Mobile devices lose 5GH after an hour or so

[LGNilsson]

Well, it's pretty clear that we have a huge problem with Apple devices.
The odd thing is that several of our developers use Apple devices, but it seems like our testing hasn't been as good as it could be.
I don't have an answer at the moment as to what is causing this problem, but as mentioned, we'll get it fixed.

[RGloverii]

You calling me a woman now 

OOPS!  Sorry about that Lars.  Isn't autocorrect fun??

-Robert

[robh5791]

I just had the loss of internet connection happen on my Mac Book Air on the 5 ghz band and found the easiest fix for it at this point was to simply switch to the 2.4 ghz band briefly and reconnect to the 5 ghz band. It seems to be some kind of time out happening on that bandwidth because that switch from one to the other band and then back within a few seconds fixed it for me. I realize this is not an ideal fix but a temporary one while the support team work on a permanent fix.

[Daeithil]

I don't know if it matters, but I've experienced similar issues when using a 5GHz connection on a different router than the Almond+, but only when using my Macbook Air, so it could also be something to do with the laptop as well.  The connection will simply drop and I'll have to switch back to the 2.4GHz before trying the 5GHz connection again, then all is well.

[LGNilsson]

Care to share which router that was?

[robh5791]

That is odd. I have had my MacBook Air since the newest model came out a few months ago and did not have the same issue on my previous router. That could mean it is a firmware issue with some routers communicating with the MacBook but not a universal issue that effects all routers. My previous router was an older one supplied as my modem/router by my DSL provider which is a technicolor c200t.

[RGloverii]

I am using a MacBook Air 11" 2013.  In the course of a typical day, my MacBook is connected to multiple access points and routers, both on 2.4GHz and on 5GHz.  As far as I am aware, the Almond+ is the only router that is exhibiting the disconnect issue for me. 

This is not based on a scientific test; just on observation during the course of actual usage.  I have two 5GHz access points installed in my home (one being the Almond+).  Both have the same SSID.  I have a dedicated DHCP router that both connect to.  At least 90% of the time, I am observing that the MacBook is ONLY connecting to the other access point, even when literally in the same room as the Almond+.  I typically have to turn the MacBook's WiFi off, wait a minute, then turn it back on to get it to connect to the Almond+.

Whatever the issue is, I am confident that it will be resolved in short order.  My purpose here is not to complain, just to inform. 

[coolcatiger]

same issue here too.....

[jlynch3]

add me to the list... iOS, OSX, android... about an hour of access. no impact on 2.4

[madfozi]

it happened to my Samsung Note 2, PS4 and iMac. i end up reboot the device and it is back to normal, then the symptom repeated.

[scottyhoss]

Not sure if this will help anyone, but it did help me.

I have a Nexus 5 which is known to have issues with 5GHz. I changed from the defaults on the A+ 5GHz radio to 40MHz channel width and picked a channel in the middle of the available band - 48 (5.240 GHz in the US). That will minimize interference. Seemed to clear things up for me and the phone stays connected now.

I don't have any other 5 GHz equipment and a quick scan shows my neighbors don't either, so I'm not sure how much interference I was really getting? Maybe it has something more to do with internals of the A+.

Give it a try, worth a shot

Just for sake of being complete, here are all the settings currently working for me

2 GHz

Mode	11NGHT20
Channel	6 (2.437 GHz)
Country	US
Encryption	WPA2-PSK
MAC Filter	Disabled
Client Limit	20
BG Protection	No Protection
Chain Mask	3-chains

5 GHz

Mode	11ACVHT40
Channel	48 (5.240 GHz)
Country	UNITED STATES
Encryption	WPA2-PSK
Cipher	auto
MAC Filter	Disabled
Client Limit	20
Chain Mask	3-chains

[LGNilsson]

Ok, can we do a quick test here? We've had a couple of reports where going from 80 to 40MHz channel width has allowed devices to connect and work just fine, so can those of you that are having this problem try and report back here? If this proves to be a solution, we can report this back as a bug with QCA and get them to fix this.