received disassoc sending station leaving (8)

dyms · Thu May 15, 2014 11:21 am

Hello,
can you please help me. I have a issue with WLAN router : RB951G, OS 6.12.
Message debug : mac@wlan1: disconnected, received diassoc: sending station leaving (8). Configured channel : 11 20/40. I tried each solutions which I could find, but no luck.
It happens only with microtik router. Do you know how to solve it once and for all?

Thanks

scotthammersley · Sat May 17, 2014 11:46 pm

There are several reasons why it might be doing that, but to give any suggestions, we will need to know a little about your setup. Can you post your wireless settings from both the AP and the client?

Trackboy · Fri Jun 13, 2014 12:39 pm

My problem is the same, the log is full of with these and similar messages.I changed frequency, but did not help.I am trying to figure out, what the problem is.

Fri Jun 13, 2014 4:00 pm

what are the wireless clients?
that log message means that the client just disconnects. If it is some mobile client - it just goes to sleep mode (turns off the wifi to save battery life).

Trackboy · Fri Jun 13, 2014 9:51 pm

Mobile phones, and laptops, the log contain some error messages:

-disconnected, group key exchange timeout
-disconnected, unicast key exhange timeout
-disconnected, received disassoc: sending station leaving (8) (3)
-wlanX: data from unknow device XX:XX:XX:XX:XX:XX, sent deauth

Save battery mode is off
The signal strength is not bad from the clients, -60-80 dBm, Tx-power set to default in the RB 751U-2HnD

Mon Jun 16, 2014 2:12 pm

Make sure that you use AES encryption and not TKIP.

How often does it happen?
When it happens all the clients are disconnected or one of the clients and the rest of the clients continue to work without problems?

Trackboy · Mon Jun 16, 2014 10:51 pm

I am using AES, some clients working well, but some disconnected every 10 minutes
This is a very annoying problem, i am trying to guess what could be the problem, but i have no idea.

saintofinternet · Wed Jul 09, 2014 5:01 am

this is now happening with almost 10 AP's i have deployed for indoor purpose.

very annoying... clients have started doubting router's capabilities.

requesting mikrotik to find a solution please.... Normis!!! help!!!!

Zorn · Thu Dec 04, 2014 2:57 pm

I have the same problem...
Version 6.22
Firmvare 3,17 (I upgrade it to 3.19, but it showing current 3.17 after reboot - it is strange)
rate-selection is advanced
only AES
Band B/G
I explore different setting to solve:

wlan sending station leaving
data unknown device sent deauth
UNICAST KEY EXCHANGE TIMEOUT

but very offen ping upping to 2000ms and devices connection breaked. All wifi mobile devices is off (andriod and IOS), only laptops and printers.

Is anybody solved this?

saintofinternet · Wed Dec 10, 2014 12:30 pm

set the DATA RATES to Legacy and keep just single Chain (CHAIN 0) selected.

reboot the router... and check the stability

please update the result here.

AndrewOo · Fri Dec 26, 2014 11:41 am

Same problem.
I have one active user, he got disconnections. i see in logs what he connected, after 2-5 mins he got disconnections again. Also i see him as active user in hotspot window. Tested on two devices iphone 4s and iphone 5s. POS 6.23 RB95. Plz Help!!!

Zorn · Tue Dec 30, 2014 10:49 am

After I lowered the output power of the wireless card to 15 (default 17) - this problem wasn't appeared.

AndrewOo · Tue Jan 13, 2015 2:19 pm

After I lowered the output power of the wireless card to 15 (default 17) - this problem wasn't appeared.

Can you clarify what exactly you make setup power on wireless card?

yarda · Thu Apr 23, 2015 1:20 pm

It coud be fine define on Mikrotik wiki "Standard" for setting working and stable wifi on Apple, Samsung, HTC and so on. Many vendors aren't compatible with default setting wifi on mikrotik include wpa2 security. Please give us simple parametters or simple import for this. We can't use any "unify" or another but functional solutions for clients if we have Mikrotik.

incahex · Mon Apr 18, 2016 12:54 am

Sorry for bump an old post.
But I have similar problems with and iphone 6 and a mikrotik AP (RB912UAG-2HPnD)
The others devices in the wifi (androids and laptops) are constantly connected but the iphone is arround 5 minutes and then print: received disassoc sending station leaving (8) in de log.
Again and again.

I think is some related with apple wireless chip power management and mikrotik AP wireless config.

If you need more info please ask for it but I think is a common problem with apple devices.
Any solution?
Thanks.

Jan89 · Sun Apr 24, 2016 5:32 pm

Hi,

I also had a problem with disconnects iphone from network, but my network setup looks as follows:
Internet -> Routerboard 850gx2 -> Wifi AP Cisco Aironet 2702.
My iphone was disconnecting from wifi every 10 minutes. The problem was with default settings of dhcp server in Mikrotik. Lease time was set to 10 minutes. When I changed it to 24h my iphone stays connected to the network

btw. when I was using Mikrotik as AP, I also observed waking up mobile devices every 20s, it was killing battery. For this reason I'm not using Mikrotik for wireless with mobile devices

Regards.

Fleury · Sat Jan 28, 2017 11:39 am

Did mikrotik failed to fix this issue?? I have upgraded to the last firmware but still have the same problem "disconnected, received diassoc: sending station leaving (8)"

dbornack · Mon Feb 27, 2017 4:56 am

I'm also having this issue with a new install. I've turned off power saving on my phone, but even when I'm using the phone the WiFi disconnects. I don't have the issue with my laptop.

Rossi1210 · Thu Mar 02, 2017 10:45 am

Hello, I have the same problem
If I connected me with the WLAN
Android says: connection
Android says: IP is related
And then disconnect

In the logfile
Xx: xx: xx: xx: xx: xx @ xxxCRS connected
And 3 seconds later
Xx: xx: xx: xx: xx: xx: xx @ xxxCRS disconnected, received disassoc: sending station leaving (8)

I find myself in another room, with reception to another CAPs,
So I get there a stable connection.

HAP lite A and hAP lite B works fine, only the CRS109 makes problems.

It is set up all over the same channel and rolled over capsman.

Thanks for your help

Use version 6.83.3 everywhere

Rossi1210 · Sun Mar 05, 2017 8:53 am

problem solved.
The android mobilphone used McAfee
And the APP blocked the mac-adress of the WLAN, because i placed it for unsafe.

shelbynetworks · Wed Apr 05, 2017 11:16 pm

force the channel; ie dont set the channel on the ap to auto

hoftel · Tue May 16, 2017 10:59 am

force the channel; ie dont set the channel on the ap to auto

This appears to have fixed my problem. In our office there is a ton of other WiFi signals on every sub channel. Maybe that is what is causing the connection to be very unstable.

I forced it onto channel 4 which is the best one here. If the problem appears again I will tell you guys here. Thank you!

p.s. I came to this thread when googling

hoftel · Tue May 30, 2017 3:06 pm

I forced it onto channel 4 which is the best one here. If the problem appears again I will tell you guys here. Thank you!

The problem actually wasn't solved for long and reappeared next day. I tried some more settings, including disabling security and lowering the tx power, but the problem persists. I think my smartphone itself is suspect. I truly hope this problem won't appear at a customer...

frittentheke · Mon Jun 12, 2017 12:15 am

I have a RB2011 with CAPSMAN and a hAP-AC connected to it, latest and greates firmware.
My various devices, but apparently only those running Android (phones and tablets) have disconnects all the time with

"disconnected, received disassoc: sending station leaving (8)"

in the logs. The device immediately reconnect, but with a short disruption.
There is somethings really really fishy as my Linux notebook does never disconnect even when walking through the building.
I played with various settings such as:

* hw-retries (tried 4 and 7)
* group-key-update (raised to 23:59:59)
* Group Encryption (set to AES)

Anyone else seeing any issues of this sort?

michlG · Wed Aug 09, 2017 11:01 am

Hi,

i have exactly the same problem as described above.
A few device types work without problems but some device types loose the connection frequently (and sometimes fail to reconnect).

The devices which have problems are:
- Apple devices
- Windows Phone 10
- Sonos Play 1 and Play 5
- HP PageWide Printer

It doesn't matter if the reception is good or bad. After some time the connection simply drops.

I am using CapsMan with 5xRB951G-2HnD and 1xRB911G-5HPacD.
I already tried a lot of different configurations, updated to newest version 6.40.1, etc... But the problem is always the same.

I really like the mikrotik products (router + SXT Lite and also capsman) but the wifi is really annoying.
Is there a way to fix the problem?
Note that the mentioned devices work without problems if I use access points of your competitors (even if we use the cheapest home-user trash).

I created this project as an evaluation scenario for our customers. We planned to switch to Mikrotik for business use.
But if the wifi is not reliable it makes no sense.

