V2.32.5: deconz.service: Main process exited, code=killed, status=11/SEGV

I updated to v2.32.5 on 2026-02-06. Today I had 3 (until now) crashes(?)

Feb 09 05:51:29 deconzpi systemd[1]: deconz.service: Main process exited, code=killed, status=11/SEGV
Feb 09 05:51:29 deconzpi systemd[1]: deconz.service: Failed with result 'signal'.
Feb 09 05:51:29 deconzpi systemd[1]: deconz.service: Consumed 2h 6min 8.467s CPU time.
Feb 09 05:51:59 deconzpi systemd[1]: deconz.service: Scheduled restart job, restart counter is at 1.
Feb 09 05:51:59 deconzpi systemd[1]: Stopped deconz.service - deCONZ: ZigBee gateway -- REST API.
Feb 09 05:51:59 deconzpi systemd[1]: deconz.service: Consumed 2h 6min 8.467s CPU time.
Feb 09 05:51:59 deconzpi systemd[1]: Started deconz.service - deCONZ: ZigBee gateway -- REST API.
Feb 09 05:51:59 deconzpi deCONZ[10631]: QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-pi'
Feb 09 05:51:59 deconzpi deCONZ[10631]: libpng warning: iCCP: known incorrect sRGB profile
Feb 09 05:51:59 deconzpi deCONZ[10631]: This plugin does not support propagateSizeHints()
Feb 09 05:51:59 deconzpi deCONZ[10631]: This plugin does not support propagateSizeHints()
Feb 09 05:51:59 deconzpi deCONZ[10631]: This plugin does not support propagateSizeHints()
Feb 09 05:52:03 deconzpi deCONZ[10631]: This plugin does not support propagateSizeHints()
Feb 09 07:13:51 deconzpi systemd[1]: deconz.service: Main process exited, code=killed, status=11/SEGV
Feb 09 07:13:51 deconzpi systemd[1]: deconz.service: Failed with result 'signal'.
Feb 09 07:13:51 deconzpi systemd[1]: deconz.service: Consumed 2min 31.993s CPU time.
Feb 09 07:14:22 deconzpi systemd[1]: deconz.service: Scheduled restart job, restart counter is at 2.
Feb 09 07:14:22 deconzpi systemd[1]: Stopped deconz.service - deCONZ: ZigBee gateway -- REST API.
Feb 09 07:14:22 deconzpi systemd[1]: deconz.service: Consumed 2min 31.993s CPU time.
Feb 09 07:14:22 deconzpi systemd[1]: Started deconz.service - deCONZ: ZigBee gateway -- REST API.
Feb 09 07:14:22 deconzpi deCONZ[11574]: QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-pi'
Feb 09 07:14:22 deconzpi deCONZ[11574]: libpng warning: iCCP: known incorrect sRGB profile
Feb 09 07:14:22 deconzpi deCONZ[11574]: This plugin does not support propagateSizeHints()
Feb 09 07:14:22 deconzpi deCONZ[11574]: This plugin does not support propagateSizeHints()
Feb 09 07:14:22 deconzpi deCONZ[11574]: This plugin does not support propagateSizeHints()
Feb 09 07:14:25 deconzpi deCONZ[11574]: This plugin does not support propagateSizeHints()
Feb 09 15:23:14 deconzpi systemd[1]: deconz.service: Main process exited, code=killed, status=11/SEGV
Feb 09 15:23:14 deconzpi systemd[1]: deconz.service: Failed with result 'signal'.
Feb 09 15:23:14 deconzpi systemd[1]: deconz.service: Consumed 14min 40.709s CPU time.
Feb 09 15:23:44 deconzpi systemd[1]: deconz.service: Scheduled restart job, restart counter is at 3.
Feb 09 15:23:44 deconzpi systemd[1]: Stopped deconz.service - deCONZ: ZigBee gateway -- REST API.
Feb 09 15:23:44 deconzpi systemd[1]: deconz.service: Consumed 14min 40.709s CPU time.
Feb 09 15:23:44 deconzpi systemd[1]: Started deconz.service - deCONZ: ZigBee gateway -- REST API.
Feb 09 15:23:44 deconzpi deCONZ[12168]: QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-pi'
Feb 09 15:23:44 deconzpi deCONZ[12168]: libpng warning: iCCP: known incorrect sRGB profile
Feb 09 15:23:44 deconzpi deCONZ[12168]: This plugin does not support propagateSizeHints()
Feb 09 15:23:44 deconzpi deCONZ[12168]: This plugin does not support propagateSizeHints()
Feb 09 15:23:44 deconzpi deCONZ[12168]: This plugin does not support propagateSizeHints()
Feb 09 15:23:48 deconzpi deCONZ[12168]: This plugin does not support propagateSizeHints()

