LM CPU/IO high load without any scripting

[Hadeel]

dear admin,

We are encoutering the situation where we disabled all of the event/resident script 
but still CPU/IO is over 1.00.
Is there any way we can try to detect the high trafic cause?

Thank you in advance!

[admin]

Install System load app and see what process is consuming all CPU.

[Daniel]

Disable KNX IP and see if this helps.

[AZ4L88]

Install System load app and see what process is consuming all CPU.

we did and found that this part using almost 100% of the CPU 
1148  avahi-daemon: running [LogicMachine.local]

---

Disable KNX IP and see if this helps.

this not worked

---

Install System load app and see what process is consuming all CPU.

System load
PID    Process    CPU    MEM
1  /sbin/init  0.00%  0.61MB
714 /sbin/syslogd-C16  0.00%  0.65MB
716 /sbin/klogd 0.00%  0.62MB
718 /sbin/hotplug2--override --persistent --set-rules-file /etc/hotplug2.rules --set-coldplug-cmd /sbin/udevtrigger --max-children 1    0.00%  0.66MB
912 /sbin/watchdog-t 5 /dev/watchdog    0.00%  0.59MB
979 /usr/sbin/gpiod-l /lib/restore/defaults.sh -d /lib/restore/restore.sh -b 9  0.00%  0.30MB
990 dropbear-R -p 22 -K 300 0.00%  0.60MB
994 /usr/sbin/ntpd-n -p 0.europe.pool.ntp.org -p 1.europe.pool.ntp.org -p 2.europe.pool.ntp.org -p 3.europe.pool.ntp.org    0.00%  0.67MB
1003    /usr/sbin/redis-server/etc/redis.conf  0.00%  1.18MB
1064    lua/lib/genohm-scada/core/groupmonitor.lua  0.00%  1.88MB
1074    lua/lib/genohm-scada/core/scenes.lua    0.00%  1.86MB
1075    lua/lib/genohm-scada/core/ipblocker.lua 0.00%  0.93MB
1077    /usr/bin/eibd-e 1.1.1 -q 100 -F n,n,n,n,0,0 tpuarts:/dev/ttymxc5    0.00%  1.03MB
1081    lua/lib/genohm-scada/plugins/modbus/daemon.lua 0    0.00%  1.96MB
1092    lua/lib/apps/daemon.lua canx /home/apps/store/daemon/canx/daemon.lua    0.00%  1.99MB
1094    lua/lib/apps/daemon.lua lmcloud /home/apps/store/daemon/lmcloud/daemon.lua  0.00%  2.13MB
1114    nginx:master process nginx -c /tmp/nginx.conf  0.00%  1.07MB
1116    nginx:worker process    0.70%  3.74MB
1122    /usr/sbin/crond-l 20 -c /etc/crontabs  0.00%  0.60MB
1128    /usr/bin/dbus-daemon--system    0.00%  0.78MB
1147    avahi-daemon:running [LogicMachine.local]  96.84%  1.07MB
1161    /usr/sbin/zerotier-one-p9993 /var/lib/zerotier-one_default  0.18%  5.88MB
1172    avahi-browse-l _http._tcp  0.00%  0.81MB
system  0.35%
total  98.07%  31.59MB

[admin]

Which firmware version do you have? Have you tried rebooting LM?

[AZ4L88]

Which firmware version do you have? Have you tried rebooting LM?

HW: LM5(i.MX6)
SW: 20230612

we did reboot and factory reset and still same probleam.
we have this probleam in 5 LM now.

---

Which firmware version do you have? Have you tried rebooting LM?

by the way, this problem occurs only when we connect LLM with internet. however when connect it directly with the PC by LAN no issues.

[admin]

Can you send access to zerotier via PM?

[Hadeel]

Thank you ! I sent you zerotier access information via PM (:

[AZ4L88]

Can you send access to zerotier via PM?

Hello Admin have you received the Zerotier access data by PM? my partner sent it to you yesterday.

[admin]

Accept my ZeroTier access.

[AZ4L88]

Accept my ZeroTier access.

access accepted.

[admin]

This can be caused by a multicast loop. I've made some config changes, see if it helps.

[AZ4L88]

This can be caused by a multicast loop. I've made some config changes, see if it helps.

Thanks admin
the changes you did looks like fixed the issue and the CPU usage now is 3%.
we have an other 4 LMs need to be fixed.
can you please let us know who we can fix this issue?
and is this relted to the new LM firmware?
do we need to do these changes from now and on?

[admin]

Can you explain the network setup? Do you have multiple LMs on the same network and also connected to the same ZeroTier network?

To fix this, run this script once on each LM:

Code:

conf = [[
[server]
use-ipv4=yes
allow-interfaces=eth0
use-ipv6=no
check-response-ttl=no
use-iff-running=no

[publish]
publish-addresses=yes
publish-hinfo=yes
publish-workstation=no
publish-domain=yes

[reflector]
enable-reflector=no
reflect-ipv=no

[rlimits]
rlimit-core=0
rlimit-data=4194304
rlimit-fsize=0
rlimit-nofile=30
rlimit-stack=4194304
rlimit-nproc=3
]]

io.writefile('/etc/avahi/avahi-daemon.conf', conf)
os.execute('/etc/init.d/avahi-daemon restart')

[AZ4L88]

Can you explain the network setup? Do you have multiple LMs on the same network and also connected to the same ZeroTier network?
To fix this, run this script once on each LM:

Code:

conf = [[
[server]
use-ipv4=yes
allow-interfaces=eth0
use-ipv6=no
check-response-ttl=no
use-iff-running=no
[publish]
publish-addresses=yes
publish-hinfo=yes
publish-workstation=no
publish-domain=yes
[reflector]
enable-reflector=no
reflect-ipv=no
[rlimits]
rlimit-core=0
rlimit-data=4194304
rlimit-fsize=0
rlimit-nofile=30
rlimit-stack=4194304
rlimit-nproc=3
]]
io.writefile('/etc/avahi/avahi-daemon.conf', conf)
os.execute('/etc/init.d/avahi-daemon restart')

thanks admin
i will try this tomorrow at the site.
we have 5 villas built in the same area. each one have its own network and LM.
each LM have difrent Zerotier network since they are in different Network

[Hadeel]

Dear admin,

Thank you for the guidance!
We executed the code you provided and it seemed the CPU state gets better
but when the LM is restarted, the CPU state gets over 1.00 again.

I attach the image of the sysload of the LM you executed the script through zerotier.
The avahi-daemon process uses 95% again.

Do you think it is better for us to execute the code you provided every time the LM is restarted?

[admin]

The script should only be run once. Looks like something is wrong in your network setup.

Try running script with this modified configuration:

Code:

conf = [[
[server]
use-ipv4=yes
allow-interfaces=eth0
use-ipv6=no
check-response-ttl=yes
use-iff-running=no
[publish]
publish-addresses=yes
publish-hinfo=yes
publish-workstation=no
publish-domain=yes
[reflector]
enable-reflector=no
reflect-ipv=no
[rlimits]
rlimit-core=0
rlimit-data=4194304
rlimit-fsize=0
rlimit-nofile=30
rlimit-stack=4194304
rlimit-nproc=3
]]
io.writefile('/etc/avahi/avahi-daemon.conf', conf)
os.execute('/etc/init.d/avahi-daemon restart')

If this does not help then run Wireshark on the same network when this issue happens. Set filter to mdns and check what's going on.