Wireless Access

Hi,

We are having an issue on new site with the 7005 controller and 325 access points.  Customers complaining that they are dropping connection throughout the day.  See the ticket trail below:

 

Hi there,

I've lost connectivity a couple of times ... connects to wireless network but no internet connectivity. Its happened a couple of times when in the meeting rooms i. also has happened to another Mac user.

Can connect after restarting the Mac - turning wireless on and off doesn't.

 

_____________

 

Just to add to this, I have had similar issues, connection completely dropped yesterday afternoon. Was unable to send emails, use skype etc. Had to reboot and connectivity eventually picked up again.

Have found connections unstable over the last couple of days.

 

______________

 

Happened to me earlier today, and I've heard others saying the same thing. Not sure if being on Apple kit is the case here, but another 2 users (Mac) have had problems that I'm aware of. Turning wi-fi on and off has no impact, and rebooting seems to be the only way to solve it.

 

_______________-

 

Same issue for me, wifi randomly drops and the only way to restore is to reboot (MacBook Pro). Also found if I move to a meeting room or take my laptop home at night, when I return to my desk its reboot time (very annoying if your running a GoToMeeting etc).

____________

 

The wifi shows I am connected but there is no connectivity. The Mac will then completely freeze if you attempt to switch wifi/turn off and needs to be rebooted.

 

Any help would be greatly appreciated as having multiple complaints as you can see

 

Thanks

We need more details:

 

- What version of ArubaOS?

- What model of access points?

- Are the users using encryption?

- What is the transmit power of the acces points?

 

Please see more details on how to troubleshoot disconnects here:  https://blogs.arubanetworks.com/solutions/removing-the-bottleneck-in-wireless/

Hi Colin,

Thanks for the quick reply.

 

Sorry I should have provided these.

 

Aruba OS 8.3.0.0

 

Access Points are 325s

What do you . mean by encryption? As in Anti Virus? They use symantec cloud AV and filevault as macbook encryption

 

Transmit power on 2.4 ghz min EIRP is 6 and Max is 12

5 ghz min EIRP is 5 and Max is 21 ( I believe these are the defaults)

 

Thanks

The wireless encryption on the SSID.

 

You should constrain the maximum transmit power on the 5ghz band to 12 as well.  You should also make sure that in the Virtual AP profile "Drop Broadcasts and Unknown multicast" is enabled.

Change the max to 12 on 5ghz?

Yes drop broadcast and multicast is enabled

 

Sorry - the wireless encryption on one of the SSIDs is WPA2-AES using 802.1x and the other SSID is just WPA2-PSK-AES

You should try changing it to 12 or 15.  Drop broadcast and multicast should be enabled on both virtual Aps.

 

You also need to identify what band each user is on when they have problems.  You also need to make sure that you are only using 40 or 20 mhz channels.

OK i will try that.

Have you heard of this issue of being connected to the SSID with no internet connectivity? Seems a bit strange to me

 

Yes it is enabled on both VAPs

If you have ArubaOS 8.3.0.0, I would consider upgrading to 8.3.0.4 because there are quite  few connectivity fixes.  Without seeing your tech support, I cannot suggest anything more specific than that.

 

Outside of that, you could have a dhcp server that is giving out the default gateway of the clients as an ip address.

 

I have just noticed that when clicking on the 2.4ghz and 5ghz symbol it shows this the default with default power levels

 

However then if i click the RF arm section to which i have created profiles it is different.

See the attached screenshots

 

What ones are being applied?

 

Thanks

 

 

If there is an MM in this network, the radio profile and Airmatch is used.  If there is no MM, the ARM profile is used for transmit power.

There is no mobility master

 

So looking at my RF Arm profile settings do they look ok to you?

 

Is there anything you would recommend changing?

 

What is the best way to tell what channel a client is on when having issues just ask them to report and then check the controller at the time and see what channel they are on?

In general the ARM settings look ok.

 

You need to find out how often the problem occurs to narrow down what could be happening.

It is occuring frequently every day maybe happening to multiple users a couple of times a day.

Also this is my dashboard does everything look ok or normal on here (attached)

 

 

Another point to note

 

On the ground floor we have 5 x 225s

 

On the first floor we have 3 x 325s

 

A user has just complained when going from ground floor walking to the first floor they have no internest access yet the SSID is connected and a reboot was required to fix.

You have too many data points to track down in a forum thread.  You should:

 

- Open a TAC case

- Find out if you should upgrade to 8.3.0.4 to fix any known bugs

- Find an end-user that is willing to work with you or a person who has the most problems

 

It is tough to understand what could be going on without troubleshooting live.

