Problem

WiFi became completely unstable — both 2.4 GHz and 5 GHz bands affected simultaneously. Wired Ethernet worked perfectly throughout. The problem persisted across router reboots and WiFi radio restarts (interface WifiMaster0/1 down/up).

Symptoms

1. Wireless clients unable to auto-connect — must toggle WiFi off/on on the client device
2. When connected, connection drops after seconds to minutes
3. Drops are distance-dependent (worse at range) but also occur within 1 meter of the router
4. After drop, clients get stuck in "zombie state" — they think they're connected but the router has already deauthenticated them
5. Affects all client types: iPhones (iOS 18), iPads, Android (Galaxy S20 FE, Redmi A7 Pro), Huawei tablet, Denon AVR, Yandex Station, IoT sensors
6. ~15 WiFi clients on the network

Log evidence (from self-test and syslog)

PTK 4-way handshake timeouts (router sends msg 1, client never responds):

MIC corruption during key exchange:

Group key handshake timeout:

RSN IE failure:

Mass simultaneous disconnection — all clients on both bands at once:

Constant "retransmits limit reached" for multiple devices across all 3 days.

What was ruled out

• RF interference: Channel scan shows <10% load on both bands, minimal neighbor networks
• Neighbor issues: Adjacent apartments report no WiFi problems
• Channel selection: Tested manual channels (ch 11 on 2.4 GHz, ch 36/80 MHz on 5 GHz) — no improvement
• Band steering: Disabled via CLI — no improvement
• WPA3/FT: Disabled both — no improvement
• Radio restart: interface WifiMaster0/1 down/up — no improvement
• Router reboot: Full power cycle — no improvement
• Hardware failure: No kernel panics, watchdog resets, thermal warnings, or PHY errors in logs

Configuration at time of issue

• SSID: same on both bands (band-steering enabled)
• Security: WPA2 + WPA3 (SAE), PMF, Fast Transition (802.11r)
• 2.4 GHz: auto channel, 20/40 MHz, BGN+AX+BE
• 5 GHz: auto channel, up to 160 MHz, AN+AC+AX+BE
• WireGuard VPN active
• USB storage (Seagate 2TB NTFS) attached

Resolution

Switching from preview channel (5.1 Beta 1) to main channel (5.0 stable) immediately resolved all WiFi issues. No configuration changes were needed — same settings work perfectly on 5.0.

Self-test diagnostic files from during the issue are available on request.

Hello,

I would like to report a limitation (or possibly request a feature) regarding firmware update channels in the Wi-Fi System (Mesh).

On the main Keenetic router, it is possible to switch between different update channels (e.g., Main, Preview, Dev), and also downgrade to an LTS version if needed.

However, for extender/mesh devices connected within the Wi-Fi System, there is no option to control or change the update channel. As far as I can see, these devices do not provide access to LTS versions or any channel selection at all.

This creates an inconsistency, especially for users who prefer running stable/LTS firmware across all devices in the mesh network.

Is there currently a way to:

• Set the update channel (especially LTS) for mesh/extender devices?
• Or make extender devices follow the main router’s selected channel?

If not, I would like to suggest adding this feature for better consistency and control in mesh environments.

For some time i'am unable to flash update, issue popped out couple of months ago. I tried different channels and workaround, no effect.

Via web i see error after 50% complete

`The server is temporarily unavailable (code 0). Please try again later.`

with replacing firmware by .bin file

Failed to switch system boot partition

via telnet following error

`Core::System::Update::Hooks error[8586241]: failed to update the firmware: unable to commit an update.`

Self test log in attachment.

What may be wrong, how to fix this?

self-test.txt

I performed all these operations on devices running the following file systems: Ext4, exFAT, and FAT32. In every case, these files and directories became hidden. The data transfer protocol did not affect this either, as I tested it using both SMB and SFTP. The result remained the same.

In Windows 11, when using the SMB protocol, I was only able to make files and directories visible after enabling the «Show hidden files, folders, and drives» option in File Explorer and disabling «Hide protected operating system files (Recommended)».

In Android 14, when using Cx File Explorer and the SFTP protocol, these files and directories are not displayed at all. If you look at the number of items in the parent directory containing such hidden elements, all files and directories are present and are copied correctly when copying the parent directory in which they are located, but there is no way to interact with them. You can find a file manager that can handle system hidden items, but the problem here is that user files and directories that shouldn’t be system files become system files.

