New setup won't start on Portainer

[ShiftyMcCool]

I initially made this comment on another issue before I realized it wasn't exactly the same.

I set up a new stack in Portainer with the details below and the container log showed the copy of default files and they appear on my mounted volume so that seems good. However, despite dropping the stack and redeploying (once with v2.4.1) I get the same output:

PROXY: Starting...
rm: cannot remove '/etc/proxmox-backup/.*.lck': No such file or directory
rm: cannot remove '/etc/proxmox-backup/*.lock': No such file or directory
/etc/proxmox-backup is a mountpoint
/var/lib/proxmox-backup is a mountpoint
/var/log/proxmox-backup is a mountpoint
/run is a mountpoint
API: Starting...

I've let it sit like this for over 30 minutes without any movement. I'm guessing I'm doing something wrong, any ideas?

docker-compose.yml:

version: '2.1'

services:
  proxmox-backup:
    image: ayufan/proxmox-backup-server:latest
    network_mode: host
    mem_limit: 2G
    volumes:
      - backups:/backups
      - pbs_etc:/etc/proxmox-backup
      - pbs_logs:/var/log/proxmox-backup
      - pbs_lib:/var/lib/proxmox-backup
    environment:
      TZ: America/Detroit
    tmpfs:
      - /run
    cap_add:
      - SYS_RAWIO
    devices:
      - /dev/mmcblk0
      - /dev/sda
    restart: unless-stopped
    stop_signal: SIGHUP
    labels:
      - "traefik.enable=true"
      - "traefik.http.routers.proxmox-http.rule=Host(`proxmox.it-clowd`)"
      - "traefik.http.routers.proxmox-http.entrypoints=web"
      - "traefik.http.routers.proxmox-https.rule=Host(`proxmox.it-clowd`)"
      - "traefik.http.routers.proxmox-https.entrypoints=websecure"
      - "traefik.http.routers.proxmox-https.tls.certresolver=basic"
      - "traefik.http.services.proxmox-service.loadbalancer.server.port=8007"

volumes:
  backups:
    driver: local
    driver_opts:
      type: ''
      o: bind
      device: /media/data/proxmox/backups
  pbs_etc:
    driver: local
    driver_opts:
      type: ''
      o: bind
      device: /media/data/proxmox/etc
  pbs_logs:
    driver: local
    driver_opts:
      type: ''
      o: bind
      device: /media/data/proxmox/logs
  pbs_lib:
    driver: local
    driver_opts:
      type: ''
      o: bind
      device: /media/data/proxmox/lib

Let me know if you need any other details Thanks!

[ayufan]

I will take a look tomorrow.

On Sun, Dec 3, 2023 at 9:17 PM Ryan Harris @.***> wrote:

I initially made this comment on another issue before I realized it wasn't exactly the same.

I set up a new stack in Portainer with the details below and the container log showed the copy of default files and they appear on my mounted volume so that seems good. However, despite dropping the stack and redeploying (once with v2.4.1) I get the same output:

PROXY: Starting... rm: cannot remove '/etc/proxmox-backup/..lck': No such file or directory rm: cannot remove '/etc/proxmox-backup/.lock': No such file or directory /etc/proxmox-backup is a mountpoint /var/lib/proxmox-backup is a mountpoint /var/log/proxmox-backup is a mountpoint /run is a mountpoint API: Starting...

I've let it sit like this for over 30 minutes without any movement. I'm guessing I'm doing something wrong, any ideas? image.png (view on web) https://github.com/ayufan/pve-backup-server-dockerfiles/assets/1320624/b3e341a8-b285-4f92-8ba1-ffd71755feb4

docker-compose.yml:

version: '2.1'

services: proxmox-backup: image: ayufan/proxmox-backup-server:latest network_mode: host mem_limit: 2G volumes:

• backups:/backups
• pbs_etc:/etc/proxmox-backup
• pbs_logs:/var/log/proxmox-backup
• pbs_lib:/var/lib/proxmox-backup environment: TZ: America/Detroit tmpfs:
• /run cap_add:
• SYS_RAWIO devices:
• /dev/mmcblk0
• /dev/sda restart: unless-stopped stop_signal: SIGHUP labels:
• "traefik.enable=true"
• "traefik.http.routers.proxmox-http.rule=Host(proxmox.it-clowd)"
• "traefik.http.routers.proxmox-http.entrypoints=web"
• "traefik.http.routers.proxmox-https.rule=Host(proxmox.it-clowd)"
• "traefik.http.routers.proxmox-https.entrypoints=websecure"
• "traefik.http.routers.proxmox-https.tls.certresolver=basic"
• "traefik.http.services.proxmox-service.loadbalancer.server.port=8007"

volumes: backups: driver: local driver_opts: type: '' o: bind device: /media/data/proxmox/backups pbs_etc: driver: local driver_opts: type: '' o: bind device: /media/data/proxmox/etc pbs_logs: driver: local driver_opts: type: '' o: bind device: /media/data/proxmox/logs pbs_lib: driver: local driver_opts: type: '' o: bind device: /media/data/proxmox/lib

Let me know if you need any other details Thanks!

— Reply to this email directly, view it on GitHub https://github.com/ayufan/pve-backup-server-dockerfiles/issues/44, or unsubscribe https://github.com/notifications/unsubscribe-auth/AASOSQN6YUHY4ARAFSPCPU3YHTM5VAVCNFSM6AAAAABAFAUH76VHI2DSMVQWIX3LMV43ASLTON2WKOZSGAZDENRZG44TQNI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[SAOPP]

Interesting, the kernel is latest, but u r use aarch64 architecture... @ayufan wating fo ya suggestion homie :+1:

[SAOPP]

...and behaviour is the same as mine.

[ShiftyMcCool]

sorry, not trying to be pushy, but have you had a chance to look into this?

[ayufan]

@ShiftyMcCool

I tested this on vanilla docker, with recent enough kernel and it seems to work. Maybe you cannot access the container? The docker-compose uses net: host which might mess with firewall. You might prefer to change this to forward only 8007 port.

[tradenet]

I am having the same issues with my portainer config. Also, I've tried running this: docker run -d \ --name=pbs \ --network=host \ --tmpfs=/run \ -e TZ=Europe/Zurich \ -h pbs \ -v /path/to/data/backup:/mnt/backup \ -v /path/to/config/pbs/config/etc:/etc/proxmox-backup \ -v /path/to/config/pbs/config/log:/var/log/proxmox-backup \ -v /path/to/config/pbs/config/lib:/var/lib/proxmox-backup \ --restart unless-stopped \ ayufan/proxmox-backup-server:latest

With the same issues.

Docker version 24.0.7 on Debian buster

[MasterIO02]

Having the same issue here, using this startup command:

sudo docker run --name proxmox-backup-server --restart="unless-stopped" \
    -e TZ=Europe/Paris \
    --volume=/path/proxmox-backup-server/backups:/backups \
    --volume=/path/proxmox-backup-server/etc:/etc \
    --volume=/path/proxmox-backup-server/logs:/var/logs \
    --volume=/path/proxmox-backup-server/lib:/var/lib \
    --tmpfs=/run \
    -p 8007:8007 \
    -d ayufan/proxmox-backup-server:latest

Using Docker 25.0.0 with kernel 6.2.16-6-pve (Proxmox VE Kernel)

[MasterIO02]

Got it to run using this command from this repository: https://github.com/Tontonjo/docker/tree/main/pbs:

docker run -d \
  --name=pbs \
  --network=host \
  --tmpfs=/run \
  -e TZ=Europe/Zurich \
  -h pbs \
  -v /path/to/data/backup:/mnt/backup \
  -v /path/to/config/pbs/config/etc:/etc/proxmox-backup \
  -v /path/to/config/pbs/config/log:/var/log/proxmox-backup \
  -v /path/to/config/pbs/config/lib:/var/lib/proxmox-backup \
  --restart unless-stopped \
  ayufan/proxmox-backup-server:latest

maybe the folders bound to the container were not having the right permissions, or in my case it was maybe because I mounted the whole /etc, /var/log and /var/lib? No idea tbh, but it works ¯_(ツ)_/¯

[tradenet]

I tried that as well. Seems most of these folks are using some NAS box running something like Unraid. I think it's my version of Debian Buster and docker that could be the issue. Even though I am running like 20 containers on it. I scrapped the notion of running it in a docker container. Realized I had an old Windows box I wasn't using and spun it up on a Hyper V VM. It's actually behaving quite nicely. I even got backs up going on it from a remotely locate PVE build. Amazing and magical I'd say.