Colin based on the settings i have in the Arm RF profiles do these power levels look correct for one of the APs

 

That does not look unusual.

 

Again, please consider upgrading to ArubaOS 8.3.0.4  Any x.x.0.0 release probably has initial bugs in it.

Ok thanks Colin, I am scheduling in the upgrade so hopefully this helps.

At least all of the settings look ok for now.

 

Thanks for your help so far on this.

Also alot of users complain that the issue is when they are roaming around the office.  Don't know if there are certain settings or levels which work better when roaming or if there has ever been issues in the past with MacBooks and roaming as I haven't had Windows users complain as of yet.

Although the ratio is 44 mac to 24 windows currently

Do you have the local-probe-request-threshold on the SSIDs set?  That could cause an issue if that parameter is too high.

No it is set to 0 which i assume is default for not being used?

 

I can't get my head around this bit though:

 

Which one of the above is taking effect? Are the min EIRP settings different from the min TX EIRP settings in the second picture?

 

I have disabled 80mhz support on the second pic but on the top pic the max channel bandwidth is showing 80mhz ?

 

The radio profile is not used if there is no MM.  The ARM profile is when there is no MM.

Were any rates on the SSID profile removed?  Do you have 802.11k, v or r enabled?

No none of the data rates have been removed should I remove some lower rates ?

802.11k and v is enabled but not r as it can cause issues with Macs I believe

You should not remove lower rates.  The clients should be able to work just fine without them.

 

How many access points in total?

You should type "show ap arm state ap-name <name of ap>" to see how many access points a single AP can see.

Sorry do you mean i should remove the lower rates or not ?

I will try that tomorrow and let you know as I'm no longer on site for this evening.

Is there a command you can use in cli to show all settings being applied to a VAP? So I can see what min max power levels are ? See for sure what 802.11k and v profiles and arm profiles are active rather than checking on the GUI

You should not remove the lower rates.  Clients should be able to work just fine with the defaults.

 

show ap configuration ap-name <name of ap>

Hi Colin,

Do these settings look ok under the WLAN HT SSID Profile:

 

Also uploaded an AP Config result for you to double check if you dont mind.

I don't see anything particularly wrong with those settings.  Did you narrow down the problem(s) and how often and when it started happening?  That is key to figuring out what is wrong here.

It is just happening randomly throughout the day quite frequently.

It appears more than often it is happening when users roam from one floor to the next.

 

Thanks

Do the 325 Aps have a different kind of radio or higher throughput radio that when roaming to and from a 225 it could cause an issue? Is there a setting that I maybe need to look at to turn of VHT 80mhz or something on the controller?

Colin,

Please find attached the AP Neighbours for the 3 APS on the top floor (325) which seem to be seeing the most problems.

Thanks

Two users reported this today:

 

Had the same issue just now. Went to the meeting room 1.5 .  The machine got disconnected there, would not connect back. Came back to my desk, still had the same issue. Then had to restart it now to be able to get connectivity back.

 

--------

 

My issue today occurred today when I went into meeting room 1.5 from my desk (which at the back of the office). So appears to have been triggered my moving location. Was about 10:30 am.

What is your regulatory domain and what are your allowed channels in the 40mhz spectrum?  You might not have enough channels to run 40mhz channels in that regulatory domain even with DFS.

 

 

How can I check this?

 

I don't believe i have ever seen these settings before or even configured them so they will still be the defaults

That is not it, but I think you should go to 20 mhz channels in the 5ghz ARM profile by setting the "allowed bands for 40mhz wide channels" to "none"

 

I dont think you have enough channels for 40mhz wide channels.

Will this not greatly reduce the channels available for users especially on newer devices? there is 200 client connections on this site spread across 8 APs in total. So no one would be able to use the 40mhz band? Would you start even first by disabling the VHT radios or anything or just completely cut the 40mhz ?

Will this not greatly reduce the channels available for users especially on newer devices? there is 200 client connections on this site spread across 8 APs in total. So no one would be able to use the 40mhz band? Would you start even first by disabling the VHT radios or anything or just completely cut the 40mhz ?

Scott McMullan | Senior Systems Engineer | Kainos | DD: +44 (0)28 9057 1517 | s.mcmullan@kainos.com

Going from 40mhz to 20 mhz channels increases the number of channels available.  VHT has nothing to do with this.  Both access points are 802.11ac (VHT) so it does not matter.  Eliminating or decreasing channel contention by going to 20mhz channels is an initial step.

