ArchiveTeam/warrior-dockerfile

Constant "stream copy error: reading from a closed fifo" messages when two Warrior containers are running

catharsis71 opened this issue · 1 comments

I have two Warrior containers so I can set them to work on different projects (I usually have one on Auto and the other on URL Team 2)

  warrior1:
    container_name: warrior1
    image: atdr.meo.ws/archiveteam/warrior-dockerfile:latest
    restart: unless-stopped
    hostname: warrior1
    ports:
      - 65535:8001
    labels:
      com.centurylinklabs.watchtower.enable: "true"
    environment:
      - DOWNLOADER=myname
      - HTTP_USERNAME=admin
      - HTTP_PASSWORD=********
      - SELECTED_PROJECT=auto
      - CONCURRENT_ITEMS=3
  warrior2:
    container_name: warrior2
    image: atdr.meo.ws/archiveteam/warrior-dockerfile:latest
    restart: unless-stopped
    hostname: warrior2
    ports:
      - 65534:8001
    labels:
      com.centurylinklabs.watchtower.enable: "true"
    environment:
      - DOWNLOADER=myname
      - HTTP_USERNAME=admin
      - HTTP_PASSWORD=********
      - SELECTED_PROJECT=auto
      - CONCURRENT_ITEMS=3

If warrior1 is running but warrior2 is stopped, all is good
If warrior2 is running but warrior1 is stopped, all is good

however if both are running at once I get constant Docker errors.

Aug 25 15:37:35 us1 docker.dockerd[139066]: time="2023-08-25T15:37:35.927488084-05:00" level=error msg="stream copy error: reading from a closed fifo"
Aug 25 15:37:41 us1 docker.dockerd[139066]: time="2023-08-25T15:37:41.230477508-05:00" level=error msg="stream copy error: reading from a closed fifo"
Aug 25 15:37:41 us1 docker.dockerd[139066]: time="2023-08-25T15:37:41.234078141-05:00" level=error msg="stream copy error: reading from a closed fifo"
Aug 25 15:39:23 us1 docker.dockerd[139066]: time="2023-08-25T15:39:23.223602004-05:00" level=error msg="stream copy error: reading from a closed fifo"
Aug 25 15:39:23 us1 docker.dockerd[139066]: time="2023-08-25T15:39:23.232893762-05:00" level=error msg="stream copy error: reading from a closed fifo"
Aug 25 15:39:23 us1 docker.dockerd[139066]: time="2023-08-25T15:39:23.335053217-05:00" level=error msg="stream copy error: reading from a closed fifo"
Aug 25 15:39:23 us1 docker.dockerd[139066]: time="2023-08-25T15:39:23.335732882-05:00" level=error msg="stream copy error: reading from a closed fifo"
Aug 25 15:39:31 us1 docker.dockerd[139066]: time="2023-08-25T15:39:31.924019609-05:00" level=error msg="stream copy error: reading from a closed fifo"
Aug 25 15:39:31 us1 docker.dockerd[139066]: time="2023-08-25T15:39:31.927519312-05:00" level=error msg="stream copy error: reading from a closed fifo"
Aug 25 15:41:05 us1 docker.dockerd[139066]: time="2023-08-25T15:41:05.722050335-05:00" level=error msg="stream copy error: reading from a closed fifo"

they were initially appearing in /var/log/syslog but I reconfigured syslog to dump them into a separate file to keep my syslog clean

other than the log messages, both containers appear to function properly

Disabling the healthcheck on both containers seems to get rid of the messages as well as hugely reducing lag on my server

the healthcheck seems weirdly intensive, more than a single-core VPS can handle