ePMP Software Release 4.7 is now available

Products ePMP
39

I’m surprised I’m not seeing anyone else with the issues we have seen. We have ~300 radios that were upgraded to 4.7.0STA - which I understand is the same firmware as 4.7.0 but with the additional 5.9 frequency.

Since that upgrade, we have had issues all over the place. None of these issues existed before 4.7.0:

  • Device Name will not remain through a power cycle. We can set the device name, it forces a reboot, device name stays but as soon as that radio is hard power cycled, the device name reverts back to what it was pre 4.7.0. I can duplicate this on 100% of our radios (Force 180, 190, 200, 300-16, 300-25). Have not tested on an AP yet.

  • GPS coordinates do not remain through a power cycle. Same behavior as Device Name. They will remain through reboot commands but as soon as you hard power cycle the radio, the GPS reverts to the data that was there pre 4.7.0 upgrade. Can reproduce on 100% of my client radios.

  • Radios crash with various commands pushed from cnMaestro. Sometimes a reboot command will crash them, sometimes a data vlan ID change will crash them (the two things we have tested). When they crash, they are up but will not associate with a tower. We can log in locally, they act weird but are accessible and will not associate until they are hard rebooted with a power cycle. The reboot button in the GUI never initiates a reboot when in this state. We have seen this on Force 180, 190, 200 and various 300-x client radios.

  • So far, 4 client radios have reverted to a factory default state when they were sent a vlan ID change command from cnMaestro. We logged in locally (after a truck roll) and it was totally reset to factory. We have the power sequence factory reset option turned off on all radios.

  • Radios lose the ability to be logged into. We can ping the IP and the IP shows up in the AP/Monitor list but trying to access it from the GUI fails. It’s like there is nothing at that IP. This is only fixed with a hard power cycle. Then it works again. See this on all SM models. No pattern we can tell when they decide to hang.

  • Every single ePMP 1000 AP we have will stop responding to SNMP requests once the airtime utilization creeps up. I can’t tell you the threshold yet but in the evening, our SNMP status board is red on most of our 1000 APs showing SNMP is no longer responding. Ping times to the AP remain unchanged. We do not see this on 2000 or 3000 series APs but we do see it on Force 180 radios in AP mode.

As an ePMP customer for 7 years, this is the worst upgrade we’ve been through. I am ready to revert everything back to a previous version but I am afraid of how many radios we will lose control of in that process and how many more truck rolls we will have to make.

Here is our vlan ID change code that is crashing things. You can see it is not doing anything special:

{
    "device_props": {
        "dataVLANEnable":	"1",
	"dataVLANVID":	"301",
        "dataVLANVP":	 "6"

    }
}
3 Likes