Is this a new bug?

Maybe this is coincidence - but I restarted Home Assistant 3 times today. And the last time at 15:19:08 so just a few moments before the crash happened

An other user have it on the version 2.32.4, I put his log for memory

Thread 1 "deCONZ" received signal SIGSEGV, Segmentation fault.
0x0000fffff64cd204 in ?? () from /lib/aarch64-linux-gnu/libc.so.6
(gdb) bt
#0  0x0000fffff64cd204 in ?? () from /lib/aarch64-linux-gnu/libc.so.6
#1  0x0000ffffe90de19c in DeRestPluginPrivate::openClientTimerFired() () from /usr/share/deCONZ/plugins/libde_rest_plugin.so
#2  0x0000ffffe8ff1dec in DeRestPluginPrivate::qt_static_metacall(QObject*, QMetaObject::Call, int, void**) () from /usr/share/deCONZ/plugins/libde_rest_plugin.so
#3  0x0000fffff6be6628 in ?? () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#4  0x0000fffff6beac40 in QTimer::timeout(QTimer::QPrivateSignal) () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#5  0x0000fffff6bda5e4 in QObject::event(QEvent*) () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#6  0x0000fffff7a4c0a0 in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /lib/aarch64-linux-gnu/libQt5Widgets.so.5
#7  0x0000fffff6ba9d60 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#8  0x0000fffff6c08d80 in QTimerInfoList::activateTimers() () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#9  0x0000fffff6c0972c in ?? () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#10 0x0000fffff5896c34 in g_main_context_dispatch () from /lib/aarch64-linux-gnu/libglib-2.0.so.0
#11 0x0000fffff5896ed8 in ?? () from /lib/aarch64-linux-gnu/libglib-2.0.so.0
#12 0x0000fffff5896f84 in g_main_context_iteration () from /lib/aarch64-linux-gnu/libglib-2.0.so.0
#13 0x0000fffff6c09ac8 in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#14 0x0000fffff6ba8278 in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#15 0x0000fffff6bb12fc in QCoreApplication::exec() () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#16 0x0000aaaaaaae1754 in main ()
(gdb)

It’s stable until I restart Home Assistant… Restarting Homebridge has no effect.

I have the same, really frustrating :frowning:
Issue started with version V2_32_01 - 14-Nov-2025 02:30 20388122

So you mean deconz alone is working but if you start HA, deconz is crashing ?
So it happen when a third app use the API or try to read the websocket ?

No, it works with Home Assistant running. But when I stop it and start it again deCONZ crashes. Restarting homebridge does not crash deCONZ

Ha ? So I don’t see a link …
BTW there is an issue here fd leak with https_port enabled (sockets to :80 in CLOSE_WAIT) · Issue #8494 · dresden-elektronik/deconz-rest-plugin · GitHub
The user have problem if he use http and https port in same time

I don’t know which Port Home Assistant is using. The Gateway was auto discovered

Your system sounds rather unconventional, and you haven’t even specified your operating system. From the context, I assume you’re using three Docker containers: one for Home Assistant, one for Homebridge, and one for deCONZ. If your system was truly stable at one point, then you can only try the current deCONZ Docker container with HTTPS disabled.

