2.7.0 and 2.7.1

[Koenkk]

Feel free to discuss the 2.7.0 and 2.7.1 release here!

Note that the 2.7.1 release is a hotfix release to resolve an issue causing Z2M to crash from excessive disk writes (e.g. #29971, #29957, #29999)

[lantech69420]

*EDIT: It now works. Thanks!

I just got a notification in HA about it.

However it won't update:

"Failed to perform the action update/install. Error updating Zigbee2MQTT: Can't install ghcr.io/zigbee2mqtt/zigbee2mqtt-amd64:2.7.0-1: DockerError(404, 'manifest unknown')"

I'm not running in Docker. My current version is 2.6.3-1

any ideas?

[chris-one-one]

same problem here... no docker, only HA OS

[Betoniara]

Why does Home Assistant report that a Zigbee2MQTT update is available even if this version isn’t intended for the HA add-on?
And why are releases not published to all installation types (manual, Docker, HA add-on) at the same time?

From a user perspective it’s confusing: HA shows an update that cannot or should not be installed yet.

[BertLDN]

Same problem here, 2.7.0-1 not available. Running HAOS on bare metal

[eigger]

same issue here.. It worked in beta 25.12 (arm), but it doesn’t update in version 25.11 (x86).

[RikvdMark]

I had the same issue on HA OS. I tried again a few minutes ago and now everything updated correctly.

[jkosharek]

Baremetal HA OS Install
Installation method: Home Assistant OS
Core: 2025.11.3
Supervisor: 2025.11.5
Operating System: 16.3
Frontend: 20251105.1

Failed to perform the action update/install. Error updating Zigbee2MQTT: Can't install ghcr.io/zigbee2mqtt/zigbee2mqtt-amd64:2.7.0-1: DockerError(404, 'manifest unknown')

[huizebruin]

Here also

[arian-nasr]

same issue for me too

[SebMagda]

same here

[WzL]

Same here, HA OS on bare metal,
[139622619436416] Error during service call to update.install: Error updating Zigbee2MQTT: Can't install ghcr.io/zigbee2mqtt/zigbee2mqtt-amd64:2.7.0-1: DockerError(404, 'manifest unknown')

Upgrading from 2.6.3-1 to 2.7.0-1

[WzL]

I did Rerun the update this morning. It finished without issues this time. Not sure what fixed it though, great work indeed!

[EinSchwerd]

Same issue when trying to upgrade 2.6.3-1 to 2.7.0-1 on

[RoFrie]

Same issue here

[bdari66]

I have the same problem too, it doesn't update :(

[BertLDN]

Seems to solved, updated to 2.7.0-1 with no issues. Thanks for the rapid action.

[RikvdMark]

Same for me. Had issues last night but I also successfully updated a few minutes ago :)

[GigiPompieru]

It seems that now when opening Z2M it defaults to Overview. Is there any way to make it remember the last selection? (I'm used to going to the device tab)

[WoBBeLnl]

No issue updating with HA OS. Keep up the great work!

[BasJansen]

Couldn't update yesterday but now it does.

[woody4165]

Hi all

I've tried to upgrade from 2.6.3 to 2.7.0 but the the service did not come up.
I was on node v20.18.0

After struggling some time, I decided to copy my zigbee2mqtt folder to a backup one and follow the official guide to install it from scratch.

I decided also to upgrade node to v22.x

I don't know if it was the node upgrade, but now it works again.

[hanifsuandy]

I had similar issue but I resolved it differently without updating node.js.

After the Z2M update, I simply did apt update and apt upgrade on the system and then restarted the Z2M service after which 2.7.0 started fine.

[BitterHA]

Hi folks,

After upgrade to 2.7.0-1 my Z2M was becoming very slow with reaction time around 2 minutes in the webbrowser that is always op. I have a dashboard with many sensors to control. Even after reboot several HA times, reboot my synology and VMM : nothing changes. I'm running all latest updates and drivers, even I putted my HA Skyconnect on the latest version : no result. it's like Z2M put enormous data on the MQTT. in the new start up dashbord you see the resent actions. every second Z2M push around 10 messages to the MQTT. I think that the problem is somewhere there. I downgraded again to 2.6.3 and all these problems were prompt solved with back again direct action if I toggle a switch.