So please Mikrotik-Team help us to fix the Wifi-problems and we will be your future-customer

Kind Regards
Michael

frittentheke · Wed Aug 09, 2017 11:30 pm

I also still have the exact same issue. Only Android devices of all sorts drop the connection. Our notebooks running Linux work just fine, never dropping the Wifi connection.
Nothing is shown in the log apart from that disassoc message.

cameronhall · Sun Aug 13, 2017 5:10 am

I have the same problem on both my wAP ac's .
Has anyone had any luck fixing it? Only happens on Android phones.

radpi · Tue Aug 29, 2017 9:43 pm

problem solved.
The android mobilphone used McAfee
And the APP blocked the mac-adress of the WLAN, because i placed it for unsafe.

Same problem, same solving. McAfee blocked my Wifi.

Megakoresh · Tue Sep 26, 2017 9:14 pm

Same problem.
I have CapsMAN running on hAP provisioning to WAP on second floor. Same SSID, AES, dual band.
Occurs on every client that connects to the 2.4Ghz antenna on either, and because of interference patterns the clients often jump back and forth between bands, resulting in insane speed drops and inconsistent connection. Same log messages constantly appear as what others have posted:

//...
20:54:32 caps,info 48:51:B7:C4:53:42@cap31 connected 
20:54:32 caps,info 48:51:B7:C4:53:42@cap32 disconnected, registered to other interface 
20:55:39 caps,info 48:51:B7:C4:53:42@cap31 disconnected, received deauth: unspecified (1) 
20:55:47 caps,info 48:51:B7:C4:53:42@cap31 connected 
20:55:49 caps,info 48:51:B7:C4:53:42@cap31 reassociating 
20:55:49 caps,info 48:51:B7:C4:53:42@cap31 connected 
20:55:49 caps,info 48:51:B7:C4:53:42@cap31 disconnected, received deauth: unspecified (1) 
20:55:50 caps,info 48:51:B7:C4:53:42@cap31 connected 
20:57:05 caps,info 44:74:6C:B2:B3:A6@cap32 connected 
20:57:06 dhcp,info dhcp1 assigned 192.168.88.162 to 44:74:6C:B2:B3:A6 
20:58:25 caps,info 48:51:B7:C4:53:42@cap31 disconnected, group key timeout 
20:58:29 caps,info 48:51:B7:C4:53:42@cap32 connected 
21:01:00 caps,info 80:57:19:E8:0B:38@cap31 disconnected, received disassoc: sending station leaving (8) 
21:01:05 caps,info 80:57:19:E8:0B:38@cap31 connected 
21:01:35 caps,info 48:51:B7:C4:53:42@cap31 connected 
21:01:35 caps,info 48:51:B7:C4:53:42@cap32 disconnected, registered to other interface 
21:02:35 caps,info 48:51:B7:C4:53:42@cap31 disconnected, received deauth: unspecified (1)
//...

Whenever one of these capXX disconnected messages appear, that wifi client gets dropped and after reconnecting has a speed of about 10kbps (i.e. unusable).

Looking at the age of this thread its quite ridiculous that the problem is still not resolved. Even if it's solvable by some advanced settings manupulations, it is clearly not an edge case and far too common to be just ignored.

alsangeorgia · Mon Nov 13, 2017 12:56 pm

solution:
dont use special characters in password

borregator · Fri Feb 02, 2018 7:02 pm

Hello, I got a new kind of message: wlan2:data from unknown device "DEVICE MAC ADDRESS CODE", sent deauth. It was received after trying to connect an Ipod device, IOS 6.1.6.
I tryed changin the wireless channel but it did not work.

rheo · Sun Apr 01, 2018 10:58 am

this was bugging me. i even returned a couple of android devices. I think i have found the cause, in my case i had to disable ipv6 on the mikrotik entirely... i don't know why but now my android devices stay on wifi

dollarstar · Wed Nov 07, 2018 12:58 pm

Good day.
The problem is still relevant since 2014 and at the moment (now November 2018) on the latest firmware everything is exactly the same.
Changing the MAC address of the name and hiding the point did not help (there were suspicions of an attempt to hack the wifi)
There are no "special characters" in the password. My password is 12345678.

Perhaps the developers should pay attention to this problem?

gotsprings · Thu Nov 08, 2018 1:27 pm

The support for wireless is non existent.

I have submitted multiple tickets and rarely get any response from Mikrotik.

I include logs and support files and still nothing.

Mikrotik can route... But I don't know what is up with the wireless.

dollarstar · Thu Nov 15, 2018 3:09 pm

The problem is observed on the equipment: 951Ui-2nD (I have 2 such problematic devices)
With the 951G-2HnD equipment, communication is stable.

Both routers are configured the same.

Temorizador · Tue Nov 20, 2018 12:35 am

Here the same situation, only the mac devices in my case are those that have the problem.

profile: guest without password, 2 devices Cap2nD, canel 1 and 6 only n 20mhz , tx by default , acceslist to 90.

Adroid devices, portable, wifi usb work fine but mac pc, mc iphone or ipad have errors.

as it is a free network if a device I see that it is connecting and disconnecting I usually filter it by mac-adress so that it is not like that all day, but I would like to find a solution.

Tested different settings conbinaciones speak for the forum and tests yourself but the problem persists.

In the private wireless ,no mac devices are used, so I do not have problems with my devices, but if i ever get a mac device i could use wireless pleasantly.

sorry the traslate

(

regards
Temorizador

TechMikhail · Tue Jan 29, 2019 3:07 pm

Hi, there was the same errors, at my network - the result was incorrect working DHCP server. Pushing 2 networks to one interface. Disable/Enable dhcp server helps.

raks · Tue Feb 05, 2019 12:14 pm

I have been experiencing this issue on 2 devices, a Macbook Pro and Nintendo switch.
I changed my DHCP Lease Time to 23:00:00 today, and my pings have been clean so far.
Strange thing is that the IP on my Macbook has been reserved, so I don't understand why this occurs.

zandhaas · Wed Feb 27, 2019 11:33 pm

I'm Experiencing this problem on my Samsung TV while watching video via the internet via the smart TV apps (Netflix, Videoland etc)
The TV +stops paying and shows the message "no internet connection". When this happens my laptop which is on the same wifi interface loses the connection also.

I'm using the HAPac2 as an access-point.

a575606 · Fri Mar 22, 2019 1:38 am

Same problem here. 2019 now. MIKROTIK, ARE YOU LISTENING?

Zadorozhko · Sun May 12, 2019 1:20 pm

hAP ac lite. 6.45beta34 (latest)
2 Android tablets (U65GT, U68GT) have this problem. Macbook, iPad, android notebook - don't.
Log entries every 5-10 min:
disconnected, received deauth: sending station leaving (3)
disconnected, received disassoc: sending station leaving (8)
Signal strength is good. Noise level low. Battery saving is off.
Tried everything I googled in 2 years. Nothing helps.
Mikrotik, any ideas? Seems this problem is well known for years and quite common, Google shows around 30k hits on this query. But no any solution.

frogale · Thu Jun 13, 2019 11:56 pm

Same problem.. hAP ac'2.. Latest stable 6.44.3.. Both 2G and 5G interfaces..

llubik · Thu Jul 04, 2019 1:15 pm

the same problem, iphone 6s, nobody's interested in it, i'm going to throw a microtik hap ac2 into the trash

mkx · Thu Jul 04, 2019 1:52 pm

i'm going to throw a microtik hap ac2 into the trash

Throw it into my trash bin ... I don't have any apple gadget nearby ...

KiLLeRRaT · Thu Jul 04, 2019 10:23 pm

i'm going to throw a microtik hap ac2 into the trash
Throw it into my trash bin ... I don't have any apple gadget nearby ...

Nothing to do with Apple. It does it with my iot gear too. I have about 10 esp8266 devices that can't connect.

I really hope they would take a look at it.

gotsprings · Fri Jul 05, 2019 3:11 am

I got a good capture of Caps-man not allowing connections. Wrote up a what how and when for Mikrotik.

They actually investigated it.

All my prior interactions with Mikrotik about wireless have been pointless. One suggestion email per month with no resolution.

This time...
"We were able to reproduce the problem. (Blah blah blah). We don't currently have a Fix."

I have stopped firmware updates on any deployed Tik wireless and will not deploy an new ones until this is fixed.

Kind of sucks as I have developed a lot of stuff for our installs that take advantage of Tik radios. But if clients can't connect... It's pointless.

Back to Ruckus Only for wireless.
(Thousands of dollars in Tik wireless is going to sit on the shelf because distro won't take it back.)

zandhaas · Fri Jul 05, 2019 1:06 pm

This time...
"We were able to reproduce the problem. (Blah blah blah). We don't currently have a Fix."

When did you get this anwser?

gotsprings · Sun Jul 07, 2019 4:49 am

This time...
"We were able to reproduce the problem. (Blah blah blah). We don't currently have a Fix."

When did you get this anwser?

June 10th.

zandhaas · Sun Jul 07, 2019 11:56 am

OK.

At least the message that they were able to reproduce the issue gives me a little hope they are able to fix this in a (near) future release.

freez07 · Fri Jul 12, 2019 12:04 pm

Hi, same problem here.
I'm using 2 routers RB751G-2Hnd with CapsMan control.

All my 10 devices are connecting good and stable.

