Pi-hole 6 Container Crashes on Login (Segmentation Fault in FTL)

[dymek37]

Details

After installing a clean version of Pi-hole 6 (without upgrading from version 5) and starting the container, the application runs. However, when attempting to log in via http://ip:8080/admin/login, the container shuts down. The logs contain an error related to generating the CLI password hash and errors related to the NTP server. Ultimately, the FTL process crashes with a Segmentation fault error.

How to reproduce the issue

1. Environment data

• Hardware: Synology NAS
• Kernel Architecture: amd64
• Docker Install Info and version:
• Software source: official docker-ce
• Supplementary Software: Portainer
• Hardware architecture: x86_64

2. Docker-compose.yml contents:

services:
  pihole:
    image: pihole/pihole
    container_name: Pi-Hole
    network_mode: host
    security_opt:
      - no-new-privileges:false
    restart: on-failure:5
    volumes:
      - /volume1/docker/pihole/dnsmasq.d:/etc/dnsmasq.d:rw
      - /volume1/docker/pihole/pihole:/etc/pihole:rw
    environment:
      FTLCONF_webserver_api_password: pass
      FTLCONF_webserver_port: 8080
      TZ: Europe/Warsaw
      DNSMASQ_USER: pihole 
      PIHOLE_UID: 1026
      PIHOLE_GID: 100
    cap_add:
      - SYS_TIME
      - SYS_NICE

3. Container logs:

2025-02-25 09:37:39.919 CET [53M] ERROR: Failed to generate CLI password hash!  
2025-02-25 09:37:40.081 CET [53/T113] INFO: Compiled 0 allow and 0 deny regex for 15 clients in 0.3 msec  
2025-02-25 09:39:38.882 CET [53/T218] INFO: ---------------------------->  FTL crashed!  <----------------------------  
2025-02-25 09:39:38.882 CET [53/T218] INFO: Received signal: Segmentation fault  
2025-02-25 09:39:38.883 CET [53/T218] INFO:      at address: 0  
2025-02-25 09:39:38.883 CET [53/T218] INFO:      with code:  SEGV_MAPERR  

[DL6ER]

Could you provide a few more lines from the container logs? Otherwise, it would be extremely helpful if you could follow the steps on https://docs.pi-hole.net/ftldns/gdb/ and provide the backtrace on this crash.

[dymek37]

Is this what you need?

(gdb) backtrace
#0  0x000000000042a8f2 in check_client_auth (api=api@entry=0x7f752ef584f0, is_api=is_api@entry=false) at /app/src/api/auth.c:79
#1  0x0000000000448268 in request_handler (conn=0x7f752f85bb20, cbdata=<optimized out>) at /app/src/webserver/lua_web.c:122
#2  0x00000000004651f4 in handle_request (conn=conn@entry=0x7f752f85bb20) at /app/src/webserver/civetweb/civetweb.c:15413
#3  0x0000000000467d99 in handle_request_stat_log (conn=0x7f752f85bb20) at /app/src/webserver/civetweb/civetweb.c:6831
#4  process_new_connection (conn=conn@entry=0x7f752f85bb20) at /app/src/webserver/civetweb/civetweb.c:19977
#5  0x00000000004698ef in worker_thread_run (conn=0x7f752f85bb20) at /app/src/webserver/civetweb/civetweb.c:20349
#6  worker_thread (thread_func_param=0x7f752f85bb20) at /app/src/webserver/civetweb/civetweb.c:20496
#7  0x000000000080946f in start (p=0x7f752ef5db00) at src/thread/pthread_create.c:207
#8  0x000000000080ab4e in __clone () at src/thread/x86_64/clone.s:22
Backtrace stopped: frame did not save the PC

[DL6ER]

Yes, thank you! Could you please check if the crash is gone for you after running

fix/randomness_crash

?

You may not be able to set a password as secure randomness seems to be unavailable on your system but we can get to that in a second step. It'd first be great if you could check if the crash is gone.

[DL6ER]

Duplicate of #2242, fixed by #2300 which will be part of the next release (FTL v6.0.4).