Aerohive AP250 needing reset irregularly

  • 1
  • Question
  • Updated 4 months ago
I have a strange issue with our aerohive AP250s
every week or so I will get an AP that nobody can access the network through.
It has happened with 4 different APs so far. 
When reset the issue is resolved.
There is no indication that there is an issue with the access point.  The only way I know that something has happened is from users.
When the issue occurs clients can still connect to the AP.
I can see the clients in hivemanager ng.
I can ping the AP itself.
I cannot ping any device connected to the AP.
This will happen for all connected devices.
Any device connected to the AP appears to be connected fine.  but can not ping anything else on the network.
Everything returns to normal after a reset.
What troubleshooting steps should I take next time to try to determine what causes this issue?
I have a limited time frame to test, as it is unknown when the issue will happen next
Photo of press

press

  • 16 Posts
  • 0 Reply Likes

Posted 8 months ago

  • 1
Photo of Rob Pritchard

Rob Pritchard

  • 86 Posts
  • 8 Reply Likes
Have you verified if when the problem occurs that any of the clients are obtaining an IP address or not? 

Is there another SSID available that they can try to connect to in order to see if they can access the Internet (that's one of my troubleshooting steps for our school techs, if they're having trouble getting to the Internet on our main SSID then they can try connecting to our guestnet SSID to see if that works just to try and narrow down the problem)?

Are you a school?  If so, are these laptops that are coming out of a cart where the laptops may have been left on while in the cart?  Does restarting the laptop resolve the problem at all?
Photo of press

press

  • 16 Posts
  • 0 Reply Likes
Hi Rob,
All the connected devices have a correct IP address.  I can see the devices still connected in the hivemanger ng (i can see the correct IP addresses in the hive manager as well).
the devices cant ping anything else on the network.  And I cant ping the devices.

There are two SSIDs, however, the issue occurs for both.  When the issue occurs everybody connected to the AP will be affected.

We are a school. Laptops are both byo and out of a cart.  Unfortunately, a restart of the devices does not help.  The only way to resolve the issue (that I found so far) is a restart of the AP.

Its almost as if the AP won't allow any connections to pass through it either way but otherwise functions correctly

Thanks for your thoughts
Photo of Rob Pritchard

Rob Pritchard

  • 86 Posts
  • 8 Reply Likes
What version of HiveOS are you running?
Photo of press

press

  • 16 Posts
  • 0 Reply Likes
it was 7.0r2, I have upgraded all versions to the latest 8.1r1 to see how that performs (seemed ok in small test trial)
Photo of Rob Pritchard

Rob Pritchard

  • 86 Posts
  • 8 Reply Likes
Hopefully the new HiveOS does the trick.  We were running 6.5r6 which was solid and in early August I upgraded our APs to 8.0r1 in our technology office as well as two of our elementary schools - so far so good.  I'm going to upgrade a couple of our secondary schools next week to 8.0r1, and if they run fine for a few weeks then I'll upgrade the rest of our schools.  We totally skipped the version 7 platform.  From some limited personal experience and from what I've read others do, if there are connection issues such as yours, rolling the APs back to a prior version seems to be the trick.  Hopefully in your case upgrading to 8.1r1 will be the resolution.  Keep us posted!
Photo of press

press

  • 16 Posts
  • 0 Reply Likes
The only major issue I have seen on 8.1r1 is with macbooks still on Yosemite.  users were able to connect but unable to ping anything else on the network.  The issue was resolved when updated to Sierra.
Hopefully, the issue is resolved.  I will keep this post updated if it happens again
Photo of Pat ORourke

Pat ORourke

  • 14 Posts
  • 0 Reply Likes
Last yer, we purchased 40 AP 250's. I worked with a dedicated tier 3 support engineer for 3 months trying to get reliable performance out of the devices. Active Directory authenticated users remain connected to the AP with an IP address, have a RSSI level of sub -70Db, less than 35 devices per AP but all network services stall for 30-40 seconds, resume connectivity, stall for 30-40 seconds, resume connectivity, repeat cycle.....

I downgraded the firmware to HiveOS 7.0r2 Bay.131568. After I demoted the AP's firmware, I was able to get better performance for about 2 weeks before I was forced to reboot the AP's. 

I am now finding that I have to restart the AP 250's weekly or sooner. I have a few that are needing to be restarted every 3-4 days. 

In mission critical areas, I have replaced the 250's with the AP120's that were intended to be replaced by the 250's. After redeploying the 120's, I find that I do not have any performance issues from the same network drop. 

99% of our clients are MacBookPro 13" Retina models. 


If you own AP250's, are you satisfied with your purchase? How often are you restarting them and what HiveOS-Firmware combo do you have deployed. If you worked with TS to improve performance, what suggestions did they offer?

Thanks in advance for any feedback. 
Photo of Luke Harris

Luke Harris

  • 265 Posts
  • 18 Reply Likes
Unfortunately Pat I don't have any experience with the AP250 - the only things that I can add to this discussion would be that I had similar occurrences of random drop outs with MacBooks on both AP130 and 230 APs. However this did not occur on Windows devices. 

Often TS will recommend updating to the latest version of firmware etc. but often this does not solve the problem in it's entirety. I would recommend implementing a known 'problem' AP250 in isolation and experimenting with a test version of your active network policy to see if you can isolate the issue down to a particular variable.  
Photo of press

press

  • 16 Posts
  • 0 Reply Likes
So I am still getting this issue, but it is more frequent now.  when it occurs, nobody on the AP can ping anything else on the network.
The devices show the user connected to the AP perfectly, no errors appear in hivemanger ng & shows all clients as connected.
the issue is resolved when ap is reset from hive manager, but this isn't good enough

Is there something I can run to try to find out whats causing this. should I run wire shark on my connected device when this issue is happening, what to look for? 
Photo of Pat ORourke

Pat ORourke

  • 14 Posts
  • 0 Reply Likes
I was sharing my screen with a Tier3 support agent when I had a the stall happen. He could not determine the cause.

I am now needing to restart our heaviest used devices every 2 days. 
Photo of press

press

  • 16 Posts
  • 0 Reply Likes
we did downgrade as we were having an issue with some of our mac users.  We are now running half of our APs on 8.1r2a as a test(i will move the rest over later in the week if things are steady).  I will update here if we see the issue again
Photo of Pat ORourke

Pat ORourke

  • 14 Posts
  • 0 Reply Likes
I am being forced to restart the entire hive 2 times a week. It is no longer just the 250's. The AP120's, 121's, 230's, 250's, 330's and 350's are now riding the wireless roller coaster of doom.
Photo of press

press

  • 16 Posts
  • 0 Reply Likes
Hi Pat, what version of hiveOS are you running now?  we have been running 8.1r2a now for around 3 weeks, I have not had the issue where I need to restart APs on the current hiveOS.  Our APs are primarily 250s and they have become a lot better on the latest build.  (We only have slight issue with some devices which only have 2.4gz wireless cards)
Photo of Pat ORourke

Pat ORourke

  • 14 Posts
  • 0 Reply Likes
I don't have any updates available for our hive past 7.1.
Photo of Pat ORourke

Pat ORourke

  • 14 Posts
  • 0 Reply Likes
I finally updated the 250's to HiveOS 8.1r2a.178408. I also generated new default radio profiles. My restart window has gone from 2-3 days to just over 2 weeks. 

I am testing a different vendor's offering that has been up for 75 days without any issues.