Feature Request - ePMP3000 Automatically connect back to Preferred APs

Hi Team,

Would it be possible for ePMP SUs to periodically re-scan from their preferred AP list and reselect based on AP list order?

Reason for this request is that if their preferred AP loses connectivity (e.g. goes offline, GPS sync, interference, whatever) and they connect to secondary AP, we then have to manually force them back to their preferred AP.

Thanks

4 Likes

I’m guessing this will also cause a couple of minutes or so radio down time when they detect the other SSID, and switch AP’s but as long as operators know this it’s not a problem. Could perhaps have a configurable min/max retry timeout and implement it so that the retry timeouts double each time up to a maximum? e.g. 5, 10, 20, 40, 80, 160, then once every 3 hours…

This feature has been marked by Cambium as “under consideration” for over 6 years now… with no movement yet. Actually, there’s a huge list of features which are “under consideration” for years and years…

2 Likes

It sure is nice on the Canopy and 450 to be able to do this. Hate that on ePMP we have to only allow each radio 1 AP to connect to else you end up with everyone piled onto the wrong AP every time you reboot an AP. And as pointed out, it’s even more frustrating that you have to manually move them back to the preferred AP which can be a PITA or a Huge PITA.

That said, I haven’t messed with PtMP on Ubiquiti or Mimosa or Tik yet . Do any of them do this (is it something the N/AC generic chips just can’t do maybe ? )

A couple WISPA Events ago, my Boss went to the Cambium booth and ask about this. He was told that it is impossible on the 802.11 based products to initiate a re-scan from the SM. It must be dropped by the AP. So this feature will never be available (at least that was the answer a few years ago)

There is a re-scan button on the web ui, so the functionality is there manually. It would be trivial to script this in the underlying OS I would think. A simple cron job would do it.

2 Likes

That just shows an embarrassing lack of imagination. Even if that’s true - EVERYONE else has figured a way around it.

For example: Put an option in the GUI on the non-primary SSIDs which says something such as “Retry primary SSID every [xx] minutes” and if there’s actually some reason it can’t rescan, then program the SM to do a ‘reset the wireless interface’ like the watchdog is capable of, and the SM will reconnect with the primary AP, if it’s available again.

There IS a way to accomplish this if you want to, even if it takes a bit of imagination.

2 Likes

We have a script that runs in the middle of the night that goes to each SM and initiates a re-scan. Crude, but better than nothing.

1 Like

Bet the people that regularly use the service late at night just love that. We had an issue once that caused me to have to knock everyone offline and cause a 1 minute outage at 3:00am 3 nights in a row and we had half a dozen people calling in outraged wanting to know if it was going to be an every night event.

We also use to limit PPPoE connections to 76 hours so connections would drop for just a few seconds every 3 days and 4 hours ( extra 4 hours there so it didn’t happen the same time every 3 days ) so that disabled accounts would be forced to re-auth (and fail). Got complaints constantly about the connection dropping “every other day” even though it was literally 20 seconds and at most once every three days and only the same time of day every 18 days (because it would be 4 hours later every 3 days)…

1 Like

I agree this would be a very useful feature when a CPE is knocked off their ideal access point

It could easily be handled within the scope of the ‘watchdog’ which is already there, already doing things in the event of certain failures. Assuming there is actually some reason Cambium can’t program it properly, It still should be relatively trivial to program the watchdog to check one more thing and reset the wireless on a SM if it’s not associated with the primary SSID after a certain period of time.

It’d still be an ugly kluge, but would at least be a functional kluge which is preferable to the other options (1-knocking everyone offline on all AP’s, 2-leaving SM’s on non-optimal Sectors, or what most people do… 3-simply not use the backup SSID feature, (since it’s really not finished without some mechanism to climb back to the top on it’s own)

Unfortunately this thread shows Cambium is just like every other middle of the road vendor, doesn’t listen to what it’s customers request.

1 Like

(There is a similar / near identical thread that’s been requesting this feature for years now, with many votes)

1 Like