Hi Colin

Going to make this change this evening, but just trying to get my head around this.

 

I always believed that if most of your clients were on 802.11a then it would be best to have the 40mhz channels enabled

 

We have 200 clients per day roughly and the split is 90% and more are on 802.11a with the rest being 802.11g clients.

 

Is this going to still be ok to turn off 40mhz channels?  Can you explain breifly how it means there would be more channels as i am keen to learn about this

 

 

scottm,

 

40mhz wide channels means that you are bonding together two 20mhz channels to make a single channel.  Numerically that means 40mhz uses half of the channels that 20mhz uses.

 

Colin,

On the G arm profile it shows the 40mhz channels also to A Only - does this need changed also or because it is set to A only and A arm profile is set to none this does not need changed?

 

Thanks

It does not need to be changed.  The 2.4ghz band will ignore if it is set to "a-only" or "none".

Hi Scott,

We are seeing a similar issue with AP-345s on ArubaOS 8.3.0.4.  Were you able to resolve your issue?

 

Thanks,

Adam

 Unfortunately not still seeing the issue was hoping it was something to do with our internet circuit but if you are seeing a similar issue then I guess it’s unlikely :(

Adam 

I don't suppose you are using VRRP are you?

 

If so in your log history you are seeing anything strange about certain VRRP IDs not in an INIT State:

 

VRID 10: vrrp is not in INIT state

 

I don't think this has anything to do with the issue i am having but would rule it out if you are not using VRRP addresses and having similar issue

OK I will try this.

Do you think this is better to do out of hours as clients would notice a drop out when the 40mhz is removed and have to reconnect?

Thanks

Scott McMullan | Senior Systems Engineer | Kainos | DD: +44 (0)28 9057 1517 | s.mcmullan@kainos.com

If clients are not reconnecting automatically, I would wait until after hours.  If you are under pressure to fix it ASAP, I would make an announcement and then make the change.

 

We are also seeing similar issues on AP-315 with MacBook Pros only.

iPads, Chromebooks, and Windows work fine.

 

I am upgrading to 8.3.0.5 as we speak, but there is nothing in the release notes about this issue.

 

iann

One user neglected to reserve the IP address of the default gateway in their DHCP scope, so a user would her the IP address of the default gateway and knock everyone off.

Hi Colin,

Checked this out however the first five addresses in each of our scopes are always reserved to make sure this doesn't happen.

 

Thanks

Let me know how you get on if you have upgraded to 8.3.0.5

 

Are you seeing users drop connection periodically ? A lot of our users are on MacBooks also

The upgrade to 8.3.0.5 went well, as expected.

MacBook users haven't had an issue since the upgrade... but I'm not ready to say "it fixed it".

 

iann

Ian, are you able to confirm if the upgrade fixed your issue?

 

We receently upgraded from 6.4.5, which TBO was perfect, no issues, to now 8.3.0.3 with MM, and are experiencing similar issues as described here... MacOS based clients.

 

Ivan,

 

An upgrade from 6.x to 8.x is not supposed to fix anything, really.  If anything it could introduce more variables.  What problems are you experiencing?

Hi CJoseph,

 

We upgraded to MM & 8 to take advatange of centralised management, easier deployment and take advatnage of new APs.

 

So wejhave a dozen offices located globally, all running 6.5.4.9 and we migrated 2 offices to 8.3.0.3 running MM. Since doing so those offices have experieced issues. Offices comprises of around 30 APs and 200 users, located across 5 floors.

 

The default is pretty much setup, that is 20, 40 and 80 used, 18/21 for TX power. Things i have done to tshoot;

 

- Last week disabled 2.4 - (didnt fix anything)

- Today reduced power to 6/15 (to match wha we had on 6.5.4.9) - (didnt fix anything)

- Thinking if disabling 80Mhz becase we had it disabled on 6.5.4.9

- Thinking of upgrading to 8.3.0.8

 

The problems we are experieicing is random loss of network eventhough comptuers show its still connected to WIFI.

Affecietd users equals around two dozen spread across floors.

Clients are comprised on 99% MacOS.

Network restors by itsself after several minutes or when clients disable/enable WIFI on computer.

 

 

Absolutely constrict the max channel bandwidth to 40mhz in the 5ghz radio profile!  Running 80mhz channels with any sort of density can create issues!

Colin,

If Airmatch is enabled on the MM can I still configure the radio profile under the AP Group which will disable the 80mhz channels?

