Mass kresd restarts in log caused by adblock

Software SW help
1

Hi all

In my /var/log/messages I see lots of log entries like the following. Up to 10 times per 1h.

procd: Instance kresd::instance1 pid 26292 not stopped on SIGTERM, sending SIGKILL instead

Does anybody have the same problem? What can be the reason for this restarts?

Device Turris Omnia
reForis version 1.4.1
Turris OS version 6.4.2
Turris OS branch HBS
Kernel version 5.15.127

Many thanks for your help.
Markus

2

I have the same versions and no error in /var/log/messages.
What is your uptime and free memory df -h?

3 
turris:~# df -h
Filesystem                Size      Used Available Use% Mounted on
/dev/mmcblk0p1            7.3G    527.5M      6.8G   7% /
devtmpfs                512.0K         0    512.0K   0% /dev
tmpfs                  1008.1M     56.0M    952.0M   6% /tmp
tmpfs                   512.0K         0    512.0K   0% /dev
/dev/sda                  7.5G     10.1M      7.0G   0% /srv
turris:~# uptime
 21:23:11 up 34 days,  9:16,  load average: 0.11, 0.35, 0.40
4

And free -h?
Problem appears even after restart or only after some hours/days?
Load average seems to be very high. Can you check usage using htop?

5 
turris:~# free -h
              total        used        free      shared  buff/cache   available
Mem:        2064516      335716      175108       57360     1553692     1534672
Swap:             0           0           0

htop looks good. Problem appears even after restart.

6

For comparison my status from htop (I have Omnia with 1 GB RAM):

  • both CPU usage less than 5% (except some spikes to 15% for 1 second)
  • load average less than 0.1
  • 50 tasks, only 1 running

Biggest CPU usage (less than 3%) has process “python3 /usr/bin/foris-controller”.

image
image1007×208 26.6 KB

7

Thanks for your hel so far, Skippi. Much appreciated. htop looks similar on my side.

I rebooted the turris yesterday evening and analyzed the logs. Some hours after turris reboot i see no kresd restarts. Instead I see lots of following messages. I do not know is this is normal.

dhcp_host_domain_ng.py: Refresh kresd leases

Problems begin 4 hours after restart, after adblock gets started the first time. (Don’t ask me why it takes 4 hours until adblock starts first time after turris reboot). I see issues loading blocklists.

log: curl: (6) Could not resolve host: raw.githubusercontent.com

Nevertheless adblock starts successfully.

adblock-4.1.3[12603]: blocklist with overall 440604 blocked domains loaded successfully

After starting adblock the mass kresd restarts begin…

procd: Instance kresd::instance1 pid 16664 not stopped on SIGTERM, sending SIGKILL instead
8

I disabled adblock and everything works fine now. No kresd restarts for hours. Somehow adblock (or something in my adblock config) causes kresd to restart.

Any ideas? Does anybody have the same problem?

9

I thought that’s how adblock causes reload of the rules updated in RPZ file. But I don’t remember.