Need help - during video calls (Zoom, Slack, Teams, etc) getting random 'hungs' for 0.5-2 seconds each 5-15 minutes.

My setup - RB4011iGS+5HacQ2HnD as main router (RSTP 6000) - have 2 WANs, CAPsMAN manages all Caps I have, 2 x RBcAPGi-5acD2nD (Cap AC) (RSTP 9000), CSS610-8P-2S+ to poe some devices and 2 Caps (RSTP 8800), 2 x RBcAPL-2nD (Cap lite) (RSTP 9000), RB952Ui-5ac2nD (RSTP 9000). No VLANs, around 200 devices connected via ethernet or WiFi, most via 2.4GHz. Have several Firewall and NAT rules to block most ports and manage fail-over of 2 WANs, some devices have priority to one of the WAN all other to another WAN. Attached is config of main router, other have default settings.

Laptop MacM1, Windows-based 2x laptops, 2x iPhone 14 are connected via 5Ghz to main router - all of them are having this issue. I've tried to monitor pings, CPU/Memory usage, switched between 2 WANs - tried one, then another one. No errors I can see in logs or anywhere else.

Should I move to another vendor or try to somehow find the issue?

they all using older wireless package or the new wifi package?
also some configuration would be useful to see

Yes - they all have new wireless package as separate one, updated to 7.13.
Configuration attached as well, thank you.

Like I mentioned on your face book post...

Unplug all that mess

Hardwire a computer to the router.
Hook up one ISP to the router.
Try the services again.

Do the problems still exist?

I had similar symptoms with with flapping ethernet ports between RB4011 and the ISP's modem, but these showed up in the log.

I've changed second channel for 5GHz wifi on RB4011 from auto to 5795 as it was in status tab, reduced radio power to 10 for all Caps except RB4011 based on testing using WiFi Explorer moving around house and checked channel interference as well. will test it next few days.
Also I've tested via LAN cable for few days and there were no issues with laptop connected, as soon as I've disconnected it from cable issue appeared once again

That's what we call a wifi issue.

Didn't help much but noticed that "Tx Queue Drops" is very high - in 25 days more than 4 millions. Started to digg this issue and found few posts suggesting to manage Queues. Also found that some Ethernet's have thousands of drops and WANs as well. Changed queues to ethernet-default and wireless-default for corresponding interfaces. For those queues changed to PCQ with 5k Queue Size and 200000 Total Queue Size. Now do not see any drops for few hours during quite active calls on 3+ devices using Teams, Zoom, Skype. Still investigating it but it seems that Queues when you have 2x1Gb WANs, SFP with 10Gb, 4k cameras, 4k TVs and other staff are quite outdated and unable to manage packets queues - on some peaks there were 10-20 000 p/s at same time 'ethernet-default' with 'pfifo' have 50 packets queue size by default, as for me it should be at least 5000.

No errors but still hangs.

One of the laptops is now connected via LAN cable to RB4011 and do not see any hangs during a calls.
Another laptop is connected via WiFi and see hangs during a calls. At same time I do 'pings' and traffic monitoring from WiFi-connected laptop - traffic hangs for same time as during a call and shows 'gaps' in graphic - see attached images

No errors but still hangs.

One of the laptops is now connected via LAN cable to RB4011 and do not see any hangs during a calls.
Another laptop is connected via WiFi and see hangs during a calls. At same time I do 'pings' and traffic monitoring from WiFi-connected laptop - traffic hangs for same time as during a call and shows 'gaps' in graphic - see attached images

As @gotsprings already written "[it is] a wifi issue". You have not given any information about the hardware of your laptops using Microsoft Windows. However by your description and Intel's Advanced Intel® Wireless Adapter Settings page, the issue is at hand could be related to either of the following settings:

• Global BG scan blocking
  By default, the Wi-Fi adapter will perform periodic scan for other available Access Points (AP).
  Disabling this behavior can be helpful when using application software that is sensitive to brief interruptions in network connectivity. [emphasis added]

  • Always: Will not perform periodic scans for other available APs.
  • Never (default): Will perform periodic scans for other available APs.
  • On Good RSSI: Will only perform periodic scans for other available APs when the current AP’s signal strength is low.

  The default setting may causes the issues like the one you are experiencing. Either select On Good RSSI or Always - the former is the safer bet if you are using your devices at other locations too.

• Roaming aggressiveness
  This setting alters the signal strength threshold at which the WiFi adapter starts scanning for another candidate AP. The default value is Medium. Depending on the environment, one option may work better than the other. You may try other values to see which works best for your environment. However, it is recommended to revert back to the default (Medium) if no improvement is observed with other values.

