Unexpected link change

[Wolfraider]

We have an issue with several of our towers where our radios suddenly drop from 100M to 10M. We can try Cisco and Tough Switches and we do not experience this issue. Does any one have any ideas?

Replaced AP (Bullet m2 HP)
Replaced Netonix mini
Replaced cable (Ubiquiti tough cable with tough cable tips)
Installed ferrite beads on both the AP and switch side
Verified grounding, #6 wire from switch to the tower ground, Tower ground is connected with #2 wire to grounding grid

This is on both Rohn 25 towers and water towers where we are the only one using them


Nov 15 20:36:34 Port: link state changed to 'down' on port 4
Nov 15 20:36:37 Port: link state changed to 'up' (10M-F) on port 4
Nov 15 20:36:37 switch[549]: unexpected link change on port 4 (WoodsAccess.AP) from 100M-F to 10M-F
Nov 15 20:36:39 sSMTP[552]: Sent mail for air@netviscom.com (221 2.0.0 Bye) uid=0 username=admin outbytes=522
Nov 15 20:37:01 Port: link state changed to 'down' on port 4
Nov 15 20:37:02 Port: link state changed to 'up' (100M-F) on port 4
Nov 16 08:24:01 Port: link state changed to 'down' on port 4
Nov 16 08:24:07 Port: link state changed to 'up' (100M-F) on port 4

Cable Diagnostics




Pair 1 Status (Pins 1,2) (Pos) (Orange)
Ok


Pair 2 Status (Pins 3,6) (Neg) (Green)
Ok


Pair 3 Status (Pins 4,5) (Pos) (Blue)
Short


Pair 4 Status (Pins 7,8) (Neg) (Brown)
Short






Pair 1 Length
68.90
ft / 21
m


Pair 2 Length
68.90
ft / 21
m


Pair 3 Length
59.05
ft / 18
m


Pair 4 Length
59.05
ft / 18
m






Firmware Version
1.4.5rc8


Board Rev
F

[sirhc]

After the link state changes login to radio and verify radio is not rebooting.

[Wolfraider]

We have verified that the radios are not rebooting, some of the radios will stay in 10M, others will flap back and forth between 100M and 10M

[sirhc]

Please post up the Device/Status Tab screen grab

I am not sure what to say, I have many of these unit in service and this is not an issue for me?

I have had link issues in the past but every time it was a cable/end issue or a bad radio.

[Wolfraider]

[sirhc]

Well swap port 4 with another port and see if problem follows or remains on port 4

[Wolfraider]

We have tried a different port before we replaced the switch, we will get someone to climb the water tower later in the week to try a different port again.

We also have the same issue at another tower.

WS-8-150-AC

Nov 14 22:41:09 switch[2576]: unexpected link change on port 4 (Fair Grove 2.4 S) from 100M-F to 10M-F
Nov 14 22:41:10 sSMTP[2579]: Sent mail for air@netviscom.com (221 2.0.0 Bye) uid=0 username=admin outbytes=541
Nov 14 22:41:32 Port: link state changed to 'down' on port 4
Nov 14 22:41:34 Port: link state changed to 'up' (100M-F) on port 4
Nov 15 15:36:46 Port: link state changed to 'down' on port 4
Nov 15 15:36:49 Port: link state changed to 'up' (10M-F) on port 4
Nov 15 15:36:49 switch[818]: unexpected link change on port 4 (Fair Grove 2.4 S) from 100M-F to 10M-F
Nov 15 15:36:51 sSMTP[821]: Sent mail for air@netviscom.com (221 2.0.0 Bye) uid=0 username=admin outbytes=541
Nov 15 15:37:13 Port: link state changed to 'down' on port 4
Nov 15 15:37:14 Port: link state changed to 'up' (100M-F) on port 4

[sirhc]

Well as I said I would debug it:
1) Change ports, if it follows not the port
2) Change radio, if it still does it not the radio
3) Crimp new ends, it continues not the ends
4) Run new cable, if it continues the you fall into a mess of possibilities


Some of the mess of possibilities:
1) Noise in the 100 MHz range like at FM towers that interferes with Ethernet communications. Possibly some old radio equipment located on the tower such as 2 way or SCADA?

2) Some sort of Static or other current flowing on the cable.

Static on the cable usually occurs during windy events, but does not have to be gale forced winds just a steady wind at the right humidity which normally shows up around sudden temperature changes such as before a rain event, dawn or dusk.

Current such as Earth Ground current on an Ethernet cable can also cause issues.

If the Ethernet cable runs close "in parallel" to an AC power cable can also cause this to occur.