But my laptop drop the connection every 10-15 minutes. It is powerfull ASUS N705UD-EH76 laptop with 2 and 5 Mhz wi-fi support.
Here what the log shows:

DHCP leases setted up to 3 days.
TKIP is off.

Don't know what else to do

garthk · Sat Jul 13, 2019 4:21 am

I've had awful 2GHz wireless performance for a few days (5GHz was fine). Upgraded my hAP ac from 6.44.3 to 6.45.1 (release notes) as preparation for posting a question here about it… and either the upgrade or the reboot had a dramatic change. I'm back to a solid 60Mbps down, 30Mbps up, through a wall and a floor at a nasty angle. Release notes say…

*) wireless - fixed frequency duplication in the frequency selection menu;
*) wireless - fixed incorrect IP header for RADIUS accounting packet;
*) wireless - improved 160MHz channel width stability on rb4011;
*) wireless - improved DFS radar detection when using non-ETSI regulated country;
*) wireless - improved installation mode selection for wireless outdoor equipment;
*) wireless - set default SSID and supplicant-identity the same as router's identity;
*) wireless - updated "china" regulatory domain information;
*) wireless - updated "new zealand" regulatory domain information;

… though 6.44.3 was fine a week ago, so maybe it was the reboot.

Zensis · Wed Jul 17, 2019 11:00 am

The problem was not solved, I bought and I have Mikrtotik Metal G-52SHPacn year 2019 and also a 2011UAS-2HnD with RouterOs 6.45.1 and they give exactly the same failure mentioned in the first message of this thread in 2014 ... disconnected, received diassoc : sending station leaving (8), comment, I have several home devices connected to metal 52AC, cell phones, tablets, etc. and none more than 6 meters away and most with direct line of sight, buy the equipment two months ago and since then connections are horrible in almost all devices that connect, (6 in total) the only device that goes well and does not fail is an old Chinese Tablet that navigates and walks as if nothing happened, the other computers all the time disconnected, received diassoc: sending station leaving (8), every 2 to 5 minutes, unbearable, I tried everything, everything, change of frequencies (channels), change of security .. change of data rates, etc. etc. I touched everything And for my last test I did Full reset to Metal 52ac, OS "Current Firmware 6.45.1" security default (without security), I only enabled wan by dhcp client and the nat in mascarade for internet access, nothing else, all default and connected by wifi only one device (sony xperia x performance with Android

to Metal 52ac and this device with little navigation load, web browsing news, 480p videos is disconnected and connects every 2 minutes approximately, being unbearable, just like before, but if the data load is maximum, for example video playback 4k .. the connection is stable does not fail, as long as it is playing and occupying all the bandwidth, if it pauses or leaves the video the connection returns to fail as before ... conclusion is that Mikrotik does not support modern energy saving system? or modern data saving? By disconnecting the devices that enter in that mode or change mode, is there any way to configure so that Mikrotik does not disconnect the devices? Try to keep them? pay for mikrotik 5 times more than the normal router Tplink, Zyxel, Huawei to have these problems ... ..
Could someone solve this problem?

and excuse my bad english

slav0nic · Sat Aug 03, 2019 2:59 pm

HAC AC, v6.45.3 , same problem

gotsprings · Tue Aug 20, 2019 1:35 pm

Use a different manufacture for wireless and just forget about Mikrotik wireless.

This "glitch" has been observed and reported for years at this point.

Get over it and move on.

llubik · Mon Sep 09, 2019 9:53 am

I also have these problems, and I have it with a maximum of xxxxxxx, I see in the log nothing else just dissasoc. . . I guess there is anything you can set up when it doesn't handle the basic functions and works worse than a simple tplink, for example, for a third of the price. . .

kolorafa · Tue Sep 10, 2019 12:03 am

I had the same message (disconnected, received disassoc: sending station leaving (8)) every 17s.

It turned out to be that this is a message when device requests to get disconnected, in my case it was "missing bridge port", as I did setup CAPsMAN with multiple ssids, and the secondary ssid was not bridged to the network, so the device waited 17s for ip and then after not receiving was disconnecting.

So yea, Mikrotik might not be the best and easiest router for wireless, but at least it work flawlessly with my stuff (if setup correctly

Hope my "mistake" will help someone with similar problem.

GregC · Mon Sep 30, 2019 6:08 pm

I have the same similar issue “disconnected, received disassoc: sending station leaving”. I have not found a fix for this problem but I think I have a workaround that has solved this issue. Wrote a script and add it to the scheduler as shown below:

/system scheduler
add interval=1w name=Recycle-Caps-Interfaces on-event=":log info \"Disable caps-man\";\r\
\n/caps-man manager set enabled=no\r\
\n:delay 10s\r\
\n:log info \"Enable caps-man\";\r\
\n/caps-man manager set enabled=yes\r\
\n" policy=ftp,reboot,read,write,policy,test,password,sniff,sensitive,romon start-date=sep/18/2019 start-time=02:00:00

This runs once a week at 2:00 AM in the morning and so far it has worked for me. Yes, I am aware that mobile clients disconnect when they go onto sleep mode to save battery life.

gotsprings · Mon Sep 30, 2019 11:41 pm

I have the same similar issue “disconnected, received disassoc: sending station leaving”. I have not found a fix for this problem but I think I have a workaround that has solved this issue. Wrote a script and add it to the scheduler as shown below:

/system scheduler
add interval=1w name=Recycle-Caps-Interfaces on-event=":log info \"Disable caps-man\";\r\
\n/caps-man manager set enabled=no\r\
\n:delay 10s\r\
\n:log info \"Enable caps-man\";\r\
\n/caps-man manager set enabled=yes\r\
\n" policy=ftp,reboot,read,write,policy,test,password,sniff,sensitive,romon start-date=sep/18/2019 start-time=02:00:00

This runs once a week at 2:00 AM in the morning and so far it has worked for me. Yes, I am aware that mobile clients disconnect when they go onto sleep mode to save battery life.

So you only see this problem after a week?

GregC · Tue Oct 01, 2019 12:14 am

I have the same similar issue “disconnected, received disassoc: sending station leaving”. I have not found a fix for this problem but I think I have a workaround that has solved this issue. Wrote a script and add it to the scheduler as shown below:

/system scheduler
add interval=1w name=Recycle-Caps-Interfaces on-event=":log info \"Disable caps-man\";\r\
\n/caps-man manager set enabled=no\r\
\n:delay 10s\r\
\n:log info \"Enable caps-man\";\r\
\n/caps-man manager set enabled=yes\r\
\n" policy=ftp,reboot,read,write,policy,test,password,sniff,sensitive,romon start-date=sep/18/2019 start-time=02:00:00

This runs once a week at 2:00 AM in the morning and so far it has worked for me. Yes, I am aware that mobile clients disconnect when they go onto sleep mode to save battery life.
So you only see this problem after a week?

Because it runs on a weekly basis is hard to say. All I know is that I don’t see it anymore due to the weekly runs. Maybe if I run it once a month I could find out, but I haven’t and see no need to do so.

gotsprings · Tue Oct 01, 2019 1:57 pm

I see this problem the same day I configure the router.

Now on some jobs the problem pops up a few days after I have left. Those ones... disabling the radio interface sometimes cures it. But the jobs where I see those messages right away... I have to change manufactures. Too many lost hours and clients.

dad2312 · Sun Nov 17, 2019 1:01 pm

I see this problem the same day I configure the router.

Now on some jobs the problem pops up a few days after I have left. Those ones... disabling the radio interface sometimes cures it. But the jobs where I see those messages right away... I have to change manufactures. Too many lost hours and clients.

Don't know if it can help but i have same trouble and i just change name of SSID on virtual wifi (i have trouble "station leaving" only on virtual wifi) with a short name without special chararcter and for moment it's ok.
before : Restaurant-Jeannala
After : Jeannala

ogekuri · Sat Nov 30, 2019 3:43 pm

I see this problem the same day I configure the router.

Now on some jobs the problem pops up a few days after I have left. Those ones... disabling the radio interface sometimes cures it. But the jobs where I see those messages right away... I have to change manufactures. Too many lost hours and clients.

I'm on CRS125-24G-1S-2HnD with some cAP L-2nD configured via CAPsMAN.

All sorftware versions are v6.45.7.

Today I've try do make some changes to resolve the same issue "disconnected, received disassoc: sending station leaving (8)" according all suggestions founded on this thread and it's seem resolve my issue.

Here my changes done:
1) Changed caps-man configuration
- distance=indoors
- hw-retries=15
- hw-protection-mode=rts-cts
- frame-lifetime=0
- disconnect-timeout=15s
2) Changed caps-man security
- encryption=aes-ccm ( removed tkip, leaved only aes-ccm)
- group-encryption=aes-ccm
- group-key-update=23h59m59s
3) Changed ip dhcp-server
- lease-time=23h59m59s

--
Ciao
Francesco.

srtrs · Wed Dec 04, 2019 10:41 pm

Unbelievable...
Here is my theme 4 years ago.
viewtopic.php?f=7&t=97487

And there is still no solution. Now I have 952Ui-5ac2nD and all the same. Where is the official vendors comment ?!

llubik · Fri Dec 13, 2019 3:01 pm

Iphone connect . . . reassociating . . . disconnect . . . connect . . . ??????????????????????? Mikrotik feature ?