  • Lowest: The WiFi adapter will trigger scan a for another candidate AP when the signal strength with the current AP is very low.
  • Medium-Low
  • Medium (default): Recommended value.
  • Medium-High
  • Highest: The WiFi adapter will trigger scan a for another candidate AP when the signal strength with the current AP is still good.

  You may want to make experiments with the Lowest and the Medium-Low settings.

Intel does not publish the general logic (including trigger thresholds) behind roaming decisions unlike Apple which does this both for its mobile devices and also for its laptops, besides the company also discloses information about 802.11k, 802.11r, 802.11v and 802.11w support in its products. These may help you in designing in your Wi-Fi network (by selecting manually the channel, the transmit power and the location for the APs).

By the way you may want to revisit the bridge priorities you've set on your devices as only the following hex(adecimal) values are valid: 0 (decimal 0), 1000 (decimal 4096), 2000 (decimal 8192), 3000 (decimal 12288), 4000 (decimal 16384), 5000 (decimal 20480), 6000 (decimal 24576), 7000 (decimal 28672), 8000 (decimal 32768), 9000 (decimal 36864), A000 (decimal 40960), B000 (decimal 45056), C000 (decimal 49152), D000 (decimal 53248), E000 (decimal 57344) and F000 (decimal 61440) as described in the Election process section of the documentation, also you may want to revisit the set port costs.

...

Thank you for extended explanation - will give it try.

Thats very SAD that I have to jump to every existing (and new) laptop and ask users to do some tweaks in somewhere to have stable Teams call. Just monitored Teams call and it uses 2-3MB/sec connection with screen sharing and there were a lot random drops. How to deal with 100MB/s and more speed that generates a lot of questions in my mind and disappointment...

Thank you for extended explanation - will give it try.

Thats very SAD that I have to jump to every existing (and new) laptop and ask users to do some tweaks in somewhere to have stable Teams call. Just monitored Teams call and it uses 2-3MB/sec connection with screen sharing and there were a lot random drops. How to deal with 100MB/s and more speed that generates a lot of questions in my mind and disappointment...

You're very welcome. Hope that it helps you to get the issue at hand settled.

As you have experienced 15 to 25 Mbit/sec bandwidth for a client is enough for work, what really matters is the steady low latency while providing this bandwidth to as many client as possible, not the high 800 Mbit/sec or more single user download bandwidth numbers that are advertised (directly or indirectly by influencers) and bragged of by quite some manufacturer. For high throughput and(/or) low latency workloads wired solutions (whether Ethernet over SFTP or optical cables, Fibre Channel, InfiniBand, or any other) are the way to go (even MikroTik has low cost and low power spine and leaf switches with 100 Gbit/s QSFP28 ports (and a not that low power one too, tough it still provides higher throughput per Watt than anything else in MikroTik's current portfolio except the other two mentioned here) and unlike the wireless products these do deliver the advertised switching speeds).

While we should all appreciate what un9edsda did for you here...

NO MOTHER F--KING WAY AM I TOUCHING ANY SETTINGS ON CLIENTS' DEVICES.

God I miss caps-man, but I do not miss all the client complaints!

Issue is solved (I hope).

There was a bottleneck in one of my configurations - hAP AC lite (RB952Ui-5ac2nD) was connected via 5GHz WiFi to RB4011. Placement is shown at this picture. Between them there are 2 brick walls. hAP AC lite has RTSP 4k camera connected that constantly sends traffic.

During monitoring I've found that "Tx/Rx CCQ" is on average 60% with quite often drops to 20-30%. At place of hAP there is quite crowded place from neighbors. So my first test was to switch WiFi bridge to use 2GHz instead of 5GHz. Then I changed channel width to 20MHz and CCQ increased to 70% with drops to 40%. During those drops if there 1 or more calls at same time there were some short 'hungs' of traffic flow (monitored it on separate PC via LAN connection) for 0.1-2 seconds. I assume there is some 'bug' in processing of breaking radio connection and traffic on RB4011 that produces 'deadlock' in queue processing as at same time sometimes there were 'spike' of CPU usage from 3-4% average to 20%.

As I have old RB2011 laying around I added it as dedicated WiFi bridge. Now RB2011 connected to RB4011 via LAN cable on 1Gb ports. On RB2011 nv2 bridge configured. On hAP AC lite nv2 station WDS configured to connect to RB2011. Both use 20MHz channel width and less crowded 2GHz frequence 2432. On this setup I see Tx/Rx CCQ on average 85% with drops to 65%.