The strange thing is : I have a second setup on another location, completly the same hardware exept the Z2M coordinator that is SLBZ (and not HA Skyconnect) and there is no problem with the upgrade and with also the same amount of devices and sensors. I don't get it.

some numbers of my setup :
HA running in VMM on Synology DS218+ with the latest DSM 7.2.2-72806 update 5 (running smoothly) - upgrade to 10GB RAM
CPU HA : 8% - RAM HA : 25% (everything healty)
HA OS - core 2025.11.3 - supervisor 2025.11.5 - OS 16.3 - Frontend 20251105.1
Z2M (back again to 2.6.3) - Frontend 2.2.3 - Zherdsman-converters : 25.59.0 - Zherdsman version 6.3.2
using HA Skyconnect coordinator on EmberZNet - revision 7.4.4 (GA)
Z2M Machine : #1 SMP PREEMPT_DYNAMIC Tue Nov 4 13:52:04 UTC 2025 - 6.12.51-haos - x64
CPU: Intel(R) Celeron(R) CPU J3355 @ 2.00GHz (x2) - RAM: 1967 MB - Node version v22.16.0
Running 60 Z2M devices

I hope that some users have the same issue, looking forward for replays or solution

[BitterHA]

I really dont understand anymore. I set up an older configuration with same synology on a 3th independent location but as an empty test, I did the upgrade and everything direct communication. What is wrong with my operational setup? @Koenkk
quick setup Z2M (by the way I'm from the Flanders) :
Zigbee2MQTT-versie
2.7.0
commit: unknown
Frontendversie
2.4.2
zigbee-herdsman-converters versie
25.80.0
zigbee-herdsman versie
7.0.1
Machine
#1 SMP PREEMPT_DYNAMIC Tue Nov 4 13:52:04 UTC 2025 - 6.12.51-haos - x64
CPU: Intel(R) Celeron(R) CPU J3355 @ 2.00GHz (x2)
RAM: 1967 MB
MQTT
mqtt://core-mosquitto:1883
Versie: 4
Node-versie
v22.16.0
Coördinator
EmberZNet
0xbc026efffe22a784
Revisie: 7.4.2 [GA]

[Yves-Be]

My log is getting spammed with this after upgrading:

[12/2/2025, 8:47:35 PM] zh:ember:ezsp: ERROR Transaction failure; status=ERROR_WRONG_DIRECTION. Last Frame: [FRAME: ID=52:"SEND_UNICAST" Seq=78 Len=24].
[12/2/2025, 8:47:35 PM] zh:ember:ezsp: The adapter has run out of buffers, causing general malfunction. Remediate network congestion, if present. Last Frame: [FRAME: ID=52:"SEND_UNICAST" Seq=78 Len=24].

I did not have this issue in 2.6.3-1. This is on a SLZB-07 with 8.0.2.

[dansteigerwald]

Running HA on a Green. The 2.7.0 update was a mess. Got back on line with ChatGPT and now running on 2.6.3.

[woody4165]

I've got issues also upgrading from 2.6.3 to 2.70 with update.sh

I still don't know what was the cause, but after some time spent trying to understand the issue, I've decided to install it from scratch and I've done these steps:

• upgraded node from v20.18.0 to v22.21.0
  • sudo rm /etc/apt/sources.list.d/nodesource.list
  • curl -fsSL https://deb.nodesource.com/setup_22.x | sudo -E bash -
  • sudo apt-get update
  • sudo apt-get install -y nodejs
  • node --version
  • npm --version
• moved actual z2m folder to another one with mv /opt/zigbee2mqtt /opt/zigbee2mqtt_old
• and followed the official guide to install z2m (https://www.zigbee2mqtt.io/guide/installation/01_linux.html#installing)
• sudo mkdir /opt/zigbee2mqtt
• sudo chown -R ${USER}: /opt/zigbee2mqtt
• Cloned Zigbee2MQTT repository
• git clone --depth 1 https://github.com/Koenkk/zigbee2mqtt.git /opt/zigbee2mqtt
• Installed dependencies (in my case as root since I'm in Proxmox container)
• cd /opt/zigbee2mqtt
• pnpm install --frozen-lockfile
• and copied back the data folder with cp -r /opt/zigbee2mqtt_old/data .

And at the end I've started z2m and it worked
cd /opt/zigbee2mqtt
pnpm start

and followed the guide to run it as a daemon with systemctl

Now it's working fine.