At first, I thought that KeeneticOS simply wouldn't return them when requested and would say they were missing, but that's not the case. All the files and directories don't disappear but become system files, which makes it impossible to work with them normally.

ENVIRONMENT

Router: Keenetic Titan KN-1812, KeeneticOS 5.00.C.7.0-0
ISP: Magyar Telekom, fiber GPON, 1 Gbps symmetric
Topology: MBP (Wi-Fi 6) → Keenetic → Magenta Box (Technicolor FGA2235, DMZ mode) → ISP
Wi-Fi client: MacBook Pro M1 Max (BCM4387), 5 GHz, 2SS/80MHz, 802.11ax
MBP is the only client on 5 GHz band
IntelliQoS config when enabled: CAKE, input 890 Mbps, output 500 Mbps

PROBLEM

When IntelliQoS is disabled (no qos), Wi-Fi upload through WAN drops to 30-50 Mbps. With IntelliQoS enabled (CAKE, output 500 Mbps), upload is stable at ~480 Mbps.

I would expect that removing the 500 Mbps shaper would increase or maintain upload speed, not collapse it by 10x.

DIAGNOSTIC RESULTS

Upload throughput (same iperf3 server, same time of day):

Pi-hole, Ethernet → Keenetic → WAN, QoS OFF:       846 Mbps
MBP, Wi-Fi 5GHz → Keenetic → WAN, QoS OFF:         23-51 Mbps
MBP, Wi-Fi 5GHz → Keenetic → WAN, QoS ON (500):    480 Mbps
MBP, Wi-Fi 5GHz → Keenetic → LAN (iperf3), QoS OFF: 595 Mbps
MBP, Wi-Fi 5GHz → Magenta Box Wi-Fi → WAN, QoS OFF: 610 Mbps
Desktop, Ethernet → Keenetic → WAN, QoS OFF:        450-640 Mbps
Desktop, Ethernet → Magenta Box directly, QoS OFF:  980 Mbps

Key observations:
1. Wi-Fi LAN throughput = 595 Mbps (iperf3) — Wi-Fi radio works fine
2. Ethernet upload through same WAN path = 846 Mbps — WAN path works fine
3. MBP on Magenta Box Wi-Fi directly = 610 Mbps — MBP Wi-Fi chip works fine
4. PHY rate during test: TX 1201, RX 960-1080 Mbps, RSSI -67...-69 — stable
5. WAN interface stats: 0 txdropped, 0 txerrors
6. Problem is specific to Wi-Fi → WAN forwarding without CAKE

What I tested (none helped):
- Disabled hardware PPE (no ppe hardware) — upload still 44 Mbps
- Disabled TSO on MBP (sysctl net.inet.tcp.tso=0) — upload still 45 Mbps

QUESTION

Is this expected behavior? It appears that the default queueing on the WAN egress path cannot handle Wi-Fi-originated traffic properly. CAKE fixes it, but requires an artificial bandwidth limit (500 Mbps) which caps upload below ISP capacity (980 Mbps).

Is there a way to enable CAKE/fq_codel without bandwidth limiting? Or another setting to fix Wi-Fi → WAN forwarding without IntelliQoS?
 

{
	"prompt": "(config)",
	"status": [
		{
			"status": "message",
			"code": "8781924",
			"ident": "Core::Legal",
			"message": "\"dpn\": accepted version 20230223."
		}
	]
}

then I pressed acquire buttom the button disappeared, so I reset to factory settings the Air, wait air comes up, but no more acquire button. So I reboot the Sprinter, again no button, I want to type again the command but can't find it more, can you please write it me here? Thank you

I'm using Keenetic Viva (KN-1910) and have a configured Wireguard tunnel between two Keenetic routers (the second one is a Keenetic Ultra (KN-1810)).

After update the Keenetic Viva to 5.0.4, I can't connect to local devices on the side of Keenetic Ultra, but the Wireguar connection looks ok in the web ui (green and has a connected status on both sides).

When I downgrade firmware of Keenetic Viva to 4.3.6.3 - everything works perfect again.

There is no difference of Keenetic Ultra's firmware - tested on 4.3.6.3 and 5.0.4.

Could you check what is going wrong with 5.0.4? I'm ready to provide any logs you need.

Thanks in advance!

In the Install Updates screen, when a large number of components are added, the Confirm button is not visible as it gets pushed down, and the page cannot be scrolled.

Please check and resolve this issue.

Thanks.