And no issues on RB4011 during high load speed tests and many parallel calls on several devices. Will still monitor it to make sure there are no issues. Also speedtest showing 350Mb download and 550Mb upload. Previously there were 100-200Mb/200-300Mb with some 'hangs' during test.

Thank you.

Last 3 days WiFi is stable 100% + speed is increased significally.

My guess is still the same - WiFi bridge on RB4011<->hAP was unstable due to 2 brick walls - it received incomplete/corrupted packets that leads to deadlocks/crashes of queues processing and 'hungs' of the processing thread that produced interruptions of existing clients connections. Taking this to separate hardware RB2011<->hAP fixed issue.

Update.

Changed RB4011 -> RB5009UPr+S+IN. 2 WANs connected to LAN 2 & 3.
Changed RBcAPGi-5acD2nD APs to:
LAN1 (2.5Gb) connected to Unifi U6 Enterprise
LAN10 (1Gb) connected to Unifi U6 Pro
SFP1 (10Gb) connected to CSS610-8P-2S+IN which powers Unifi U6 Pro

Now all laptops have 1.2Gb WiFi connection. Speedtest 700/700. Download file from NAS at speed 920Mbps.
No hangs, no issues at all. Also smooth roaming between all APs.

RB5009 is better than UDM SE due to its hardware architecture. Also I'm in love of the flexibility and power of Mikrotik configurations and ability to control everything in details.

That's what we call a wifi issue.

Update.

Changed RB4011 -> RB5009UPr+S+IN. 2 WANs connected to LAN 2 & 3.
Changed RBcAPGi-5acD2nD APs to:
LAN1 (2.5Gb) connected to Unifi U6 Enterprise
LAN10 (1Gb) connected to Unifi U6 Pro
SFP1 (10Gb) connected to CSS610-8P-2S+IN which powers Unifi U6 Pro

Now all laptops have 1.2Gb WiFi connection. Speedtest 700/700. Download file from NAS at speed 920Mbps.
No hangs, no issues at all. Also smooth roaming between all APs.

RB5009 is better than UDM SE due to its hardware architecture. Also I'm in love of the flexibility and power of Mikrotik configurations and ability to control everything in details.

Yup

Yes - they all have new wireless package as separate one, updated to 7.13.
Configuration attached as well, thank you.

Well, "wireless" package is not wifi-qcom-ac package. According to the attached config you were still using legacy wireless. My condolences.

No errors but still hangs.

Indeed. Oh my dear, 2.4ghz on MT legacy wireless. I experienced all these troubles myself. 5ghz was okayish - not as fast as one would expect but reliable at least. As long I did not turn off and on the wifi on my Huawei tablet - that action triggered a "DFS radar detection". I used a scheduler script to work around. However, I have no need for high throughput/speeds as my ISP connection is only 100/50 LTE. But 2.4ghz I only used that band for portable devices and streaming on TV which could live with those "instability". Ping loss - check. Highly varying throughput - check. Random transmission timeouts because of mysterious interferences - check. One known problem at my home was the microwave. I am using that microwave maybe once or twice a week, but I noticed: every time the microwave was running - 2.4ghz wireless was dead. There was a 100% correlation. But I did not care much for this particular "noise" - because who cares for twice a week. But there were other "interference sources" apparently too - because where did the timeouts and ping losses came from? And I had one wireless client that lost connection to AP after some time and did not reconnect until restarted (it was a audio streaming device).

One day wifi-qcom-ac landed in MT universe. Uninstalled legacy wireless, installed wifi-qcom-ac. Resolved. Rock stable 2.4ghz wireless - check. Microwave running for minutes (making popcorn) and 2.4ghz wireless still useable - check. Reliable pings on 2.4ghz - check. Reliable connection - check. No more fake DFS detections - check. Formerly problematic clients now connect flawlessly - check. Device roaming - check. Stupid "poor man's band steering" access list no more needed. Can't say anything bad at all. Overall satisfaction with MT wireless increased. So the hardware was not the issue. It was all the way software/driver. Did a personal comparison with Unifi U6+ and I could not find a single benefit for me (other than: "I now have big wifi6 balls") that would justify the upgrade for me. Thanks again MikroTik who made it possible.

And specifically to the new devices:
Well done OP. Exchange old devices like "CAP lite" and even "CAP AC" with Unifi devices from a) different price range and b) vastly better specs. I mean: way better than any current MT wifi6 devices. It would be a shame to not see a tremendous improvement in all belongings (stability, speed, etc.) when upgrading a crapola old devices fleet of "CAP Lite" and "hap ac lite" with "U6 Pro" and "U6 Enterprise". MT does not even offer such powerful devices in their current product range. I am not aware of any MT wireless device which could do more than 2x2.