Should I also be reducing the EIRP on the 5ghz radio currently set min 15 and max 21?

Just to be clear this can be set at the site level and won't be over turned by Airmatch?

 

Thanks

The radio profile under the ap-group is how Airmatch determines the range of transmit power for the access points in that ap-group.

Whatever the radio profile configuration is in the ap-group will be enforced. Sites “lower down” simply inherit radio profiles from ap-groups “higher up”. You would just configure the radio profiles higher up and just assign them to an-groups. After that, you would just put aps in those ap-groups.

I would do all my configuration at a single level globally, so it will be available to all devices at that level and lower. The hierarchy is designed so that you do the lion’s share of the configuration at a single level and you reference that configuration at lower levels. You would occasionally make an override for say, a radius server that you would want to be different at a lower level, but in general, the lion’s share of your configuration should exist at a single level. At the controller level you would do the specific interface configuration for that md and/or specific vlan name to vlan number translations for that specific device.

I repeat: The majority of configuration should be done at the same level, and lower down you might make an exception for something like a different radius server, and you would reserve controller-specific configuration like IP addresses under that device (MD) itself.

I hope that even makes sense.

Ivan did you ever get a resolution to your problem ? We are seeing exact similar problems as you described on one of our sites. Mostly Mac users. Also the WiFi diagnostics tool on Mac is really good for troubleshooting look it up if you haven't used it before. Gives you lots of statistics to work off

We are curreently experiencing the exact same issues with one of our sites, we have been battling this issue for the past few weeks with Aruba TAC and our internally engineers.  Aruba ERT team finally sent the issue to the DEV OPS team and are looking into it. Any findings I will update all.

Hi

Can i ask what the makeup of your site is that is having the problem?

 

Are they mostly OS X on macbook?

 

Is the version 10.14.6 (mojave latest) ? 

Seeing weird behaviour on our AP logs:

 

Rebootstrap Information
-----------------------
Date       Time     Reason (Latest 10)
--------------------------------------
2019-09-30 09:30:22 Switching to LMS 192.168.12.5: Missed heartbeats: Last Sequence Generated=8570 Sent=8570 Rcvd=8563; eth Sent=1488306 Drop=0; gre Sent=1488306 Drop=0 First=1; ipsec Sent=0 Drop=0. Last Ctrl message: KEEPALIVE len=77 dest=192.168.12.5 tries=1 seq=21
2019-09-30 09:30:46 Power report failed: UNKNOWN_AP
2019-09-30 09:31:15 Trying new VRRP master for LMS 192.168.12.5. Old VRRP master switch IP is 3.25.199.44
2019-09-30 09:31:40 Trying new VRRP master for LMS 192.168.12.5. Old VRRP master switch IP is 3.25.197.180
2019-09-30 12:19:57 Switching to LMS 192.168.12.5: Missed heartbeats: Last Sequence Generated=10072 Sent=10072 Rcvd=10064; eth Sent=1498442 Drop=0; gre Sent=1498442 Drop=0 First=1; ipsec Sent=0 Drop=0. Last Ctrl message: KEEPALIVE len=77 dest=192.168.12.5 tries=1 seq=23
2019-09-30 12:20:31 Trying new VRRP master for LMS 192.168.12.5. Old VRRP master switch IP is 3.25.197.180
2019-09-30 14:46:58 Switching to LMS 192.168.12.5: Missed heartbeats: Last Sequence Generated=8764 Sent=8764 Rcvd=8756; eth Sent=1507236 Drop=0; gre Sent=1507236 Drop=0 First=1; ipsec Sent=0 Drop=0. Last Ctrl message: KEEPALIVE len=77 dest=192.168.12.5 tries=1 seq=21
2019-09-30 14:47:33 Trying new VRRP master for LMS 192.168.12.5. Old VRRP master switch IP is 3.25.197.180
2019-09-30 16:35:24 Trying new VRRP master for LMS 192.168.12.5. Old VRRP master switch IP is 3.25.199.44
2019-09-30 16:36:10 Trying new VRRP master for LMS 192.168.12.5. Old VRRP master switch IP is 3.25.197.180

This is happening on all APs on this site.