Having Proxmox, I've got a backup/snapshot of the container so in any case the issue was going badly, I had the opportunity to rollback to 2.6.3 working z2m

I hope it can help

[anateodora140381]

To me, some lights are randomly keep turning on at minimum brightness. I tried repairing, restart coordonator, nothing works....
Finally downgrade was the solution.....

[snok3r-krk]

After updating to version 2.7.0 of the HAOS add-on, I noticed a significant increase in disk writes—approximately 200MB per hour with around 140 connected devices. 😨 My logs are set to 'Warning,' and I have verified that they are functioning correctly.

Has anyone else experienced this behaviour? I checked the config/logs/db, but I didn't find anything unusual. After downgrading to version 2.6.3-1, the situation returned to normal.

I have noticed in the release notes that 'Activity' and other features were added to the frontend. Could this be related? In the changelog, I could not find anything that would explain such a situation.

[BitterHA]

After updating to version 2.7.0 of the HAOS add-on, I noticed a significant increase in disk writes—approximately 200MB per hour with around 140 connected devices. 😨 My logs are set to 'Warning,' and I have verified that they are functioning correctly.

Has anyone else experienced this behaviour? I checked the config/logs/db, but I didn't find anything unusual. After downgrading to version 2.6.3-1, the situation returned to normal.

I have noticed in the release notes that 'Activity' and other features were added to the frontend. Could this be related? In the changelog, I could not find anything that would explain such a situation.

exactly the same, the activity is that big that I have a hugh delay on actions that I want to send to devices. after downgrading to 2.6.3 everything went back to normal. I did not find already a solution for it. Waiting for the clever guys :-)

[Koenkk]

Issue has been fix in the dev branch, let us know if it works, then we'll do a hotfix release.

Changes will be available in the dev branch in a few hours from now and in the next release which is every 1st of the month.

[tlovitz]

Issue has been fix in the dev branch, let us know if it works, then we'll do a hotfix release.

Changes will be available in the dev branch in a few hours from now and in the next release which is every 1st of the month.

Has anyone tested this out? If not, I'll bite the bullet and test it in Home Assistant. Zigbee becomes unusable and the log just gets hammered. For what it might be worth for diagnostic purposes, I've pasted the settings yaml below

data_path: /config/zigbee2mqtt
socat:
enabled: false
master: pty,raw,echo=0,link=/tmp/ttyZ2M,mode=777
slave: tcp-listen:8485,keepalive,nodelay,reuseaddr,keepidle=1,keepintvl=1,keepcnt=5 options: "-d -d"
log: false
mqtt: {}
serial: port: >- /dev/serial/by-id/usb-Nabu_Casa_SkyConnect_v1.0_70e5e293c15aed11a21eef165820eef3-if00-port0
adapter: ezsp
server: mqtt://localhost:1883
user: xxx
password: xxx
port: >- /dev/serial/by-id/usb-Nabu_Casa_SkyConnect_v1.0_70e5e293c15aed11a21eef165820eef3-if00-port0

Apologies if I've generated a lot of email notifications trying to get the formatting to work. I forgot there was a preview pane.

[yassore]

in Proxmox, zigbee2mqtt as LXC, SLZB-06p7(20240710, core v3.1.3)

• update to 2.7.0 from 2.6.3 is work good.
• #10896 Detect _TZ3210_a04acm9s as Tuya TS0001_fingerbot (@homeclubber) is fixed, but i dont get battery level , its always 1%, its usb c battery fingerbot.

[joonasgalaxy-eng]

After update Zigbee network got super slow!
Can't get the map made -> takes forever to make
Data still flows from energymeters but cant use Zigbee plugs -> commands wont work

Sonoff dongle has been updated, devices are Neo, Sonoff, Blitzwolf etc thta won't work.

Before update there was no worries.

BR Joonas

[joonasgalaxy-eng]

How does one downgrade?

[BitterHA]

restore your backup from before

[anateodora140381]

Restore from HA backup

[dansteigerwald]

All that needs to be restored is the Add-on. It was quick and effective.

[Vaarinvaliliha]

All that needs to be restored is the Add-on. It was quick and effective.

Worked for a moment, stopped working even with the older version.

[4slash1977]

After upgrading to 2.7.0-1 z2m stopped working in few hours after installation all devices went offline and when I tried to restart z2m it din't start enymore. My coordinator is SLZB-06P7 and error that I get is:
Failed to connect to the adapter (Error: SRSP - SYS - ping after 6000ms)

[Vaarinvaliliha]

My devices wont work, plugs and such. Sonoff dongle (latest firmware) + HAOS (latest core). Did the hotfix update last night. Worked for a brief moment and after an hour nothing. In the morning I did restart for home assistant and no control over Zigbee devices at all.

