Cisco 2960X and 9200L POE support for Paxton

[Andy Mackie]

Hi,

We currently have a mix of Cisco switches from 2960s, 2960X and now the 9200L.  However we are having issues with our Paxton POE door controllers.

The 2960s work with all of our Paxton Controllers

The 2960X up to hardware version 4 work with the Paxton POE controllers.

2960X Hardware version 5 and higher and 9200L will not power up the Paxton Controllers.

Has anyone else seen this issue? We are currently rolling out more Paxton controllers and struggling to keep up with the demand on power injectors.

Switch details:-

Switch Ports Model SW Version SW Image Mode
------ ----- ----- ---------- ---------- ----
* 1 52 C9200L-48P-4X 16.11.1 CAT9K_LITE_IOSXE INSTALL

Here is the power inline details:-

****#show power inline gig 1/0/23 detail
Interface: Gi1/0/23
Inline Power Mode: auto
Operational status: on
Device Detected: yes
Device Type: Ieee PD
IEEE Class: 4
Discovery mechanism used/configured: Ieee and Cisco
Police: off

Power Allocated
Admin Value: 30.0
Power drawn from the source: 15.4
Power available to the device: 15.4

Actual consumption
Measured at the port: 0.0
Maximum Power drawn by the device since powered on: 0.0

Absent Counter: 0
Over Current Counter: 0
Short Current Counter: 0
Invalid Signature Counter: 0
Power Denied Counter: 0

Power Negotiation Used: None
LLDP Power Negotiation --Sent to PD-- --Rcvd from PD--
Power Type: - -
Power Source: - -
Power Priority: - -
Requested Power(W): - -
Allocated Power(W): - -

Four-Pair PoE Supported: No
Spare Pair Power Enabled: No
Four-Pair PD Architecture: N/A

[Blacktip69]

The answer i think you will find is ;-

terminal pager 0

[nbutlev]

Hi there,

My name is Nikola and I'm a Cisco Engineer with Small Business Devices Team.

To determine the width of the display for the echo input to CLI sessions, use the terminal width User EXEC mode command. To return to the default, use terminal no width. The command is per session and will not be saved in the configuration database.So please  type these commands 

terminal width number-of-characters 

terminal no width

To enable the terminal prompts, use the terminal prompt User EXEC mode command. To disable the terminal prompts, use terminal no prompt command. The command is per session and will not be saved in the configuration databa

[Andy Mackie]

Unfortunately not.  It is the Paxton Net2 device that isn't POE compliant and not an issue with the switch.  

[balaji.bandi]

Good to know and thank you for the information.

what was the solution you arrived as of now ?

[Andy Mackie]

We have gone with..

a) all new Net2 Controllers we purchase make sure they come with a power supply

b) We install a Netgear Unmanaged POE switch into the cabinets where the controllers go to and connect that back to the Cisco Switch

c) Use Cisco Wireless AP POE blocks for single Net2 controllers

d) upgrade the Net2 Controllers from POE to powered.

Cheers

[curtis.rumble]

Just an update to this issue. We did some testing on some new 9200 switches we purchased that are going to be running the Paxton PoE door controllers and ran into this issue again. Previously we set the power statically at 30000 which gave us false hope that this fixed the issue however the door controllers were still not happy with the amount of power being sent to them. I stumbled across a Cisco article yesterday (https://www.cisco.com/c/en/us/td/docs/switches/lan/catalyst9200/software/release/16-9/configuration_guide/npl/b_169_npl_9200/configuring_perpetual_poe.pdf), this identifies that if a PD that doesn't support LLDP is plugged in to configure '2-event' or statically assign power to the device. So the interesting thing is, for the 3850s we ended up having to use 'power inline port 2-event' command to get them to work. With the 9200s this command doesn't work. We had to use 'power inline static max 15400' and this has now resolved the issue. Be aware that you may need to have no other power inline commands set on the interface. 

Interestingly I logged a case with Paxton support, which were pretty unhelpful and said to factory default it. Though I emailed them back after I resolved it and they said this:

"Thank you for the update, that's actually really helpful. I will chat with the product team because currently the statement is "Ensure you disable LLDP at the PoE switch side as Paxton10 PoE splitter doesn't support that. (The same rule is applied to the Net2 PoE splitter.)"."

Hope this helps. 

