ePMP Force 425 Bridging Issues still continue 2 years and a half later...

Hi everyone!

My name is Bruno and I’m suffering the same issue described here: Force 425 PTP - Bridging Issues?

Exactly the same…

Teh customer had an old Ubiquiti radio link 9 Km distance and with a very low speed, we uprade to a ePMP Force 425 devices and there we go! From one end you can reach everything on the other end but, from the other end you can ping (for example) but if you want to login via http/https to a device like a video camera… no way.

I have recreated the scenario in my lab. Obviously the radios are about 10 meters far. 2 laptops, 2 switches, 2 ePMP devices, one acting as Access Point and the other one acting as Subscriber Module 1 lab firewall simulating the client’s firewall (same VLANs, etc…) It is a very easy deployment. Same issue… from one side you can access everything via http/https, ssh, etc…, from the other side you can get ping responses but not http/https, ssh or whatever communication.
ePMP Force 425 devices are running the last firmware 5.9.1

Just to inform:
No Spanning-Tree issues
No Network issues (Think about just puting back the ubiquiti devices everything works properly it’s a bridge mode config!)
No VLAN mistakes…

It is something inside the Cambium devices but what?

I’ll continue doing my own tests in my lab but any suggestions will be very welcome.

Thanks in advance!

Hello @Bruno_Pasanisi,

if there are no vlans at all I can assume that Managment via Ethernet and Wireless is not enabled. This can prevent some traffic to pass in uplink.

If it is not the case then I suggest you to Raise Ticket in the top right of this page and attach techsupport files from Master and Slave. Our engineer will assist you to isolate it. It is free of charge.

@Bruno_Pasanisi have you tried running 5.9.0? I have had this same issue for years and it either magically fixed itself using 5.9.0 or disabling the acceleration engine on the radios fixed the issue. It would also help if you diagram the ip addressing side of things. Are you using something similar to a /30 management subnet on each side of the link?

Hi @aka
Thanks for your quick reply.
Sorry for the misunderstanding. Yes there are VLANs but what I meant is the switch ports are the same no changes in configuration which is as follows in both switches:

interface 1/1/24
no shutdown
no routing
vlan trunk native 170
vlan trunk allowed all
exit

VLAN 170 is the radio link management VLAN.

I’ve already created a ticket months ago ticket 415969 and it finished with a RMA but the problem still persists with the new unit received.

Any other suggestions wil be very welcome, thanks in advance.

Hi @Jacob_Turner

Thanks very much for your reply, I’ll check this ASAP and I’ll keep you updated.
Attached you can find the diagram.

Hi again @Jacob_Turner

There you go the IP Addressing as I can only attach one image per post…
There are many other VLANs in the production environment, these ones are a little example.

Hi @Jacob_Turner!!

Eureka!!!

You made my day!

It works!!

By downgrading the firmware to version 5.9.0 and disabling the acceleration engine in both radios everything is working properly!

Note: I disabled the acceleration engine in version 5.9.1 but it doesn’t work.

Thanks so much!!

@aka , Could you please notice your colleagues to improve this in further firmwares?

Thanks so much you both!

1 Like

Hi @Bruno_Pasanisi how did you disable the acceleration engine? is it by setting the value to 3 on the cli ?

Hi @IT_Infra ,
Yes! That’s it.

set cambiumAccelerationEngine 3
commit
save

Very Important NOTE: The firmware version is very important, it doesn’t work with any of them, just with a few, in my case with version 5.9.0

Kind Regards,

Hello @Bruno_Pasanisi and @IT_Infra,

we have a fix in upcoming 5.10 release. A beta may be published next week.
Now it should work with acceleration engine enabled.

2 Likes

Hi @aka !

Thanks very much for this update.

We will test in in our lab before upgarding both devices in the production environment.

1 Like