Why is it trying for a new VRRP master? I have no idea where it is getting any of the IPS marked 3.25.xx.xx

Also noticed this:

 

Sep 30 14:47:43  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [277928.401101] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 14:47:43  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [277928.401164] Enabled radar detection on channel 5640
Sep 30 14:47:43  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [277928.401164] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 14:47:43  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [277928.401195] dfs_radar_enable: duration multiplier is 72
Sep 30 14:47:43  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1510645.594953] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 14:47:43  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1510645.595016] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 14:47:43  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1510645.595047] Enabled radar detection on channel 5520
Sep 30 14:47:43  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1510645.595047] dfs_radar_enable: duration multiplier is 72
Sep 30 14:47:43  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1510639.262932] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 14:47:43  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1510639.262995] Enabled radar detection on channel 5660
Sep 30 14:47:43  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1510639.262995] dfs_radar_enable: duration multiplier is 72
Sep 30 14:47:43  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1510639.262995] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 14:47:44  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [277929.401070] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 14:47:44  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [277929.401133] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 14:47:44  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [277929.401164] Enabled radar detection on channel 5640
Sep 30 14:47:44  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [277929.401164] dfs_radar_enable: duration multiplier is 72
Sep 30 14:47:44  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1510646.594922] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 14:47:44  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1510646.594985] Enabled radar detection on channel 5520
Sep 30 14:47:44  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1510646.594985] dfs_radar_enable: duration multiplier is 72
Sep 30 14:47:44  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1510646.594985] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 14:47:44  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1510640.262870] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 14:47:44  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1510640.262963] Enabled radar detection on channel 5660
Sep 30 14:47:44  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1510640.262963] dfs_radar_enable: duration multiplier is 72
Sep 30 14:47:44  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1510640.262963] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 16:36:16  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1517152.280864] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 16:36:16  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1517152.280958] Enabled radar detection on channel 5660
Sep 30 16:36:16  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1517152.280958] dfs_radar_enable: duration multiplier is 72
Sep 30 16:36:16  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1517152.280958] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 16:36:17  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [284442.258646] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 16:36:17  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [284442.258740] Enabled radar detection on channel 5640
Sep 30 16:36:17  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [284442.258740] dfs_radar_enable: duration multiplier is 72
Sep 30 16:36:17  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [284442.258740] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 16:36:17  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1517153.280864] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 16:36:17  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1517153.280958] Enabled radar detection on channel 5660
Sep 30 16:36:17  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1517153.280958] dfs_radar_enable: duration multiplier is 72
Sep 30 16:36:17  KERNEL(Birmingham-AP08-23362@xx.xx.xx.15): [1517153.280958] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 16:36:18  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [284443.258615] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 16:36:18  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [284443.258709] Enabled radar detection on channel 5640
Sep 30 16:36:18  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [284443.258709] dfs_radar_enable: duration multiplier is 72
Sep 30 16:36:18  KERNEL(Birmingham-AP01-23366@xx.xx.xx.16): [284443.258709] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 16:36:20  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1517162.538409] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 16:36:20  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1517162.538471] Enabled radar detection on channel 5520
Sep 30 16:36:20  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1517162.538471] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 16:36:20  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1517162.538503] dfs_radar_enable: duration multiplier is 72
Sep 30 16:36:21  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1517163.538409] dfs_init_radar_filters: dfsdomain=2, numradars=9, numb5radars=0
Sep 30 16:36:21  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1517163.538471] dfs_radar_enable: firpwr=0, rssi=0, height=0, prssi=0, inband=0, relpwr=0, relstep=0, maxlen=0
Sep 30 16:36:21  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1517163.538503] Enabled radar detection on channel 5520
Sep 30 16:36:21  KERNEL(Birmingham-AP02-23369@xx.xx.xx.10): [1517163.538503] dfs_radar_enable: duration multiplier is 72

Which ties in with the AP timestamps below for AP02:

 

2019-09-30 16:35:24 Trying new VRRP master for LMS xx.xx.xx.5. Old VRRP master switch IP is 3.25.199.44
2019-09-30 16:36:10 Trying new VRRP master for LMS xx.xx.xx.5. Old VRRP master switch IP is 3.25.197.180

The heartbeats missed message represents a connectivity issue between and access point and am MD on the wired network. The radar message is an AP turning radar detection on when it first connects to an MD.

This thread has been open for almost a year, and it looks like your access points are having a connectivity issue to your controllers so I am closing it.  

 

We too have users reporting this since migration to Mobility Master : “The problems we are experieicing is random loss of network eventhough comptuers show its still connected to WIFI.”

Only difference again is like you said the 80mhz channels being enabled as before they were disabled – but this time I have left them enabled as I believed Airmatch being enabled it could pick the best power/freq & channels based on surrounding environment.

Scott McMullan | Senior Systems Engineer | Kainos | DD: +44 (0)28 9057 1517 | s.mcmullan@kainos.com