Dec/10/2019 20 43 25 disk wireless. inlo 3B SE 2A connecled. signál slrenglh -6S
Dec/10/2019 20 49 39 disk wireless. inlo 33 SE 2A reassocialing
Dec/10/2019 20 49 35 disk wireless. inlo 3B 3E 2A disconnecled. ok
Dec/10/2019 20 49 39 disk wireless. inlo 3B 8E 2A connecled. signál slrenglh -6S
Dec/10/2019 20 52 53 disk wireless. inlo 3B 3E 2A reassocialing
Dec/10/2019 20 52 53 disk wireless. inlo 3B SE 2A disconnecled. ok
Dec/10/2019 20 52 53 disk wireless. inlo 3B 3E 2A connecled. signál slrenglh -71
Dec/10/2019 20 55 14 disk wireless. inlo 3B 3E 2A reassocialing
Dec/10/2019 20 55 14 disk wireless. inlo 3B SE 2A disconnecled. ok
Dec/10/2019 20 55 14 disk wireless. inlo 33 SE 2A connecled. signál slrenglh -70
Dec/10/2019 20 57 2S disk wireless. inlo 3B 3E 2A reassocialing
Dec/10/2019 20 57 2S disk wireless. inlo 3B 8E 2A disconnecled. ok
Dec/10/2019 20 57 2S disk wireless. inlo 3B 3E 2A connecled. signál slrenglh -70
Dec/10/2019 20 53 27 disk wireless. inlo 3B SE 2A disconnecled. etfensive dala loss
Dec/10/2019 20 59 44 disk wireless. inlo 3B 3E 2A connecled. signál slrenglh -70
Dec/10/2019 21 1 5S disk wireless. inlo 3B 3E 2A reassocialing
Dec/10/2019 21 1 53 disk wireless. inlo 3B SE 2A disconnecled. ok
Dec/10/2019 21 1 5S disk wireless. inlo 33 SE 2A connecled. signál slrenglh -70
Dec/10/2019 21 2 45 disk wireless. inlo 3B SE 2A reassocialing
Dec/10/2019 21 2 45 disk wireless. inlo 3B SE 2A disconnecled. ok
Dec/10/2019 21 2 45 disk wireless. inlo 3B SE 2A connecled. signál slrenglh -70
Dec/10/2019 21 3 27 disk wireless. inlo 3B SE 2A reassocialing
Dec/10/2019 21 3 27 disk wireless. inlo 3B 3E 2A disconnecled. ok
Dec/10/2019 21 3 27 disk wireless. inlo 3B SE 2A connecled. signál slrenglh -70
Dec/10/2019 21 4 11 disk wireless. inlo 3B SE 2A reassocialing
Dec/10/2019 21 4 11 disk wireless. inlo 33 SE 2A disconnecled. ok
Dec/10/2019 21 4 11 disk wireless. inlo 3B SE 2A connecled. signál slrenglh -74
Dec/10/2019 21 4 47 disk wireless. inlo 3B SE 2A disconnecled. etfensive dala loss
Dec/10/2019 21 5 0 disk wireless. inlo 3B SE 2A connecled. signál slrenglh -74
Dec/10/2019 21 5 5S disk wireless. inlo reassocialmg

Iphone 6s (2x), iphone 5s problem
samsung S7, A50, Honor 10, ntb Dell, acer, thinkpad, tv samsung, HTC, redmi4A, Huawei P10, - no problem

Test Asus WL500gp 2.4GHz . . . iphone OK

????????????????????????????????????

SidoroOFF · Thu Feb 06, 2020 12:08 am

Router - RB951
Software ID S2VM-GWK3
Serial Number 6F1206B8461F
Level 4

Client - notebook HP Pavilion g6 WiFi
Problem periodic^

disconnected, received disassoc: sending station leaving (8)

llubik · Wed Feb 19, 2020 3:28 pm

I prefer a plastic toy (Asus, etc ....) as a "professional" MKT device that will not hold the connection.

rooted · Wed Feb 19, 2020 3:57 pm

I prefer a plastic toy (Asus, etc ....) as a "professional" MKT device that will not hold the connection.

You could disable the wireless on the Tik and use an external AP, then you have all the great features of RouterOS with wireless that doesn't give you issues.

llubik · Sat Feb 22, 2020 6:54 pm

This is not the solution. Look at how many years the problem is not solved. How is it 99% mikrotik? Chinese toys have no problem. What should be done in order to get started in mikrotik? Should I send a photo as please on my knees? If that helps, I'm glad to do it, it doesn't make me a problem.

llubik · Sat Feb 22, 2020 9:13 pm

In most of the answers to a similar problem, it is recommended to set up Beacon, but especially DTIM. In Mikrotik required since 2010. . . if not early. . . and still nothing.

Retral · Sat Feb 22, 2020 9:18 pm

Personally I believe it comes down to Mikrotik's lack of WiFi alliance certification (Note: virtually every other manufacturer I've looked up on wi-fi.org has at least one device certified. Even Google's Pixel 2,3 and 4 for crying out loud have been certified). I firmly believe that if they went through that with at least one product they would be left with no other choice than making their product work properly with the protocols and in turn that would help them do 802.11 correctly on their other products. That's just my 2 cents though.

llubik · Sun Feb 23, 2020 5:30 pm

Is this normal?

llubik · Mon Feb 24, 2020 9:42 pm

I have here iPhone 5s, ios 12.4.3.
There is no problem, at least to look at. connects to wifi, gets static IP, ping its IP. I lock the phone, make a few pings, then time out. After about 2.5 min there will be 15 pings, then again the timeout until 10 min - I did not try again. During this time, no logs were entered in the log except IP allocation. Let's feel nice in the wireles tables.
Now I take the iPhone 6s, I suppose the last (even 2 phones), these phones have no physical reach, but I have an idea when what is happening, so the wreckage with reserve.
When it is turned on - without running on stream tv, there is no problem. Once locked, the lof will appear every half minute 3 minutes - reassiociating, - disconected, ok, - connected.
When I ping on such a phone, as soon as this sequence runs, so pingam, 10x, then timeout until the next reassociating. . . and so on. At this point, Uptime is reset in the wireless tables. Ip is assigned statically, both 5s and 6s connect to the same wifi. in 5s there is not much to set up for wifi. I could not save battery.
I wonder who wakes up the 6s every half minute. Is it iOS that checks, say, some notifications, etc., or it is a microtik that sends a packet, frame or anything that wakes up the phone. Older iOS, say, has not responded to this until later. But let's say, every half a minute, according to me, moooooc often - batteries, write uptime. . .
To compare Android. 5 minutes after locking pingam only timeout, within 10 min. then I canceled it. When I unlock it, no log in the log, uptime normane running, as on the iPhone 5s.
Does this look like setting up s6 / new iOS / bug?
Perhaps the whole world

internet, no solution. . . I found some settings what I recommend for iphone, I set them (enable in mm, long preamble, do not use tkip, etc etc), but I did not find DTIM settings - microtik offer - here recommends to change from 1 to 3 for Apple. with iphone on microtics?
Excuse my English, perhaps it's understandable.

maretodoric · Fri Mar 27, 2020 1:17 pm

Year 2020 and still the same issue.

I have two clients constantly disconnecting. I forced the specific one from screenshot bellow to connect to another non mikrotik AP and no disconnects ever since.
I'm using hAP ac2 with firmware version 6.46.3 and this is going on for quite some time. I don't want to rely on another AP as i only need 2 one. I've placed third one basically next to hAP ac2 to troubleshoot this issue and seems to work. Don't want to radiate myself because of this issue! Why is it constantly getting disconnected from MikroTik ?!
Please for the love of God, HELP!

tik dropouts.png

[admin@MikroTik] > /interface wireless print 
Flags: X - disabled, R - running 
 0  R name="wlan1" mtu=1500 l2mtu=1600 mac-address=74:4D:28:6D:4E:1D arp=enabled 
      interface-type=IPQ4019 mode=ap-bridge ssid="Don't look at me" 
      frequency=auto band=2ghz-b/g/n channel-width=20/40mhz-XX 
      secondary-channel="" scan-list=default wireless-protocol=802.11 
      vlan-mode=no-tag vlan-id=1 wds-mode=disabled wds-default-bridge=none 
      wds-ignore-ssid=no bridge-mode=enabled default-authentication=yes 
      default-forwarding=yes default-ap-tx-limit=0 default-client-tx-limit=0 
      hide-ssid=no security-profile=default compression=no

llubik · Fri Mar 27, 2020 10:24 pm

Anyone who has such a problem, I suggest to send a microtik log every week.

maretodoric · Sat Mar 28, 2020 8:02 pm

My God, even their support is out of whack........ Just signed up to support portal at https://help.mikrotik.com/servicedesk and an email I've received for confirmation leads to domain jira.mt.lv which resolves to nothing... Changed that to help.mikrotik.com and kept the other part of the link with confirmation token and worked..... How unprofessional this is if they cannot even send proper info via email..? What happened to MikroTik?