I’m using Home Assistant Core in a FreeBSD jail and deCONZ on a dedicated Raspberry Pi 4 Model B Rev 1.5 with Debian GNU/Linux 12 (bookworm).

I wanted to try it again this morning. Opened deCONZ GUI, enabled ERROR and ERROR_L2.

Stopped Home Assistant

[homeassistant@homeassistant ~]$ date
Thu Feb 12 07:01:24 CET 2026

Started Home Assistant again.

07:01:48:989 bool parseXiaomiSpecial(Resource*, ResourceItem*, const deCONZ::ApsDataIndication&, const deCONZ::ZclFrame&, const QVariant&),1240: assertion 'at == 0xff01 || at == 0xff02 || a                              t == 0x00f7' failed
07:01:48:992 bool parseXiaomiSpecial(Resource*, ResourceItem*, const deCONZ::ApsDataIndication&, const deCONZ::ZclFrame&, const QVariant&),1240: assertion 'at == 0xff01 || at == 0xff02 || a                              t == 0x00f7' failed
Segmentation fault
$ date
Thu Feb 12 07:06:09 CET 2026

So - again - 5 minutes after I restarted Home Assistant deCONZ crashes. This is reproducible. Home Assistant and deCONZ running on 2 separates machines.

Whatever setup I have, deCONZ should not crash…

I’m trying to get on the bottom of it but can’t replicate so far. Can you produce a coredump when the crash happens? That may help to check in the debugger what’s going on.

I enabled core dumps but now I can’t reproduce it anymore… maybe it will come back after a few days of uptime? I will check again tomorrow

So, after 19 hours of uptime of the raspberry pi I restarted Home Assistant. A few moments later deCONZ crashed:

(gdb) bt
#0  __memcpy_generic () at ../sysdeps/aarch64/multiarch/../memcpy.S:169
#1  0x0000007f759ee19c in DeRestPluginPrivate::openClientTimerFired() () from /usr/share/deCONZ/plugins/libde_rest_plugin.so
#2  0x0000007f75901dec in DeRestPluginPrivate::qt_static_metacall(QObject*, QMetaObject::Call, int, void**) () from /usr/share/deCONZ/plugins/libde_rest_plugin.so
#3  0x0000007f823e6628 in ?? () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#4  0x0000007f823eac40 in QTimer::timeout(QTimer::QPrivateSignal) () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#5  0x0000007f823da5e4 in QObject::event(QEvent*) () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#6  0x0000007f8324c0a0 in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /lib/aarch64-linux-gnu/libQt5Widgets.so.5
#7  0x0000007f823a9d60 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#8  0x0000007f82408d80 in QTimerInfoList::activateTimers() () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#9  0x0000007f8240972c in ?? () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#10 0x0000007f81096c34 in g_main_context_dispatch () from /lib/aarch64-linux-gnu/libglib-2.0.so.0
#11 0x0000007f81096ed8 in ?? () from /lib/aarch64-linux-gnu/libglib-2.0.so.0
#12 0x0000007f81096f84 in g_main_context_iteration () from /lib/aarch64-linux-gnu/libglib-2.0.so.0
#13 0x0000007f82409ac8 in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#14 0x0000007f823a8278 in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#15 0x0000007f823b12fc in QCoreApplication::exec() () from /lib/aarch64-linux-gnu/libQt5Core.so.5
#16 0x000000555fc51754 in main ()
(gdb)

I discovered that reloading Deconz integration also triggers a crash, it takes a 5-10 min and deconz is crashing. :frowning:

deCONZ 2.32.05
Windows 10 x64
Crash triggered by Home Assistant restart
Exception: 0xc0000005
Faulting module: libde_rest_plugin.dll

Still reproducable. Anything I can do to help fixing this?

Can, you try to swap the used protocol ? for exemple use the http/ws if you use https/wss or the reverse ?

In don’t really know how. But shouldnt it be impossible to crash deconz remotely?

same for me after update to 2.32.5 deconz crashed. i can’t see any dependency to HA or similiar.
How can i downgrade?

will there be any solution?