Core::Configurator: not found: "show/sc/mws/wlan" [http/rci 192.168.1.29]

Core::Configurator: not found: "show/mws/wlan" [http/rci 192.168.1.29]

https://ibb.co/yngC12Ch

After installing the Alpha 3 version, I started experiencing issues with firmware updates.

On the Hopper, Sprinter, and Viva models, I am unable to install either the main (stable) or the dev versions.

The main unit updated to the preview version but it is impossible to update the member to the same version as it stays on the Main channel.

- it shown in interface as ON, and configured

- doesn't show up on network player

- doesn't show up in logs (there's no `Dlna::Server: service started.` message)

- reinstalling component doesn't help

- downgrading to 5.0.6 helps

context: router has "recommended" components installed + dlna, transmission, ftp, smb, DoH, DoT, ext4 (no opkg)

I have 2 Titan units, one router and another mesh access point.
A lot of my devices who are connected to the access point show 2500 Mbits/s instead of Speed and security protocol as show on this pic.

Is this a bug?
Thanks

This bug appeared in all the applications I used for testing. The only application that worked correctly with an SFTP connection and did not allow moving and renaming items was FileZilla (Windows 11). Why does everything work correctly only in this application? The SFTP server sends the correct access attributes to the file system elements, which are displayed in the FileZilla application. Most likely, only this application takes these access attributes into account and does not even attempt to rename or move files, while all other applications simply ignore the access attributes and perform the operation.

When using the SMB protocol, all permissions work correctly in all applications.

Here is what I used for the tests:

• KeeneticOS: v4.3.6.2 (delta version from https://osvault.keenetic.net)
• Router: ZyXEL Keenetic Ultra 2
• Apps:
  • Cx File Explorer (Android 14)
  • Cyberduck (Windows 11)
  • FileZilla (Windows 11)

Despite the fact that v3.5.10 is the latest stable version for the router, and I installed the latest «delta» version, I don't think this bug only affects this router. I didn't check for the bug in v3.5.10 because I'm completely satisfied with v4.3.6.2. Everything suits me, except for this bug. Although I have the option to install version v5.* for this router, I have not seen any mention in the «Changelog» section of this forum that this bug has been fixed, so I am in no hurry to change the operating system version.

Hello, I'm currently connected to the router via Wireguard VPN from another computer, but I can't connect to other devices on the router's mesh network via acme default. The latest update notes list a fix, but the problem persists.

Hello, I'd like to report an issue with the DNS-based route addition page. I configured my Warp Wireguard connection as shown in the image. Both initially worked without issue, but after turning the setting off and on, adding new domains, or deleting and re-adding them completely, the "domain" redirection in the route list became corrupted. Even deleting profile and re-adding them didn't work properly. The IP side appears to be working fine. Do you experience this issue as well? I'd appreciate it if you could try it and report back.

5.00.B.3.0-2 - 1811

Перезагрузка может произойти в рандомное время, раз в сутки, либо 2 раза в сутки.

В журнале никаких ошибок, свет не отключался, роутер подключен к бесперебойнику который работает 30 минут после отключения света, к нему же подключен ПК.

Only EFI is visible, nothing else


On version 4.2.6 everything is working fine.

Logs:

Авг 13 08:01:54

kernel

sd 0:0:0:0: [sda] Very big device. Trying to use READ CAPACITY(16).

Авг 13 08:01:54

kernel

sd 0:0:0:0: [sda] 9767475200 512-byte logical blocks: (5.00 TB/4.55 TiB)

Авг 13 08:01:54

kernel

sd 0:0:0:0: [sda] 4096-byte physical blocks

Авг 13 08:01:54

kernel

sd 0:0:0:0: [sda] Write Protect is off

Авг 13 08:01:54

kernel

sd 0:0:0:0: [sda] Mode Sense: 47 00 10 08

Авг 13 08:01:54

kernel

sd 0:0:0:0: [sda] No Caching mode page found

Авг 13 08:01:54

kernel

sd 0:0:0:0: [sda] Assuming drive cache: write through

Авг 13 08:01:54

kernel

sd 0:0:0:0: [sda] Very big device. Trying to use READ CAPACITY(16).

Авг 13 08:01:54

kernel

sda: sda1 sda2 sda3

Авг 13 08:01:54

kernel

sd 0:0:0:0: [sda] Very big device. Trying to use READ CAPACITY(16).

