aostanin/dockerfiles

Just a bunch of Dockerfiles

UNMAINTAINED

This repository is no longer maintained. Some of the Dockerfiles here are now maintained in their own repository.

dockerfiles's Issues

Update Btsync

Please rerun a build so we can get the latest version of btsync.

NZBGet could use Python and unzip support

I'm currently evaluating NZBGet on my Synology. Now that I'm playing around with extension scripts, I found that many of them are written in Python (2.x) which isn't installed in this Dockerfile. Also, I can't even write a simple script to unzip zipped NZBs because there's no unzip or a full 7za installation (7zr/p7zip doesn't support zip files.).

Plex docker container needs to specify br0 at start

docker run -d -n=false \
...

should probably read

docker run -d -b br0 -n=false \
...

Subsonic runs as root

Plex-media-server removed?

Commit cdf9ad5...cc60122 seems to have removed your plex files.

Just wondering if this is intentional?

What is the best way to handle upgrades to the btsync docker image?

What is the best way to upgrade btsync when an update comes out? Upgrade btsync within the docker container? Rebuild the container?

Deluge (at least) can't be a managed service because of broken signal handling

The start.sh script for (at least) deluge doesn't use exec to spawn its process. This is a Dockerfile best practice for custom entrypoints because it means the docker run invocation will correctly forward signals to the underlying command.

The result is that, for example, running this container as a systemd service, and issuing systemctl stop, does not behave correctly. The TERM sent by systemd is ignored and the container does not stop.

This may not be as easy as prefixing the deluge-web command with exec since the entrypoint also starts the deluge daemon. It may be required to bring in something like supervisord to run the daemon and web clients under a parent process, which can either be the entrypoint itself or be (the only thing) invoked in the entrypoint with exec.

Here's a more specific bug report about how this plays out under systemd:

Setup

Have the following systemd unit definition loaded as docker-deluge.service

[Unit]
Description=Deluge
Requires=docker.service
After=docker.service

[Service]
TimeoutStartSec=0
Restart=always
ExecStartPre=-/usr/bin/docker stop deluge
ExecStartPre=-/usr/bin/docker rm deluge
ExecStartPre=/usr/bin/docker pull aostanin/deluge
ExecStart=/usr/bin/docker run --name deluge \
  --publish 8112:8112 \
  --publish 58846:58846 \
  --volume /mnt/media:/media \
  --volume /data/deluge:/data \
  aostanin/deluge

[Install]
WantedBy=multi-user.target

Steps

systemctl start docker-deluge
systemctl stop docker-deluge

Expected result

The service and container should stop.

Actual result

The systemctl stop takes a long time before returning. The docker container is still running and can only be stopped with docker kill (or docker stop because it eventually times out and issues kill's behavior).