DHCP requests stop working

Help
1

Please follow the below template, it will help us to help you!

Expected Behaviour:

DHCP requests work 24/7

Actual Behaviour:

Hi,

I've a problem with DHCP not working and I have to restart pihole. It works for a few minutes / hours then crashes again.

My setup:

ISP --> cable modem > ASUS RT-AC68U as router running Asuswrt-Merlin 384.12 --> LAN --> pihole.
DHCP is turned off in the router, and enabled in pihole. Any idea where I can start troubleshooting? If I grep for the MAC of a device not getting the IP, I only get entries from dnsmasq-dhcp like DHCPDISCOVER, DHCPOFFER,
DHCPACK and DHCPREQUEST, nothing about the process dying.

Debug Token:

https://tricorder.pi-hole.net/lmy4szpfvh

2

Look in /var/log/pihole.log and /var/log/pihole-FTL.log for any errors.

3

Not finding anything obvious. I see some of these SERVFAIL from dnsmasq in pihole.log:

$ grep -A3 -B3 -i error /var/log/pihole.log | tail -n10
--
Jul  3 12:32:41 dnsmasq-dhcp[599]: DHCPREQUEST(eth0) 192.168.1.106 c0:f2:fb:b4:bb:63
Jul  3 12:32:43 dnsmasq-dhcp[599]: DHCPACK(eth0) 192.168.1.106 c0:f2:fb:b4:bb:63 iPhone
Jul  3 12:32:45 dnsmasq[599]: reply error is SERVFAIL
Jul  3 12:32:45 dnsmasq[599]: reply error is SERVFAIL
Jul  3 12:32:45 dnsmasq[599]: reply error is SERVFAIL
Jul  3 12:32:45 dnsmasq-dhcp[599]: DHCPREQUEST(eth0) 192.168.1.106 c0:f2:fb:b4:bb:63
Jul  3 12:32:47 dnsmasq-dhcp[599]: DHCPACK(eth0) 192.168.1.106 c0:f2:fb:b4:bb:63 iPhone

pihole-FTL.log below:

 $ grep -i "fail\|warn\|error"  /var/log/pihole-FTL.log* | grep -v "(0 errors)"
/var/log/pihole-FTL.log:[2019-07-03 02:15:01.777 2178] IPv4 telnet error: Interrupted system call (4)
/var/log/pihole-FTL.log.1:[2019-07-02 16:48:53.450 11640] IPv4 telnet error: Interrupted system call (4)

Finally:

$ sudo systemctl status --full --no-pager pihole-FTL.service
● pihole-FTL.service - LSB: pihole-FTL daemon
   Loaded: loaded (/etc/init.d/pihole-FTL; generated; vendor preset: enabled)
   Active: active (exited) since Wed 2019-07-03 12:39:42 UTC; 4h 52min ago
     Docs: man:systemd-sysv-generator(8)
  Process: 6069 ExecStop=/etc/init.d/pihole-FTL stop (code=exited, status=0/SUCCESS)
  Process: 6116 ExecStart=/etc/init.d/pihole-FTL start (code=exited, status=0/SUCCESS)
    Tasks: 0 (limit: 4915)
   CGroup: /system.slice/pihole-FTL.service

Jul 03 12:39:31 raspberrypi systemd[1]: Starting LSB: pihole-FTL daemon...
Jul 03 12:39:31 raspberrypi pihole-FTL[6116]: Not running
Jul 03 12:39:41 raspberrypi su[6170]: Successful su for pihole by root
Jul 03 12:39:41 raspberrypi su[6170]: + ??? root:pihole
Jul 03 12:39:41 raspberrypi su[6170]: pam_unix(su:session): session opened for user pihole by (uid=0)
Jul 03 12:39:42 raspberrypi pihole-FTL[6116]: FTL started!
Jul 03 12:39:42 raspberrypi systemd[1]: Started LSB: pihole-FTL daemon.

I'll do some more digging.

4

Updated token: https://tricorder.pi-hole.net/4gomf0t0jb

The only thing I can think of is pihole-FTL maxing out the CPU, causing DHCP requests to time out or not even being acknowledged at all.

How can I find the source of the pihole-FTL CPU usage? Is there a maximum recommended amount of blocked domains, for the Raspberry Pi's?

[2019-07-03 19:31:19.896 31100] /etc/pihole/gravity.list: parsed **1030489 domains** (took 8297.5 ms)

top - 20:38:10 up  8:32,  2 users,  load average: 0.40, 0.34, 0.36
Tasks: 104 total,   2 running,  57 sleeping,   0 stopped,   1 zombie
%Cpu(s):  5.1 us,  0.2 sy,  0.0 ni, 94.2 id,  0.4 wa,  0.0 hi,  0.0 si,  0.0 st
KiB Mem :   999036 total,   325448 free,   138104 used,   535484 buff/cache
KiB Swap:   102396 total,   102396 free,        0 used.   790876 avail Mem

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
31100 pihole    20   0  156912  85060   3648 R  94.1  8.5  18:42.07 **pihole-FTL**
 9168 pi        20   0    8292   3352   2760 R  17.6  0.3   0:00.04 top
    1 root      20   0   28120   6092   4892 S   0.0  0.6   0:07.39 systemd
5

I believe a DNS loop can cause high CPU load.
Make sure DNS queries dont get looped back to Pi-hole.
One such example is if you configure Pi-hole to use its own IP address for upstream DNS resolution.

6

That makes sense, but pihole is only using external DNS resolvers. The high CPU load is not constant, only seen every now and then.

7

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.