kilokilo · Thu May 14, 2020 8:20 pm

i had this problem too.
"received deauth: sending station leaving"

netinstall a few times.... long term release, stable release....

it works for a few days then the wifi start acting funny...

adding mac address to the ACL and connection list in wifi doesn't help
changing attenna gain won't help
changing channels between auto or manual doesn't help (i only have 3 ssid surround and my place is rather a clean interferance area)
disable nstreme / nv2 won't help
disable some low bit rates in 802.11 protocol doesn't help
reducing Rx / Tx chains doesn't help.

can we have a fool proof wifi connection, i can sacrifice all the custom / advance features... i can't even implement capmans since i had difficulties on basic wifi connection with a qualcomm chipset mobile...
i had 2 boardcom apple device but i don't have problems with it.
but i had 7 qualcom android device only a few has problems...

can we have a stable wifi connection? i don't think this isn't a huge request, aren't we?
my hap ac2 and hap lite is even worst then a 10yrs old TP-link 841nd...
is much horrible then a linksys wrt54g....
my old netgear r6250 has better 5G than my hap ac2....

am i asking too much?

bpwl · Fri May 15, 2020 1:12 am

I had this too... looked very weird .... could not believe it because it worked before without a glitch..... viewtopic.php?f=7&t=160376

It was MY FAULT. There was something wrong in the "bridge port" setting. The MAC address of the client was not learned by the bridge during its association. Any other action that made the bridge learn the MAC address (like doing a manual DHCP renewal immediatly after connecting) made it work. As long as there was traffic it remained somewhat stable. But ultimately it dropped with the known message. (probably bridge cache expiration)

Solution was to fix the bridge port setting with "learn=auto". Other settings immediatly reintroduced the problem. It took a long time investigating, and I considered removing my post, but who knows, it could help someone to look at a non-intuitive place.
Check your bridge port setting, and check your DHCP leases.

ksuuk · Fri May 15, 2020 8:54 pm

I had this too... looked very weird .... could not believe it because it worked before without a glitch..... viewtopic.php?f=7&t=160376

So as You had problems, then You actually lied in Your other post - viewtopic.php?f=7&t=160943#p792785

"Why do I have to proof myself. The content of my contributions should be just of value enough.
And yes all my Mikrotiks are up to spec right now. I even get borred in checking. (Not a single glitch in the last 6 months, monitored continously with DUDE)"

Shame You...

llubik · Wed Jun 03, 2020 11:57 pm

Solution was to fix the bridge port setting with "learn=auto".
Check your bridge port setting, and check your DHCP leases.

learn=auto, dhcp leases 1d 00:00:00 . . .

iPhone dissacoc . . .

maretodoric · Fri Jun 05, 2020 1:06 pm

I actually.. Well.. I don't know what got this resolved.
But basically, learn on bridge port was set to "auto" for all interfaces.
DHCP Lease was 23 hours but the issue stopped when i installed RB2011iLS instead of my main router.
So now:

RB2011ilS ether2 -> hAP ac2 ether1
RB2011iLS is default gateway and DHCP server. Ever since that setup i have no more issues, so i have on idea what was going on but I'm happy now. And to be honest, I'm afraid to revert back to test it since it was really stressing me out!

salikhoff · Thu Jun 18, 2020 1:33 am

Only flashing OpenWrt worked for my hap lite. No more trust to mikrotik

Wombat1986 · Sat Jul 25, 2020 5:08 pm

Hi,
i had the same issue on cAP ac only with the 5GHz interface (seperate SSIDs).
I solved the problem by performing a factory reset and reconfiguration with standard settings and a fixed channel at first (bridge AP mode without own DHCP server - my Router TP-Link TL-WDR4300 v1 is doing that). Everthing was fine and works well.
Afterwards i set the country code to germany, set the encryption to WPA2 only (AES) and everything is still working. I'm so happy and would like to share this experience here

.

Also the speed of the 5GHz Wifi was increased up to ~500MBit from ~300MBit. Before the factory reset it was not possible to reach higher speed with 20/40/80MHz channel width. Now with the std. config it works.

Regards,
Felix (Wombat1986)

mfliorko · Tue Aug 11, 2020 5:29 pm

I was experiencing similar issues with connecting some devices like Samsung Galaxy A7 (2018), old Dell laptop etc., whereas Macbook Pro (mid 2015), iPhone 6S Plus, iPad Pro (2020) were connecting to my RBD52G-5HacD2HnD (hAP ac2) seamlessly. I have four such devices configured through CAPsMAN.

Rather accidentally I managed to fix the issue. Hopefully, my settings would be helpful for others.

CAPsMAN -> Rates -> <<select one>> -> (select only one MCS, "0"):

screenshot 1 - rates.png

Other settings:
CAPsMAN -> Security Cfg. -> <<select one>> -> (set "Group Key Update"):

screenshot 2 - security cfg.png

CAPsMAN -> Channels -> <<select one>> -> (lower "Tx Power"):

screenshot 3 - channels.png

IP -> DHCP Server -> <<select one>> -> (increase "Lease Time):

screenshot 4 - dhcp server.png

P. S.
I'm not an expert in network technologies. Probably, these settings are not optimal and someone can suggest improvements. Anyhow it works for me.

bpwl · Tue Aug 18, 2020 11:05 pm

Surprising combination of settings. I'm not sure if I should be reacting on this.

all interface rates for 802.11b and g set
only MCS 0 for 802.11n (= 6,5Mbps only !). That is very low.
this is for 2.4 GHz so no 802.11ac that could be helping with VHT MCS settings (and solve the MCS 0 only limit)
basic rates up to 54Mbps? But which one will be used? 1 Mbps??

improvement: remove 802.11b (and speeds 1, 2 5.5 and 11 Mbps)
improvement: set all MCS rates supported. The AP and client will adjust dynamically based on # of re-transmits
improvement: set basic rates at 6Mbps to start with. If all clients have a very good reception you could increase till 9 Mbps, even 12 Mbps
If it is 54 Mbps for broadcast, you must be in an EMC/EMI testroom, as no single packet loss is allowed here, and there is no ACK expected, nor retransmits possible.
improvement all supported rates : OK but with 802.11b gone, the 4 lowest will not be there

Security: what is it now WPA2-PSK (private shared key) or WPA2-EAP (= WPA2 Enterprise, based on 802.1X and RADIUS server)? Can't be both, can it?

srtrs · Wed Aug 19, 2020 12:53 am

There is no solution to this problem. Mikrotik devices are not certified by the Wi-Fi alliance

bpwl · Wed Aug 19, 2020 1:48 am

Solution was to fix the bridge port setting with "learn=auto".
Check your bridge port setting, and check your DHCP leases.

learn=auto, dhcp leases 1d 00:00:00 . . .

iPhone dissacoc . . . :-(

I was not clear on "check your DHCP leases". It's not about the duration of the lease. It's about the fact that the DHCP lease gets to the state "offered" , a point where it is usable for the receiver, but never makes it to the state "bound". After a short period of time a DHCP lease that stays in "offered" gets dropped. The client looses, or is revoked, it's leased IP address, and as a reaction disconnects.
The reason for the DHCP lease staying in "offered" seems very related to "wireless interface timing", "bridges and their STP setting", and "DHCP server being on a bridge".
Like with the MAC address not being learned in the bridge hosts table (and not having data flowing as a consequence) does it for sure , getting or better losing an IP address seems enough for some stations to leave the AP.
There might be other checks. Those "internet" checks done by station devices (testing for reachability of "some" public server), also make the station device start trying other stored SSID's in its list, if the "internet check" test fails. Even if the current wifi connection is excellent (but gives no access to internet).

darek1981 · Wed Sep 16, 2020 10:24 am

Hello, I also have a problem with disconnecting, one device - Iphone, software 6.47.1
I read this topic but it didn't help in my case.
Mikrotik technical support?

sep/15 20:49:13 wireless,info 28:ED:6A:AA:...@wlan1: disconnected, received disassoc: sending station leaving (8)

jbl42 · Sat Sep 19, 2020 3:30 pm

I think the important point here is the reason code 8 means the STA has left (disassociated) the AP on its own decision (https://blogs.arubanetworks.com/industr ... tus-codes/). It's not the AP dropping the STA, it's the STA leaving the AP. So the question is what causes the STA to decide to leave the AP.

In a single AP setup, if this happens with good signal levels (-60dbi or better), the AP "does" something making the STA thinking something is wrong causing the STA to leave the AP. This might be the STAs or the APs fault (most likely a bug). It is almost impossible to debug this for anyone else than the developers of the AP and STA.

