Jump to content

Recommended Posts

Posted
Have you gone through their support process beginning with chat and escalating or posted on their forums. If this was a widespread problem I'm sure their forums would be full of similar posts etc.
Posted

Have version controller are you running and where is it hosted?

 

I've just rebuilt mine onto a Pi, if you have one spare one or can setup a new server and get a few AP's connected to that and see if it makes a difference.

Posted

Moved all Nano HDs from cloud controller to software controller installed on MS server 2019. Two days no disconnections. However today one reported disconnected, responding to ping.

SSH info command did not produce any output.

Did a reset (set-default) and controller had "adoption error".

SSH into the AP and it said:

Status not adopted (http://unifi:8080/inform) ?? I could ping unifi controller from SSH and yet it could not resolve it for adoption???

Set inform to the ip address and it then became available for adoption in controller.

Upgraded firmware to latest one...

I like post from SystemsAD "We have 5 AC-Pro, they never drop off at all"...

 

 

Posted

I have a similar situation. I actually bought a UNIFI switch to see if it would make any difference and it was rock solid for 5 days.. till today one reported as disconnected. (can still ping and connect)

 

I rebooted the port and it came back. Haven t had chance to look further into it yet.

Posted
It doesn't sound like they've contacted support and escalated it through engineering so a bit difficult to blame Ubiquiti. I also haven't heard of this happening in a widespread manner so once again, hard to blame Ubiquiti.
  • 4 weeks later...
Posted

No disconnections over 17 days during Christmas break. Because the site was closed wifi was not used. So, into the first week, 2 Unifi Nano HD already disconnected. They must not like being used then?

Reading though posts on Unifi forums, lots of this type of disconnecting issue. Some have tried static IP addresses only to realise it did not help after couple of days, some said overheating and poor ventilation caused lock-ups.

Lets' be clear about this. When they lock-up they stop responding and lose connectivity with controller, some still reply to ping and can be reset via SSH but most do not respond to pings any more.

There must be a root cause of this equipment failing in this way. So far we tried to isolate power supply as potential issue but they were disconnecting even when powered via Unifi injector. Latest firmware also did not help. Moving to software controller instead of cloud controller did not help. Would Unifi switch make them more stable? If so how and how different Unifi switch is to Cisco SG300-28PP that never had issue with Aruba APs in previous setup. They sit in dedicated wifi vlan, switch ports are set accordingly and they have dhcp with dedicated scope on 2012R2 server. Infrastructure checked many times for any issues, it's just that couple of Unifi AC Pro models have been more stable than Nano HDs. When they lock up, syslog server shows lots of "Kernel errors", so its really back to their design. If their hardware is pushed to the limits by clients and controller demands, lack of processing power will cause lock-ups.

Posted

Since it seems to be when they're being pushed is your switch able to supply the max power drawer they quote for the Nano HDs?

 

Given the price of Unifi switching it has to be worth trying doesn't it? They're so cheap!

Posted (edited)
Since it seems to be when they're being pushed is your switch able to supply the max power drawer they quote for the Nano HDs?

 

Given the price of Unifi switching it has to be worth trying doesn't it? They're so cheap!

 

Cisco switch is on "class limit" and High priority level POE setting and there is spare power in the switch (nominal power 180W, consumed power 84W, available 96W). Typically I see poe port max power allocation 30000 mW and actual consumption approx 6600 mW.

Anyway when APs lock up even on dedicated Unifi injector, power supply issue is ruled out. Btw according to spec sheet, max power for Nano HD is 10.5W (44-57 VDC). More than enough power in Cisco switch.

Edited by evasion
Posted
So did you go back to them and say it didn't help so they could escalate it further?

 

Will leave that to our supplier, they have opened the case, I have opened the case in September and have given up (combination of slow response and suggestions I already tried that did not work).

In the meantime I have added option 43 in DHCP to help with controller discovery function. For example two APs that appeared disconnected, one was pingable so I opened SSH sessions and "info" command returned nothing. Controller host name was resolving properly, ping came with replies. So I had to reset the AP to defaults, then give it set inform command because it could not be adopted due to "inform error". One would think that controller host name in dns would help, but it did not.

If discovery function did not work, it should now thanks to option 43. Will wait and see.

Posted
Will leave that to our supplier, they have opened the case, I have opened the case in September and have given up (combination of slow response and suggestions I already tried that did not work).

In the meantime I have added option 43 in DHCP to help with controller discovery function. For example two APs that appeared disconnected, one was pingable so I opened SSH sessions and "info" command returned nothing. Controller host name was resolving properly, ping came with replies. So I had to reset the AP to defaults, then give it set inform command because it could not be adopted due to "inform error". One would think that controller host name in dns would help, but it did not.

If discovery function did not work, it should now thanks to option 43. Will wait and see.

 

That's up to you but if you're going to rely on someone else then they're probably not going to chase as often or enthusiastically as you are.

Posted
Those nano hd 5 packs were my great hope for better priced wifi, but if it's now into month 13 of them not fixing the problem I guess that's why they were selling them off so cheap
Posted
Those nano hd 5 packs were my great hope for better priced wifi, but if it's now into month 13 of them not fixing the problem I guess that's why they were selling them off so cheap

 

It's an interesting one isn't it - the price was "too good to be true" as it had all the features of the much higher end price APs, but without the high price!

Posted

Just adding that I've had one ap of our 13 Nano HDs disconnect since September. Otherwise they seem to be fine.

 

Can any of yours disconnect or just certain ones? Have you tried a newer one or moving them around?

Posted (edited)
. One would think that controller host name in dns would help, but it did not

 

When you say you have the controller host name in DNS for the Inform - what hostname are you using?

 

I have 2 DNS enteries for mine both resolving to the same IP address.

 

Unifi-Controller for the host name - for the web interface

 

I also have unifi used for inform resolution.

 

Only options I have in DHCP for the APs are 03 - Router, 06 - DNS server and 15 - DNS Domain Name

Edited by mdrabble

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...