interesting wireless power injector issue
I wanted to let you know about an issue we encountered with a power injector.
We had an access point that was bursting traffic on a switch port (2-10 mbits)
When it had no clients associated.
I believe it may have been causing or contributing to some transitory spanning tree loops
In the local wiring closet’s redundant link causing an errordisable condition that would repeat itself every time
We enabled the link.
After troubleshooting many things we finally changed the power injector and
The condition subsided.
We examined the injector and found a bent pin.
I believe this pin caused a “short”.
The link still “worked”
I believe that caused stray voltage (noise) to cause the phantom “bandwidth” On the link.
My question is:
Have you ever heard of anything like this occurring with a power injector ?
We had an access point that was bursting traffic on a switch port (2-10 mbits)
When it had no clients associated.
I believe it may have been causing or contributing to some transitory spanning tree loops
In the local wiring closet’s redundant link causing an errordisable condition that would repeat itself every time
We enabled the link.
After troubleshooting many things we finally changed the power injector and
The condition subsided.
We examined the injector and found a bent pin.
I believe this pin caused a “short”.
The link still “worked”
I believe that caused stray voltage (noise) to cause the phantom “bandwidth” On the link.
My question is:
Have you ever heard of anything like this occurring with a power injector ?
rm -rf /
Comments
-
tiersten
Member Posts: 4,505
It wouldn't be counted as traffic since noise won't look anything like a valid ethernet frame. -
darkuser
Member Posts: 620 ■■■□□□□□□□
understood .... but changing the injector fixed the problem
if voltage crossed it could impair connectivity
cisco engineering is looking ito itrm -rf / -
networker050184
Mod Posts: 11,962 Mod
Might be unrelated and the issue just resolved when it was powered down while switching the injector. This is assuming you didn't try a power cycle first.An expert is a man who has made all the mistakes which can be made. -
darkuser
Member Posts: 620 ■■■□□□□□□□
networker050184 wrote: »Might be unrelated and the issue just resolved when it was powered down while switching the injector. This is assuming you didn't try a power cycle first.
we changed both the ap and switchport
problem followed it
when we brought the "injector" back to our desk
we plugged it in and it instantly produced 2mb "bandwidth"
my best guess is the bent pin caused voltage to leak onto somewhere it shouldn't have been.
i am personally going to deliver it to cisco engineering
to reproduce the problemrm -rf /