For multi-AP setups, such spontanious disconnects often are caused by poor roaming decisions of STAs. As widely known, in WiFi roaming decisions are taken by the clients. In difference to Mobile Phone networks, there is no proper way in WiFi to force STAs to (re)connect to a specific AP. To improve this situation, Apple and Cisco developed together what later was standardized as 802.11k/v. This provides STAs with additional informations on neighbor APs and their channels, among other things. The roaming decision is still taken by the STA, but it can do smarter decisions because it has additional informations.
As a coveveloper of 802.11k/v, Apple devices heaviliy relies on it for roaming. They roam/reconnect better than most other vendors if 802.11k/v is available. But they have in my experience more roam/connect issues than other brands if there is no 802.11k/v available.
As long as MikroTik has no support for 802.11k,v, Apple devices will never work as reliable with Tik APs in multi-AP setups as they do with other brands supporting 802.11k/v.
(Btw: 802.11r, often named together with k,v) does not directly influence the roaming decision, but allows STAs to reassociate much quicker with the new AP when roaming).

And Apple is a special case anyway:
Big vendors do extensive testing for compatibilty with Apple iOS devices. If they find WiFi bugs in iOS (and there are), they implement work arounds to fix iOS WiFi STA bugs at the AP end whenever possible. Because in the coorporate world, having issues with Apple iOS is a no-go. For customers, WiFi issues with Apple iOS are always the AP vendors fault. Even when its technically not true. Because "with Vendor XY all our iOS devices work, so its your fault". It is hard for MT to catch up here, Cisco's testlab alone has more employees than MT in total. And for sure Cisco, Aruba etc. have exclusive access to Apple Engineers and internal documenation small vendors as MT don't.
This is just a matter of fact due to the big market share of iOS devices in western companies: A WiFi setup having issues with the management's Apple gadgets won't stay long. Even if the issues are caused by Apple iOS bugs.

raskaton · Wed Feb 24, 2021 6:30 pm

There might be other checks. Those "internet" checks done by station devices (testing for reachability of "some" public server), also make the station device start trying other stored SSID's in its list, if the "internet check" test fails. Even if the current wifi connection is excellent (but gives no access to internet).

Yes! You're right! It's not a "whireless" problem.
You'll be surprised how simple it is. Just do:

IP -> DNS Settings
Max. Concurrent Queries = 500  (Default: 100)
Max. Concurrent TCP Sessions = 100  (Default: 20)

And no more communication losses at all!
The explanation is obvious.
1. After connecting, the client requests an IP address. If there is a problem with DHCP, the client will decide that the connection is not established and will disconnect.
2. After successfully obtaining the IP address, the client checks the availability of the Internet and services. It requests the DNS server received from DHCP. The problem is that not only some 8.8.8.8 is requested to check for Internet availability. Many services on your phone are trying to reach their "clouds" at once. There is a large peak load on the DNS. If there are more than 20 requests (as set default), the client is rejected and disconnected.
The problem is of a floating nature, because sometimes the client has time to make sure that the Internet is available. And sometimes other DNS queries skip ahead and block this check.
For me, it was enough to increase the allowed number of requests and sessions by five times.

And about other settings that were discussed. They are not relevant to the problem. For example, I have DHCP "Lease time = 1 hour", wi-fi Profile "Group key update = 1 hour", wi-fi "SSID name = abcd_B16". And all fine now becouse only DNS.
Of course, you need to use "WPA2 PSK" and "AES". Preferably "bridge port Learn = auto", "bridge RSTP", "DHCP Snooping".

There hasn't been a break for 21 hours. Previously, every 10-15 minutes, the devices fell off. Proof:

wi-fi up time.png

Tested on connected devices 2011UAS-2HnD (wi-fi) ++ Gigabit link ++ 2011L (DHCP and DNS servers) all in one net 192.168.1.0/24
and on simple hAP 951Ui-2nD (one bridge wi-fi, DHCP, DNS).

I will test more. But all the problems with my smart home devices and SIP softphone via wi-fi have already disappeared. No gaps, everything has become incredibly stable!
I'm happy.

P.S. Sorry for my English, I use a translator.

zandhaas · Thu Feb 25, 2021 10:47 am

Thank you for your tip.
I changed the values and keep an eye on the disconnects. :)

raskaton · Tue Mar 02, 2021 7:49 pm

Tested for a long time. Everything is fine. I only get error "leaving (8)" or "leaving (3)" when the phone goes out of wi-fi range. Smart home devices are stable.

wi-fi up time 2.png

By the way, another way to configure it is to specify the DNS of your Internet provider directly in the settings of the DHCP server. Who said that you need to process "Remote requests" in DNS yourself?

bandini981 · Wed Mar 03, 2021 6:26 pm

Same problem here.
Clients suddenly starts to randomly disconnect from APs.
Tried to set DNS queries and TCP session to 500/100 but did not solved the problem.

My CAPsMAN has Local Forwarding and Client to Client Forwarding enabled.

Any other suggestion?

bpwl · Wed Mar 03, 2021 8:53 pm

... look in the LOG for the reason mentioned there. (enable System >logging > Topics:wireless to see more information). Is it "station leaving (8)" ... or something else ?

bandini981 · Fri Mar 05, 2021 10:55 am

... look in the LOG for the reason mentioned there. (enable System >logging > Topics:wireless to see more information). Is it "station leaving (8)" ... or something else ?

Let me add some useful information about the network topology.
I have four CAPs:
1. Outside the build, outdoor installation
2. Ground floor
3. First floor
4. Second floor

Each CAP is connected by cable and carries three SSID (one real + two virtual); I know this may introduce some interference, but it worked fine until last week.

Here latest logs from CAPsMAN:
(from the client point of view, there is a disconnection due to "no internet connection available on WiFi network")

08:45:11 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 rejected, forbidden by access-list
08:45:30 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 rejected, forbidden by access-list
08:45:32 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 rejected, forbidden by access-list
08:45:42 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 rejected, forbidden by access-list
08:45:44 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap63 rejected, forbidden by access-list
08:45:47 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap69 connected, signal strength -57
08:46:15 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap69 disconnected, received deauth: sending station leaving (3)
08:48:03 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 connected, signal strength -67
08:48:19 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 connected, signal strength -54
08:48:19 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 disconnected, registered to other interface
09:03:08 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 disconnected, group key timeout
09:05:13 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 connected, signal strength -50

Here some CAPsMAN configuration details:

/caps-man access-list
add action=accept allow-signal-out-of-range=10s disabled=no interface=any signal-range=-80..120 ssid-regexp=""
add action=reject allow-signal-out-of-range=10s disabled=no interface=any signal-range=-120..-81 ssid-regexp=""

-

/caps-man configuration
add channel=channel1 datapath=loccal_forwarding_datapath datapath.arp=disabled hide-ssid=no mode=ap name=cfg1 rx-chains=0,1,2 security=security1 ssid=xxxxxxx tx-chains=0,1,2
add channel=channel6 datapath=loccal_forwarding_datapath hide-ssid=no mode=ap name=cfg6 rx-chains=0,1,2 security=security1 ssid=xxxxxxx tx-chains=0,1,2
add channel=channel11 datapath=loccal_forwarding_datapath hide-ssid=no mode=ap name=cfg11 rx-chains=0,1,2 security=security1 ssid=xxxxxxx tx-chains=0,1,2
add channel=channel3 datapath=loccal_forwarding_datapath datapath.vlan-id=10 datapath.vlan-mode=use-tag hide-ssid=yes mode=ap name=cfgSmartWifi security=security2 ssid=YYYYY
add channel=channel5 datapath=loccal_forwarding_datapath datapath.vlan-id=50 datapath.vlan-mode=use-tag hide-ssid=yes mode=ap name=cfgCameraWiFi security=security2 ssid=ZZZZZ

-

/caps-man datapath
add arp=reply-only client-to-client-forwarding=yes local-forwarding=yes name=loccal_forwarding_datapath

-

/caps-man channel
add band=2ghz-b/g/n control-channel-width=20mhz frequency=2412 name=channel1
add band=2ghz-b/g/n control-channel-width=20mhz frequency=2437 name=channel6
add band=2ghz-b/g/n control-channel-width=20mhz frequency=2462 name=channel11
add band=2ghz-b/g/n control-channel-width=20mhz frequency=2422 name=channel3
add band=2ghz-b/g/n control-channel-width=20mhz frequency=2432 name=channel5

-

/caps-man provisioning
add action=create-enabled comment="CAP 1" master-configuration=cfg1 radio-mac=XX:XX:XX:XX:D8:FD slave-configurations=cfgCameraWiFi,cfgSmartWifi
add action=create-enabled comment="CAP 2" master-configuration=cfg6 radio-mac=XX:XX:XX:XX:0D:A9 slave-configurations=cfgCameraWiFi,cfgSmartWifi
add action=create-enabled comment="CAP 3" master-configuration=cfg11 radio-mac=XX:XX:XX:XX:5A:7C slave-configurations=cfgCameraWiFi,cfgSmartWifi
add action=create-enabled comment="CAP 4" master-configuration=cfg1 radio-mac=XX:XX:XX:XX:E3:5E slave-configurations=cfgCameraWiFi,cfgSmartWifi

-

/caps-man security
add authentication-types=wpa-psk,wpa2-psk encryption=aes-ccm name=security1
add authentication-types=wpa-psk,wpa2-psk encryption=aes-ccm name=security2

-

/ip dns
set allow-remote-requests=yes max-concurrent-queries=1000 max-concurrent-tcp-sessions=200 servers=1.1.1.1,8.8.8.8,9.9.9.9,208.67.220.220,208.67.222.222,8.8.4.4,1.0.0.1

Many thanks,

ML

bpwl · Fri Mar 05, 2021 11:47 am

(3) is different from (8) as reason. (8) is the client who disconnects, (3) is because you kicked the client out !
https://aboutcher.co.uk/2012/07/linux-w ... son-codes/

This rejected and disconnected .....


