the test? My question is even if the endpoint is running at 100mb do we still look at pairs C and D? and if they are short we have a bad cable and they should be open? Or are pairs C and D not relevant at all since the device is running at 100mb? Thanks.

[Leo Laohoo]

---

@Ciscorocks wrote:
This is a 9300 running IOS-XE 16.6.5. So we should not trust the test?

---

I know I would NOT. 

100 Mbps only tests Pairs A & Pairs B and Pair D will always be "short". 

[6502]

I have the following TDR results from a 9200L running IOS-XE 17.9.1r.  The port (gig1/0/12, an RJ-45 port) is connected to a 10/100 only capable, non-PoE device:

I read @Leo Laohoo 's caveats about the IOS-XE 16 & 17 caveats, but this output doesn't look like what's being reported in those bugs.  Given that it's only a 10/100-capable end device, does it matter if pair's C & D are reporting as "shorted"?  Pair A reported as "open" appears to be an issue, as well as the remote pairs reporting N/A.

[Leo Laohoo]

IOS-XE has a "bug" with TDR.  This is particularly evident if the switch is PoE-capable. 

Turn off PoE to the port using the command "power inline never" and re-run the TDR.  After the result is out, turn back PoE to the port using the command "no power inline never".  

I have reported this bug feature to a 9300 Technical Product Manager and provided the following (below) screenshots: 

(Above)  Pair A & B distance is "0".

 Please refer to CSCvw97924 & CSCwd97177.