[anateodora140381]

CK-BL702-AL-01(7009_Z102LG03-1)
These bulbs are turning on to minimum brightness, by themselves after the update.
Only downgrade to 2.6.3 foxes the problem, I have tried all the rest... restarting, repairing and tweaking all bulb settings exposed to z2m...

[tomahawk5000]

can someone confirm that 2.7.1v fixed issues with 2.7.0 which won't appears on 2.6.3v? I had to downgrade my z2m

ok now I can confirm that 2.7.1 works fine for me. Thanks for this hotfix

[CCupler]

Updated from 2.6.3 just now (to 2.7.1), and, same issue that I had when I updated to 2.7.0: Z2M will not start. Downgraded again to 2.6.3 to get functionality back.
I'm not seeing here why 2.7 is having issues with the adapter (an HA ZBT-1). Just that it is having issues.

[21:54:03] INFO: Preparing to start... [21:54:03] INFO: Socat not enabled [21:54:03] INFO: Starting Zigbee2MQTT... Starting Zigbee2MQTT without watchdog. [2025-12-08 21:54:04] info: z2m: Logging to console, file (filename: log.log) [2025-12-08 21:54:04] info: z2m: Starting Zigbee2MQTT version 2.7.1 (commit #unknown) [2025-12-08 21:54:04] info: z2m: Starting zigbee-herdsman (7.0.4) [2025-12-08 21:54:04] info: zh:adapter:discovery: Matched adapter=ember path=/dev/ttyACM0, score=2 [2025-12-08 21:54:04] info: zh:ember: Using default stack config. [2025-12-08 21:54:04] info: zh:ember: ======== Ember Adapter Starting ======== [2025-12-08 21:54:04] info: zh:ember:ezsp: ======== EZSP starting ======== [2025-12-08 21:54:04] info: zh:ember:uart:ash: ======== ASH Adapter reset ======== [2025-12-08 21:54:04] info: zh:ember:uart:ash: ASH COUNTERS since last clear: [2025-12-08 21:54:04] info: zh:ember:uart:ash: Total frames: RX=0, TX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Cancelled : RX=0, TX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: DATA frames : RX=0, TX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: DATA bytes : RX=0, TX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Retry frames: RX=0, TX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: ACK frames : RX=0, TX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: NAK frames : RX=0, TX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: nRdy frames : RX=0, TX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: CRC errors : RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Comm errors : RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Length < minimum: RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Length > maximum: RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Bad controls : RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Bad lengths : RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Bad ACK numbers : RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Out of buffers : RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Retry dupes : RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: Out of sequence : RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: ACK timeouts : RX=0 [2025-12-08 21:54:04] info: zh:ember:uart:ash: ======== ASH stopped ======== [2025-12-08 21:54:04] error: zh:ember:uart:ash: Failed to init port with error Error: Error Resource temporarily unavailable Cannot lock port [2025-12-08 21:54:04] error: z2m: Error while starting zigbee-herdsman [2025-12-08 21:54:04] error: z2m: Failed to start zigbee-herdsman [2025-12-08 21:54:04] error: z2m: Check https://www.zigbee2mqtt.io/guide/installation/20_zigbee2mqtt-fails-to-start_crashes-runtime.html for possible solutions [2025-12-08 21:54:04] error: z2m: Exiting... [2025-12-08 21:54:04] error: z2m: Error: Failed to start EZSP layer with status=HOST_FATAL_ERROR. at EmberAdapter.initEzsp (/app/node_modules/.pnpm/[email protected]/node_modules/zigbee-herdsman/src/adapter/ember/adapter/emberAdapter.ts:666:19) at EmberAdapter.start (/app/node_modules/.pnpm/[email protected]/node_modules/zigbee-herdsman/src/adapter/ember/adapter/emberAdapter.ts:1539:24) at Controller.start (/app/node_modules/.pnpm/[email protected]/node_modules/zigbee-herdsman/src/controller/controller.ts:143:29) at Zigbee.start (/app/lib/zigbee.ts:70:27) at Controller.start (/app/lib/controller.ts:101:13) at start (/app/index.js:149:5)