08:45:11 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 rejected, forbidden by access-list
08:45:30 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 rejected, forbidden by access-list
08:45:32 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 rejected, forbidden by access-list
08:45:42 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 rejected, forbidden by access-list
08:45:44 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap63 rejected, forbidden by access-list
08:45:47 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap69 connected, signal strength -57
08:46:15 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap69 disconnected, received deauth: sending station leaving (3)
08:48:03 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 connected, signal strength -67
08:48:19 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 connected, signal strength -54
08:48:19 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 disconnected, registered to other interface
09:03:08 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 disconnected, group key timeout
09:05:13 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 connected, signal strength -50

Is caused by this .... (and even forbidden for 10 seconds, after the disconnect)
"action=reject allow-signal-out-of-range=10s " will keep the reject decision for 10s , even if the signal is out of the forbidden range

/caps-man access-list
add action=accept allow-signal-out-of-range=10s disabled=no interface=any signal-range=-80..120 ssid-regexp=""
add action=reject allow-signal-out-of-range=10s disabled=no interface=any signal-range=-120..-81 ssid-regexp=""

bandini981 · Fri Mar 05, 2021 12:11 pm

(3) is different from (8) as reason. (8) is the client who disconnects, (3) is because you kicked the client out !
https://aboutcher.co.uk/2012/07/linux-w ... son-codes/

This rejected and disconnected .....


08:45:11 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 rejected, forbidden by access-list
08:45:30 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 rejected, forbidden by access-list
08:45:32 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 rejected, forbidden by access-list
08:45:42 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 rejected, forbidden by access-list
08:45:44 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap63 rejected, forbidden by access-list
08:45:47 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap69 connected, signal strength -57
08:46:15 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap69 disconnected, received deauth: sending station leaving (3)
08:48:03 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 connected, signal strength -67
08:48:19 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 connected, signal strength -54
08:48:19 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap66 disconnected, registered to other interface
09:03:08 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 disconnected, group key timeout
09:05:13 192.xx.xx.xx caps,info XX:XX:XX:XX:07:F6@cap60 connected, signal strength -50

Is caused by this .... (and even forbidden for 10 seconds, after the disconnect)
"action=reject allow-signal-out-of-range=10s " will keep the reject decision for 10s , even if the signal is out of the forbidden range

/caps-man access-list
add action=accept allow-signal-out-of-range=10s disabled=no interface=any signal-range=-80..120 ssid-regexp=""
add action=reject allow-signal-out-of-range=10s disabled=no interface=any signal-range=-120..-81 ssid-regexp=""

Ok, thank you for the explanation. Do you have any suggestion to mitigate or resolve the issue?

bpwl · Fri Mar 05, 2021 3:09 pm

What I would try ...

add action=accept allow-signal-out-of-range=30s disabled=no interface=any signal-range=-86..120 ssid-regexp=""
add action=reject allow-signal-out-of-range=3s disabled=no interface=any signal-range=-120..-87 ssid-regexp=""

.
If you want to push off slow connections, rise the "basic rate" from 6 Mbps to 18 Mbps. (some experimenting to tune may be required. 12Mbps? 24Mbps?)

bandini981 · Fri Mar 05, 2021 7:16 pm

What I would try ...
Code: Select all
add action=accept allow-signal-out-of-range=30s disabled=no interface=any signal-range=-86..120 ssid-regexp=""
add action=reject allow-signal-out-of-range=3s disabled=no interface=any signal-range=-120..-87 ssid-regexp=""
.
If you want to push off slow connections, rise the "basic rate" from 6 Mbps to 18 Mbps. (some experimenting to tune may be required. 12Mbps? 24Mbps?)

Many thanks. Going to try.
Do you exclude some interference problem between real and virtual APs?

bpwl · Fri Mar 05, 2021 9:08 pm

Getting off topic (dissasoc station leaving (8)) so I keep it short.
- real and virtual channel share the same radio (so they can only send in turn, one by one, think as 3 input queue's feeding one output queue)
- co-channel interference (cooperation) is for all radio's on the same channel. They always wait for each other, and roll dice when the channel is free to choose who may transmit.
- adjacent channel interference ( channels overlapping in frequency) don't wait and disturb/destroy each others transmission

Channel 1, 6 and 11 = non overlapping. Channel 3 and 5 do overlap with channel 1 and 6. (Well ...euh... 1-5-9-13 is a possibility if no "b" is used.)

Valeriu · Fri Dec 31, 2021 11:41 am

I see this problem the same day I configure the router.

Now on some jobs the problem pops up a few days after I have left. Those ones... disabling the radio interface sometimes cures it. But the jobs where I see those messages right away... I have to change manufactures. Too many lost hours and clients.
Don't know if it can help but i have same trouble and i just change name of SSID on virtual wifi (i have trouble "station leaving" only on virtual wifi) with a short name without special chararcter and for moment it's ok.
before : Restaurant-Jeannala
After : Jeannala

Hello.

I have the same problem - when I tried to connect a new device, the log only had "Received disassoc sending station leaving" and "Unicast key exchange timeout" entries.
The new device is an Inventor Wi-Fi dehumidifier.

Before it was connected, everything worked fine - laptops, tablets, phones, IP cameras, a TV, Raspberry Pi, Arduino, and sensors for Google Home were connected. With the exception of only Samsung phones - it was necessary to enable TKIP - only AES did not work.

But yesterday I tried everything that was written above - I changed the channel width, frequency, encryption method, data rates, turned off chains except zero, etc. ...
Nothing helped.

Except change SSID - left only alphanumeric - other characters erased. It was "Devices-2Ghz" and changed to "Devices2Ghz". Dehumidifier сonnected instantly.
Sensors for Google-Home (SonOff) connected to the old SSID "Devices-2Ghz" without any problems.

Hence the conclusion - depending on the logic of the client side - that the creators of the device programmed in the firmware of the WiFi module. And, accordingly, you need to select your WiFi settings for the device.

No need to hait Mikrotik - I think it would be the same with another AP.

tommix · Wed Jan 12, 2022 2:32 am

[No need to hait Mikrotik - I think it would be the same with another AP.

Oh please. this problem continues 6 years! I NEVER had any problems with wifi until i started using Mikrotik.
I tried EVERYTHING written here-nothing helps. I use 5Ghz wifi, when i put load on it, lets say speedtest-it crashes. Laptop disconnects from router,,after some 10s reconnects.
Im using Mikrotik LTE 12 Chateau device. FW 7.2.RC1, i have no firewall rules or any cpu or ram loading scripts. CPU load stay at 1-3% ram 187MB.
There are such entries in log like "disconnected, received deauth: unspecified (1), signal strength -50" or "unicast key exchange timeout". But it doest happen all the time... if i reboot router and after reboot start speed test it will drop connection almost always. Sometimes there is no error even if debug wireless logging is enabled..all you see is " disconnected, ok, signal strength -52".
I DO NOT HAVE SUCH PROBLEM USING OTHER ROUTERS! It's not me, it's not my settings, it's Mikrotik.

I like MIkrotik it have some geeky charizma..but all those problems... kills any adtvantages. And yo cant return device... seller wont accept just because of "wifi disconnecting".

dbrowning2 · Thu May 12, 2022 12:10 am

I recently ran into this same issue after migrating from opnsense to MikroTik. I was having problems with some TP-Link kasa smart switches, specifically HS200 model. At DHCP renewal they would stop responding to Home Assistant. Put opnsense back and things worked great so I started to do packet captures. What I found was that the TP-Links never send DHCP option 55, which is a list of what it wants the server to send. Because of this MikroTik wasn't sending options 1 (subnet), 6 (router), and 12 (dns server). This is actually correct on MikroTiks part per rfc but opnsense always sends these, requested or not. I found with RouterOS7 you can force DHCP options. See:
https://help.mikrotik.com/docs/display/ ... HCPOptions

And specifically:
According to the DHCP protocol, a parameter is returned to the DHCP client only if it requests this parameter, specifying the respective code in the DHCP request Parameter-List (code 55) attribute. If the code is not included in the Parameter-List attribute, the DHCP server will not send it to the DHCP client, but since RouterOS v7.1rc5 it is possible to force the DHCP option from the server-side even if the DHCP-client does not request such parameter:
ip/dhcp-server/option/set force=yes

Once I configured options 1, 6, and 12 and applied those to my network under dhcp, everything started working. Doubtful most are running into this particular problem but hopefully it helps someone.

ivicask · Thu Sep 29, 2022 11:00 am

Getting this issue almost on daily on all locations and various clients, its driving me nuts, im watching on my Samsung phone and pressing connect and it keeps disconnecting then settles after 2-3 mins and then stays connected..

