Container disappears after reboot.

If you're using compose files, you'll need to have the docker compose plugin installed.

What plugin is needed? Docker Compose Manager?

I am currently using Portainer-CE + Portainer-Agent and it has been working fine with a total of 6 compose files (Stacks).

I believe it is a problem when I set the network to br0 (bridge), and after restarting the container it simply disappears.

I've seen a few posts on the forum, but without a definitive solution, so I'm posting.

 

I installed Docker Compose Manager, but nothing changed. I get the same error after restarting the server.

 

Sep  7 21:16:03 Unraid0 rc.docker: pihole: Error response from daemon: network 33947c2bdd9f667c209d465520a46f09492af05774237a9f6b267bb18e03740d not found
Sep  7 21:16:03 Unraid0 rc.docker: Error: failed to start containers: pihole

 

Edited September 8, 2022 by AyslanAlves

If you're using Portainer to manage, install etc compose files then what happens if you use Portainer to make the changes?

 

Not saying that there's an issue per se, but it really simplifies things to use one method to manage the containers you install via that method and not another one (docker tab) that has no idea about how the containers were installed etc

1 hour ago, Squid said:

If you're using Portainer to manage, install etc compose files then what happens if you use Portainer to make the changes?

 

Not saying that there's an issue per se, but it really simplifies things to use one method to manage the containers you install via that method and not another one (docker tab) that has no idea about how the containers were installed etc

 

 

I've already tested using the unraid manager and portainer, you know?

It makes no difference, the error is practically the same, with a small variation depending on the manager. In practice it always fails to start the container.

I really believe it's a bug in unraid as I tested it in a VM in VirtualBox and it works normally after restarting.

There is some configuration in docker that erases the network interface and when it restarts it shows the following error in the unraid logs: Error response from daemon: network 33947c...

OK.  Post your diagnostics and one of the relevant xml's (from /config/plugins/dockerMan/templates-user) before it disappears and afterwards

3 hours ago, Squid said:

OK.  Post your diagnostics and one of the relevant xml's (from /config/plugins/dockerMan/templates-user) before it disappears and afterwards

 

Note that there is no template related to pihole, as it was installed via compose file, not by unraid app store.

 

If I set network_mode: "br0" and restart unraid, the container disappears and displays the error message "Error response from daemon: network".

 

If I don't set network_mode, it will create pihole_default and it will work, but I can't get an IP directly from the router and I'm stuck with the unraid IP.

 

Another point is that pihole uses port 53, which is also used if the virtual machines are powered on.

The central problem is to set the network_mode to get the ip directly on the router, acting as a bridge.

 

Currently I have disabled the virtual machines and I am using it without setting the network interface, as you can see in the image. That way it works fine, but I'm stuck on the docker IP (10.0.0.100) and I can't use the virtual machines at the same time.

 

Follows current diagnostics, and current compose file, with network_mode disabled:

version: "3"

# More info at https://github.com/pi-hole/docker-pi-hole/ and https://docs.pi-hole.net/
# Listas em https://firebog.net/
services:
  pihole:
    container_name: pihole
    image: pihole/pihole:latest
    # For DHCP it is recommended to remove these ports and instead add: network_mode: "host"
    ports:
      - "53:53/tcp"
      - "53:53/udp"
      - "8082:80/tcp"
    environment:
      TZ: 'America/Maceio'
      WEBPASSWORD: 'admin'
    #  FTLCONF_LOCAL_IPV4: '10.0.0.2'
      PIHOLE_DNS_: '8.8.8.8;8.8.4.4'
    # Volumes store your data between container upgrades
    volumes:
      - '/mnt/user/appdata/pihole/pihole:/etc/pihole'
      - '/mnt/user/appdata/pihole/dnsmasq.d:/etc/dnsmasq.d'    
    #   https://github.com/pi-hole/docker-pi-hole#note-on-capabilities
    #cap_add:
    #  - NET_ADMIN # Recommended but not required (DHCP needs NET_ADMIN)      
    #network_mode: "br0"
    restart: always

 

unraid0-diagnostics-20220908-2057.zip

OK. The only way I managed to make it work is to install directly from the apps tab.
Otherwise it was not possible.

Changed Status to Closed