[Switcher2022]

Hi curtis.rumble -- What code base are you running on the 9200? I have mine at 17.03.05 (Amsterdam) and still unable to get the Net2 controllers to power up with just "power inline static max 15400" on the config. Logs show it goes into a loop of powering the device. Do you mind sharing your full interface config if you have anything else added? I also tried disabling CDP, LLDP and using the "power inline port 2-event" which does still existing on the code version I have running.

[Switcher2022]

Anyone else been able to get the Paxton Net2 controllers working on 9200 switches? 

[ScottKelley2696]

Going to throw my 2 cents here as this is something I've been troubleshooting for a couple of years and have watched Paxnet POE+ handling go from ok to bad to back to ok. I'm fortunately I've got a good vendor to work with but I'd love to talk to an Engineer and their VP at Paxnet and discuss chips and modes.  Just had another discussion on POE recently due to one door not always releasing the striker.

Cisco switching - 2960s and 9200Ls - full POE+ power, not over budget (not even close)

I have 75 doors and trackers throughout my campus, all either on Cisco switching for POE or af/at injectors (passive 15w, active 30W). Just about Pandemic time, all the new doors we installed no longer powered from our switching environment. I could pull a power module from one can and put it into another and watch the behavior go with that controller, not the port. Old units continued working fine, but the new ones didn't in our switching environment. We went to af/at injectors and were able to get them powered but only on POE. No amount of inline power commands would force the power or negotiation to see the NET2 power controller light up the second LED. Then lo and behold, after Pandemic, newer doors started working straight to our Cisco switches again. I have a whole campus of doors only running on POE power in this way. I have 2 doors over all that time that has some problem with the strikers not always getting power to full unlock. I know it is the power controller chips that are used. They (like everyone) had a supply problem, went to cheaper chips, then went back after suppliers caught back up but still use the original chips that don't negotiate well.

Smart switching is too smart for these guys. I've had vendors bring me a cheapy soho POE+ switch and go "See it works" and show both the POE and POE+ indicators light up. Yes, your cheapy dumb switch is using AT and defaulting the negotiation to max. Throw a single inline POE+ dumb injector and it works just fine too.  Unfortunately, similar commands like power inline static don't work to resolve this.

So what you need are non-negotiating, 30W only POE+ (AT) injectors.  Anything else, Paxnet can't handle the negotiation for higher power.

[ScottKelley2696]

Updating with additional info.  No 9200L up - yet.

The paxnet POE supplies are not fully LLDP or CDP compliant - that I have determined.
Enabling LLDP (lldp run) + enable transmit/receive on port + power inline port 2-event - works for my 2960X switches, giving me full 30W power on their PS.  Using power inline static max 30000 did nothing for any model for me.

Like everyone else, not so much on my 9200L (17.05...) and I'm getting ready to update to latest IOS XE.  I got it once to power up against the switch with these commands but subsequent retries didn't work.  I need to check for to see if CDP is conflicting with LLDP on the port.  I will follow up if I crack the 9200 mystery.

[TheRivanKing]

Hi, Did you have any luck resolving this problem? We have recently upgraded to Cisco 9300 switches and are finding that the Paxton net2 air bridges randomly stop working. They connect at first and work for a few days then drop off. We use DNAC assurance and this does not show any disconnection from the network so we are scratching our heads with what could be causing the connectivity issue. If we reset the air bridges they work again for a while but then drop off. 

[ScottKelley2696]

Not yet, I got reprioritized with other projects.  I'm intending to swing back soon though so I'll update here again (got a forum elsewhere that I cross posted as well).

I have noticed the paxnet controllers themselves (regardless of POE in use), will go to sleep on the network or don't respond right away.  Do a simple ping test and the first response is always dropped if it has been several minutes since a communication with it.

[ScottKelley2696]

Had to reprioritize for other projects again.  I have an open ticket with TACS for this now on the 9200L.  They've verified that setting event-2 and static max power should have been the ticket.  Now they want to look at on a call tomorrow so we'll see what they find *fingers crossed

[ScottKelley2696]

Finally got around to submitting this to TACAS and getting a final answer.

According to Cisco Support:
The Paxnet POE+ controllers are non IEEE compliant.  The architecture on 3600, 3800, and 9000 is different enough to not work where as the 2900 line would with event-2 command (simpler power architecture I think)

There is no Cisco level solution for 9000 series switches - Paxnet would need to change their controllers to be compliant.

Back to looking for dumb POE+ rack mountable injectors for closets and demand Paxnet change but I don't own a pike long enough to influence them.