Edit: This is the startup log after downgrading to 2.6.3.
[22:15:30] INFO: Preparing to start... [22:15:30] INFO: Socat not enabled [22:15:30] INFO: Starting Zigbee2MQTT... Starting Zigbee2MQTT without watchdog. [2025-12-08 22:15:31] info: z2m: Logging to console, file (filename: log.log) [2025-12-08 22:15:31] info: z2m: Starting Zigbee2MQTT version 2.6.3 (commit #unknown) [2025-12-08 22:15:31] info: z2m: Starting zigbee-herdsman (6.3.2) [2025-12-08 22:15:31] info: zh:adapter:discovery: Matched adapter: {"path":"/dev/ttyUSB0","manufacturer":"Nabu Casa","serialNumber":"5c69e68c7231ef1191c24fcfdfbc56eb","pnpId":"usb-Nabu_Casa_Home_Assistant_Connect_ZBT-1_5c69e68c7231ef1191c24fcfdfbc56eb-if00-port0","vendorId":"10c4","productId":"ea60"} => ember: path=/dev/ttyUSB0, score=2 [2025-12-08 22:15:31] info: zh:ember: Using default stack config. [2025-12-08 22:15:31] info: zh:ember: ======== Ember Adapter Starting ======== [2025-12-08 22:15:31] info: zh:ember:ezsp: ======== EZSP starting ======== [2025-12-08 22:15:31] info: zh:ember:uart:ash: ======== ASH Adapter reset ======== [2025-12-08 22:15:31] info: zh:ember:uart:ash: RTS/CTS config is off, enabling software flow control. [2025-12-08 22:15:31] info: zh:ember:uart:ash: Serial port opened [2025-12-08 22:15:31] info: zh:ember:uart:ash: ======== ASH starting ======== [2025-12-08 22:15:32] error: zh:ember:uart:ash: Received ERROR from adapter while connecting, with code=RESET_WATCHDOG. [2025-12-08 22:15:33] info: zh:ember:uart:ash: ======== ASH connected ======== [2025-12-08 22:15:33] info: zh:ember:uart:ash: ======== ASH started ======== [2025-12-08 22:15:33] info: zh:ember:ezsp: ======== EZSP started ======== [2025-12-08 22:15:33] info: zh:ember: Adapter EZSP protocol version (13) lower than Host. Switched. [2025-12-08 22:15:33] info: zh:ember: Adapter version info: {"ezsp":13,"revision":"7.4.4 [GA]","build":0,"major":7,"minor":4,"patch":4,"special":3,"type":170} [2025-12-08 22:15:34] info: zh:ember: [STACK STATUS] Network up. [2025-12-08 22:15:34] info: zh:ember: [INIT TC] Adapter network matches config. [2025-12-08 22:15:34] info: zh:ember: [CONCENTRATOR] Started source route discovery. 1248ms until next broadcast. [2025-12-08 22:15:34] info: z2m: zigbee-herdsman started (resumed) [2025-12-08 22:15:34] info: z2m: Coordinator firmware version: '{"meta":{"build":0,"ezsp":13,"major":7,"minor":4,"patch":4,"revision":"7.4.4 [GA]","special":3,"type":170},"type":"EmberZNet"}'
The two errors ("code=RESET_WATCHDOG", as well as "Adapter EZSP protocol version (13) lower than Host. Switched.") have been present for some time, and haven't seemed to created any issues before.

I'll also add that I (after creating this post) updated the stick to firmware 7.4.4.3 (from 7.4.4.0), and those errors continue.
I have not tried to update Z2M since that firmware update.

[Vaarinvaliliha]

I also had to downgrade core (back to 2025.12.0) and z2m (back to 2.6.3) to make devices work again. Dongle version is 7.4.4.0.

[IanKohout]

Gentlemen, still a problem.
Zigbee version 2.7.1-1.
Dongle firmware 8.0.2 [GA]