Авг 13 08:01:54

kernel

sd 0:0:0:0: [sda] Attached SCSI disk

Авг 13 08:01:54

ndm

Storage::Manager: "Media0": enabled a disk spin down.

Авг 13 08:01:54

ndm

Storage::Manager: "Media0": enabled TRIM support.

Авг 13 08:01:54

ndm

Core::System::DriverManager: loading /lib/modules/4.9-ndm-5/nf_conntrack_ftp.ko.

Авг 13 08:01:54

ndm

Core::System::DriverManager: loading /lib/modules/4.9-ndm-5/nf_nat_ftp.ko.

Авг 13 08:01:54

ndm

Storage::Manager: created "Media0", port 1.

Авг 13 08:01:54

ndm

Storage::Partition: vfat "67E3-17ED:": filesystem initialized.

Авг 13 08:01:54

kernel

tfat: fail_safe is enabled

Авг 13 08:01:54

kernel

tfat: cluster_heap_lbo 0x317c00

Авг 13 08:01:54

kernel

tfat: 9 blkbits for normal inodes

Авг 13 08:01:54

kernel

tfat: fat start lbo 0x4000

Авг 13 08:01:54

kernel

tfat: 9 blkbits for main FAT32 inode

Авг 13 08:01:54

kernel

tfat: cluster size 512

Авг 13 08:01:55

ndm

Core::Server: started Session /var/run/ndm.core.socket.

Авг 13 08:01:55

upnp

HTTP listening on port 1900

Авг 13 08:01:55

upnp

Listening for NAT-PMP/PCP traffic on port 5351

Авг 13 08:01:55

kernel

tfat info: FAT32 volume name 'EFI', version 0.0.

Авг 13 08:01:55

ndm

Storage::Manager: "Media0": registered partition "67E3-17ED:".

Авг 13 08:01:55

kernel

thfsplus: unable to parse mount options

Авг 13 08:01:55

ndm

Storage::Partition: hfsplus "64043940-5674-3e73-8904-6f56f3b2eed4:": filesystem initialized.

Авг 13 08:01:55

ndm

Core::FileSystem::Native: failed to mount "/tmp/mnt/64043940-5674-3e73-8904-6f56f3b2eed4": invalid argument.

Авг 13 08:01:55

ndm

Storage::Partition: failed to mount.

Авг 13 08:01:55

ndm

Storage::Partition: hfsplus "64043940-5674-3e73-8904-6f56f3b2eed4:": unable to mount.

Авг 13 08:01:55

ndm

Storage::Manager: "Media0": failed to register partition "64043940-5674-3e73-8904-6f56f3b2eed4:".

Авг 13 08:01:55

ndm

Storage::Partition: "/dev/sda3" has an unknown partition type, ignored.

Авг 13 08:01:55

kernel

sd 0:0:0:0: [sda] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08

Авг 13 08:01:55

kernel

sd 0:0:0:0: [sda] tag#0 Sense Key : 0x5 [current]

Авг 13 08:01:55

kernel

sd 0:0:0:0: [sda] tag#0 ASC=0x20 ASCQ=0x0

Авг 13 08:01:55

kernel

sd 0:0:0:0: [sda] tag#0 CDB: opcode=0x93 93 08 00 00 00 00 00 00 20 e7 00 00 00 01 00 00

Авг 13 08:01:55

kernel

blk_update_request: critical target error, dev sda, sector 8423

When I use regular DNS (ISP DNS, Google, Cloudflare, etc.), PPPoE connects just fine. But as soon as I enable DoH/DoT, the PPPoE session starts always showing "Authentication failed". In addition, the connection problem occurs when restarting the PPPoE service or modem.

After looking into it, here’s what seems to be happening:

• With DoH/DoT enabled, the router first tries to resolve the hostname of the secure DNS server.

• It attempts to do this through the PPPoE interface, but at that point the PPPoE session hasn’t been fully established yet. (Because when I add a normal DNS address, it connects after a while.)

• This premature attempt causes the router to trigger a PPPoE authentication error, even though the username and password are correct.

• With normal DNS, there’s no need for an early hostname lookup, so the PPPoE session comes up cleanly and only afterwards DNS queries flow through meaning no errors appear.

Problem isn’t wrong credentials, but rather Keenetic’s DoH/DoT “bootstrap” mechanism trying to use the PPPoE link before authentication has completed.

Both of their firmware ver is  4.3.6

Roaming for wireless clients is working without problem but roaming is not working for guest segment.

Any clue?

Best Regards,