Znevna · Thu Sep 29, 2022 11:37 am

Something makes the phone decide to disconnect, that can be hard to pinpoint without some wireless sniffing.

pe1chl · Thu Sep 29, 2022 11:45 am

It looks like it has trouble getting a DHCP lease. That is enough reason for "modern" devices to disconnect the L2 connection as well.
Debug it from the DHCP side first.

ivicask · Thu Sep 29, 2022 11:50 am

It looks like it has trouble getting a DHCP lease. That is enough reason for "modern" devices to disconnect the L2 connection as well.
Debug it from the DHCP side first.

Well whats user @dbrowning2 above saying about forcing 1,6,12 options on DHCP ?I dont understnad hes post fully.

I dont have any DHCP issues on wired clients, its just problem on wireless.

gotsprings · Thu Sep 29, 2022 1:33 pm

Something makes the phone decide to disconnect, that can be hard to pinpoint without some wireless sniffing.

Nah... Its pretty easy to pinpoint. You just have to accept reality.

gotsprings · Thu Sep 29, 2022 1:39 pm

It looks like it has trouble getting a DHCP lease. That is enough reason for "modern" devices to disconnect the L2 connection as well.
Debug it from the DHCP side first.

I get that exact message and see that pattern all time. I can keep my Mikrotik Routers and Switches... But I can't keep Mikrotik wireless in there. Switch to anyone of 6 other vendors and problems stop.

The exception is I have been able to run Wave 2 drivers on an Audience and in a hAP AC3. These do not show these disconnect messages. And they actually keep clients connected. HOWEVER... I CAN'T USE CAPS-MAN AT THAT POINT. So their is no real reason to keep messing with it

ivicask · Thu Sep 29, 2022 1:40 pm

I also dont have this issue with other WIFI vendor on same network, had 2 weeks aruba APS zero issue, with mikrotik almost every day, sometimes 2-3 times a day..

Znevna · Thu Sep 29, 2022 1:45 pm

And any of you two sniffed the traffic to see what's pushing the phone(s) to disconnect?

gotsprings · Thu Sep 29, 2022 1:51 pm

And any of you two sniffed the traffic to see what's pushing the phone(s) to disconnect?

Sure have.

Did extensive trouble shooting with Mikrotik too. They finally copped to the fact that they can't deal with interference. They called my environment "unique."

Except I had the exact same problem, across nearly 100 UNIQUE SITES, all over the United States.

ivicask · Thu Sep 29, 2022 2:00 pm

And any of you two sniffed the traffic to see what's pushing the phone(s) to disconnect?
Sure have.

Did extensive trouble shooting with Mikrotik too. They finally copped to the fact that they can't deal with interference. They called my environment "unique."

Except I had the exact same problem, across nearly 100 UNIQUE SITES, all over the United States.

And exactly the same case my side, i got this issue more or less on all sites.

At home i have single AP, zero neighbour networks, couldnt be any interference, and still happens.

pe1chl · Thu Sep 29, 2022 2:04 pm

I get that exact message and see that pattern all time. I can keep my Mikrotik Routers and Switches... But I can't keep Mikrotik wireless in there. Switch to anyone of 6 other vendors and problems stop.

I agree with that. I do not roll out indoor office wireless using MikroTik APs until they have equipment that can cope with the standards of today. But that wasn't the question.

gotsprings · Thu Sep 29, 2022 3:17 pm

I get that exact message and see that pattern all time. I can keep my Mikrotik Routers and Switches... But I can't keep Mikrotik wireless in there. Switch to anyone of 6 other vendors and problems stop.
I agree with that. I do not roll out indoor office wireless using MikroTik APs until they have equipment that can cope with the standards of today. But that wasn't the question.

My problem...

My really serious problem...

Is there are some really smart people around here. Who are well written and have a deep understanding of Mikrotik.

But they have this enormous f--king blind spot when it comes to THE REAL WORLD. I get really irritated as I am no longer having intelligent conversations with these people. They write these long form impressive as s--t excuses. I mean that's what they are... "Blah blah blah blah... To get the most out of... Blah blah blah."

It always amounts to closing your eyes and plugging you ears and yelling "nah nah nah nah nah". While people point to actual results. Actual facts. Reality as some might call it.

I try to reel them back in, and point out that we are stuck on an incomplete 2014 wireless driver. I mean who around here has a primary device that old? Seriously who here is typing on this forum with their iphone 6? Who is walking around confused about why that thing's radio is completely outdated and antiquated?

Let me drive that point home ACV1 IS A 2014 WIRELESS PROTOCOL. It's enhancement/update was ratified to ACv2 in 2016.

So let's put that in perspective... Mikrotik has been getting it's a-- kicked by other vendors since 2014... And still puts out the exact same stuff. And the fan boys around here just won't give it up. They actively refuse to use other vendors. And you can show them in live time just how badly Mikrotik wireless is performing compared to a competitive solution.

Then the typical crap starts down a different path... "Well look at how much more that costs..."

See it no longer maters that it is inferior... It's cheap so you have to accept the tradeoffs....

Ohh.... See how that works.

own3r1138 · Thu Sep 29, 2022 3:35 pm

Seriously who here is typing on this forum with their iphone 6?

Me, don't judge me.

2022-09-29_16-01-13.jpg

GregC · Thu Sep 29, 2022 3:50 pm

I also have the same problem with all my sites across the United States. Complete agreement with @gotsprings and @pe1chl. My question to you all is what alternatives for APs do we have? If using APs not controlled by the CAPsMAN will be more of a challenge.

bpwl · Thu Sep 29, 2022 7:40 pm

It looks like it has trouble getting a DHCP lease. That is enough reason for "modern" devices to disconnect the L2 connection as well.
Debug it from the DHCP side first.
Well whats user @dbrowning2 above saying about forcing 1,6,12 options on DHCP ?I dont understnad hes post fully.

I dont have any DHCP issues on wired clients, its just problem on wireless.

There are known but unresolved problems with MT DHCP on wireless.
Very visible with eg the use of wifi-repeaters that will not pass MT DHCP leases, but MT seems to be less stable than other DHCP servers in general for wifi.
It's the combination of MT DHCP (since ROS 6.38) and MT wifi (wlan startup with connection as cause?).
Replace MT DHCP server 'OR' MT wifi and the problem is gone is most cases. But having the DHCP server on another MT sometimes solves it as well.
The DHCP "use broadcast" mode is weird, as 255.255.255.255 IP address is used in the DHCPoffer but the MAC address used in the DHCPoffer is still the CHAddr of the DHCPdiscover.
For a wifi-repeater the MAC address in the packet header is NOT the MAC address in the DHCPdiscover. (ARP IP-MAC and DHCP IP-MAC table is different). This DHCPoffer never reaches the DHCPclient. It should be flooded to all interfaces as the MAC-address is not in the "Node" table of the bridge, but for some reason it does not go to the WLAN interface(s).
And without valid DHCP, the client drops the wifi connection, "reason (8)", if not restored in time.
Without the wifi-repeater the DHCP problem on wifi is not consistent, actually it mostly works, but now and then there is a "deassign"

E.g. ... in the middle of the night, with dhcp3 only handling IOT devices (lease time is 4 hours, simple queue is DHCP scripted)

Klembord-2.jpg

I have no clue for the above DHCP deassigns, they seldom happen. Very bizarre, the wifi networks for dhcp0 and dhcp3 are not related (different AP, different GHz band).
So I added more logging via DUDE for all MT devices involved (Because DHCPs server, internet access, wifi interfaces, are all on different MT devices). Didn't happen yet.

Just wonder, as I do see my laptop loose its internet connection from time to time. even with the wifi still on. Devices (OSes) have all their own methods to figure out if a network is delivering Internet access or not. And they do drop the wifi connection in the case Internet access is not provided (anymore), and will jump to the next available SSID. (Maybe maybe, while dropping the current SSID, they might tell the DHCP server.) I mean, the root cause of the wifi disconnects (8) by the client, could be the Internet uplink, and that is causing the DHCP deassign. (ISP doing maintenance last night? It happens on VDSL, also that IP address changes sometimes.)

Devices do drop wifi if they perceive "no internet" conditions. [That is one of the challenges when programming my Airco with an APP on my smartphone, halfway through the installation, the smartphone disconnects the wifi connection with the Airco because there is no internet access that way.] That connection dropping is a major difference with ethernet connections.

own3r1138 · Wed Oct 12, 2022 5:53 pm

I successfully added DHCP options 1, 6, and 12. It works fine. I didn't lose connectivity with my phone anymore.
Thanks to @dbrowning2

2022-10-12_17-49-09.jpg

2022-10-12_18-08-42.jpg

Export

/ip dhcp-server option
add code=1 name=subnetmask value=0xffffff00
add code=6 force=yes name="WLAN-Domain Server" value=0xc0a85901
add code=12 force=yes name=Hostname value="\$(HOSTNAME)"
add code=6 name="LAN-Domain Server" value=0xc0445801
/ip dhcp-server option sets
add name=WLAN options="subnetmask,WLAN-Domain Server,Hostname"
add name=LAN options="subnetmask,LAN-Domain Server,Hostname"

Who is online