WS-12-250-DC added feature

[bgoode]

I'm loving the new WS-12-250-DC that arrived last week!

One idea I'd like to see added to future firmware is a "bounce now" feature....port x poe off...wait x seconds...port x poe on. It would retain/assume the voltage level set prior to the bounce. Maybe incorporate it into the present "Port Bounce" tab by adding a "Now" to the pull-down "When" list.

[lligetfa]

I don't have a DC model but it exists in all the other models so I cannot see them taking it out.

Selection_004.jpg (183.85 KiB) Viewed 17163 times

[sirhc]

I'm loving the new WS-12-250-DC that arrived last week!

One idea I'd like to see added to future firmware is a "bounce now" feature....port x poe off...wait x seconds...port x poe on. It would retain/assume the voltage level set prior to the bounce. Maybe incorporate it into the present "Port Bounce" tab by adding a "Now" to the pull-down "When" list.

Yes Les is correct, try clicking on the little gear to the far right of each port row on the Status page.

I also believe in v1.3.X that gear exists on the Ports Tab as well.

It's called Port Bounce which powers off the POE for several seconds then re-applies the POE power.

[bgoode]

Thanks....I found the "Port Bounce" function in the gear icon on both the Status page and the Ports page.

The function seems to work perfectly when selected on the Ports page but not when selected on the Status page. Selected on the Status page, the poe does turn off but never turns back on. You have to go back to the Ports page and reset the poe voltage and save.

I'm also getting a frequent pop-up "Attempting to reestablish communication with the device ..." and
"The configuration has changed on the device. You will not be able to save any configuration changes. Click here to reload the configuration."

Using Firefox 40.00 and IE 11.0 with the same result...

[sirhc]

Thanks for the notice, I sent a request to Eric to investigate and fix it

Please see if Chrome yields the same result and let me know

[bgoode]

Same behavior with Chrome....

[sirhc]

ok thanks will fix

[Eric Stern]

This is being cause by the PoE Smart feature that disables ports that are unplugged. That part of PoE Smart is being removed so this won't happen anymore.

[bgoode]

Upgraded to 1.30 final.

The "Bounce" function has changed significantly from the above mentioned behavior. Now, the bounce on-off-on cycle is way too fast...just a couple of milliseconds rather than the stated 5 seconds. Its so fast that the poe indicator light on the port doesn't go off.

...and...

I think that has caused a serious problem with one of the ports. I didn't change anything from the original lab setup mentioned above. Port 3 is connected to an AF5x powered by 24v poe (not 24HV). As I did with the original post, I tested the new 1.30 final bounce feature several times, using different browsers to see if the new short cycle times would change (they didn't). After 4 or 5 times bouncing the port...it would not power up. I ran the cable diagnostics and I think I found the problem...

Pair 1 Status Open
Pair 2 Status Short
Pair 3 Status Open
Pair 4 Status Open

Plugged the same cable and same radio into Port 2 and everything worked fine (I didn't repeat the bounce tests fearing I would create the same problem).
With the AF5x plugged into the new port, enabled 24v poe, the cable diags show:

Pair 1 Status Short
Pair 2 Status Ok
Pair 3 Status Short
Pair 4 Status Short

Poe off, with cable and radio connected:

Pair 1 Status Open
Pair 2 Status Open
Pair 3 Status Open
Pair 4 Status Open

So...it appears that the new very rapid power cycling may have blown pair 2...that port will not power any ubnt device AF5x, Rocket, or NS.

[sirhc]

First OFF why on earth would you be powering an AF5X with standard 24V when the port you are in support 24VH? -

OK I just LAB'd this and this is what I found:

I think that has caused a serious problem with one of the ports. I didn't change anything from the original lab setup mentioned above. Port 3 is connected to an AF5x powered by 24v poe (not 24HV). As I did with the original post, I tested the new 1.30 final bounce feature several times, using different browsers to see if the new short cycle times would change (they didn't). After 4 or 5 times bouncing the port...it would not power up. I ran the cable diagnostics and I think I found the problem...

Pair 1 Status Open
Pair 2 Status Short
Pair 3 Status Open
Pair 4 Status Open

Plugged the same cable and same radio into Port 2 and everything worked fine (I didn't repeat the bounce tests fearing I would create the same problem).
With the AF5x plugged into the new port, enabled 24v poe, the cable diags show:

Pair 1 Status Short
Pair 2 Status Ok
Pair 3 Status Short
Pair 4 Status Short

Poe off, with cable and radio connected:
Pair 1 Status Open
Pair 2 Status Open
Pair 3 Status Open
Pair 4 Status Open

If you do a cable diagnostics with GOOD cabling (NO Ethernet Surge Protection) with POE OFF with an AFX
Pair 1 Status Open
Pair 2 Status Open
Pair 3 Status Open
Pair 4 Status Open


If you do a cable diagnostics with GOOD cabling (NO Ethernet Surge Protection) with POE ON with an AF5X
Pair 1 Status OK
Pair 2 Status OK
Pair 3 Status OK
Pair 4 Status OK


Above are my cable diagnostics I just ran here in the LAB, yours are WRONG!?

Not sure why you are getting the cable diagnostics you are getting unless you are using a surge suppressor???????

The only devices that report Shorts on pairs 3 and 4 are LEGACY 10/100 devices, NOT 1G devices

I WOULD BE CONCERNED WHY YOU ARE SEEING SHORT SHORT ON AN AF5X

If you do a Port Bounce the POE power cycle is indeed not 5 seconds but it does reboot the device as called for.

You more then likely damaged your Ethernet Transformer because you were repeatedly "in rapid succession" issuing the port bounce over and over again.

This exposed the Ethernet transformer to IN-RUSH current of the AF5X on just a single pair since you were only using 24V instead of 24VH.

I guess I could do the OWL test to see how many rapid repeats are needed to POP it but I would not do that.

You can power AFX with a standard 24C port if the wire is less than 150' but WHY WOULD YOU SINCE YOUR SWITCH SUPPORTS 24VH?????????