Log here:
[2025-12-08 04:14:06] error: zh:ember:uart:ash: Received ERROR from adapter, with code=ERROR_EXCEEDED_MAXIMUM_ACK_TIMEOUT_COUNT.
[2025-12-08 04:14:06] error: zh:ember:uart:ash: ASH disconnected | Adapter status: ASH_NCP_FATAL_ERROR
[2025-12-08 04:14:06] error: zh:ember:uart:ash: Error while parsing received frame, status=ASH_NCP_FATAL_ERROR.
[2025-12-08 04:14:06] error: zh:ember: Adapter fatal error: HOST_FATAL_ERROR
[2025-12-08 04:14:06] error: zh:ember:uart:ash: Error while parsing received frame, status=ASH_NCP_FATAL_ERROR.
[2025-12-08 04:14:06] error: zh:ember:uart:ash: Error while parsing received frame, status=ASH_NCP_FATAL_ERROR.
[2025-12-08 04:14:06] error: z2m: Adapter disconnected, stopping
[04:14:19] INFO: Preparing to start...
[04:14:19] INFO: Socat not enabled
[04:14:20] INFO: Starting Zigbee2MQTT...
Starting Zigbee2MQTT without watchdog.
[2025-12-09 04:14:40] error: zh:ember:uart:ash: Received ERROR from adapter, with code=ERROR_EXCEEDED_MAXIMUM_ACK_TIMEOUT_COUNT.
[2025-12-09 04:14:40] error: zh:ember:uart:ash: ASH disconnected | Adapter status: ASH_NCP_FATAL_ERROR
[2025-12-09 04:14:40] error: zh:ember:uart:ash: Error while parsing received frame, status=ASH_NCP_FATAL_ERROR.
[2025-12-09 04:14:40] error: zh:ember: Adapter fatal error: HOST_FATAL_ERROR
[2025-12-09 04:14:40] error: z2m: Adapter disconnected, stopping
[04:14:53] INFO: Preparing to start...
[04:14:53] INFO: Socat not enabled
[04:14:54] INFO: Starting Zigbee2MQTT...

....

[2025-12-10 04:13:39] error: zh:ember:uart:ash: Received ERROR from adapter, with code=ERROR_EXCEEDED_MAXIMUM_ACK_TIMEOUT_COUNT.
[2025-12-10 04:13:39] error: zh:ember:uart:ash: ASH disconnected | Adapter status: ASH_NCP_FATAL_ERROR
[2025-12-10 04:13:39] error: zh:ember:uart:ash: Error while parsing received frame, status=ASH_NCP_FATAL_ERROR.
[2025-12-10 04:13:39] error: zh:ember: Adapter fatal error: HOST_FATAL_ERROR
[2025-12-10 04:13:39] error: z2m: Adapter disconnected, stopping
[04:13:52] INFO: Preparing to start...
[04:13:52] INFO: Socat not enabled
[04:13:53] INFO: Starting Zigbee2MQTT...
Starting Zigbee2MQTT without watchdog.

as you can see, every night outage at the same time, restart of the add-on and start. Duration about 20 seconds. Once a night, at the same time. There is nothing else in the system at that moment. Server log is clean too...

[Testmangh]

With 2.7.0 my network became too unstable all devices went offline 4 times in two days, updated to 2.7.1 sometimes the network was fast sometimes super slow and devices kept going offline... Downgraded back to 2.6 now network ia stable But memory usage is constantly 25% before It was 0.6~0.8%
HA 2025.12.2

[MSL-DA]

Downgraded back to 2.6 now network ia stable But memory usage is constantly 25% before It was 0.6~0.8% HA 2025.12.2

This is an HA thing. 2025.12.0 and 2025.12.2 are among the worst updates in 2025

[IanKohout]

Yes, exactly... last version of HA is the worst... same problems... with Zigbee, with Asus integration...

[Testmangh]

I have been observing something interesting about my two separate network setups.

One network uses an SLZB-06 coordinator with the CC2652P chip, and the other uses an SLZB-06M coordinator with the EFR32MG21 chip.

The EFR32MG21-based network has consistently shown outstanding performance through every firmware update. The difference that initially caught my attention was the clock speed: 48 MHz (CC2652P) versus 80 MHz (EFR32MG21).

To make the EFR32MG21's performance even more notable, its network contains the most demanding devices (spammers), such as mmWave sensors, energy meters, and smart power plugs.

I recently updated the second network to version 2.7.1, and it remains as stable as it was previously: I've observed no offline devices and no complete network drops. This reliability has been consistent across all updates.

Due to this superior performance, I have decided to change my main network's coordinator to the SLZB-06-MG26 (or similar EFR32-based device).

[PeterChrHansen]

I just bought a Sonoff ihost hub 4gb. I installed HA on in but the zigbee2mqtt is greyed out. Guess its because of the 32bit vs 64bit. But is there no way to get it installed ?

[Koenkk]

Yes, HA dropped support for 32-bit (and so was the add-on forced to)