Container cheat sheet

Lions contend that most containers are only used for downloading torrents in a VPN.  Most ISP's these days disable your account for nefarious downloading.

The mane VPN program is openvpn.  The mane command line torrent program these days is transmission-cli. 

apt install openvpn transmission-daemon transmission-cli

Getting it going in a VPN container is a long & hard process.

The daemon is a systemd service.  It immediately gives "unauthorized" for all commands.  You have to systemctl stop transmission-daemon, edit /etc/transmission-daemon/settings.json, set rpc-authentication-required to false, & systemctl start transmission-daemon to get around this.

Another new dance is the download-queue-enabled, queue-stalled-enabled  options have to be set to false or it'll set every new torrent to queued while waiting forever for the unseeded torrents.

Generally, there's an openvpn command which configures the VPN.  It runs a command after the VPN starts & another command before the VPN dies, to ensure no data intended for the VPN goes to the insecure network.

openvpn --script-security 2 --config [.ovpn file] --auth-user-pass [userpass file] --comp-lzo --up-delay --up [startup script] --down-pre --down [shutdown script]

For some reason, the --comp-lzo option may have to be taken out if it can't access anything from inside the VPN.

The VPN nameserver is contained in a foreign_option_1 environment variable passed to the startup script.

At minimum, the startup & shutdown scripts have to manage the torrent daemon & set resolv.conf.

startup script:

#!/bin/sh

systemctl start transmission-daemon

cat > /etc/resolv.conf << EOF
nameserver THE_VPN_NAMESERVER
EOF

shutdown script:

#!/bin/sh

systemctl stop transmission-daemon

cat > /etc/resolv.conf << EOF
nameserver THE_ISP_NAMESERVER
EOF

All the transmission downloads go in /var/lib/transmission-daemon/downloads.  The lion kingdom made this a mount point pointing to a host directory in the lxc config.

lxc.mount.entry = /home/mov/sin /var/lib/lxc/sin/rootfs/var/lib/transmission-daemon/downloads none bind 0 0

The permission has to be 777 since transmission-daemon runs as a normal user.

The torrents are all in /var/lib/transmission-daemon/.config/transmission-daemon/torrents & /var/lib/transmission-daemon/.config/transmission-daemon/resume

 New containers have to be routinely created as VPN's migrate to new ubuntu releases.  To transfer all the torrents between containers, you have to copy the 2 torrent directories, make sure the permissions are 777, & the user exists. 

-------------------------------------------------------------------------------------------------------------------------

Key commands:

Start downloading a torrent:

transmission-remote -a "magnet link"

List status & ID's of all torrents:

transmission-remote -l

Stop a torrent by ID:

transmission-remote -t [ID] -S

Resume a torrent by ID:

transmission-remote -t [ID] -s

Remove a torrent by ID:

transmission-remote -t [ID] -r

There is no easy way to select individual files for downloading.  The general idea is to poll the torrent contents by torrent ID.

transmission-remote -t [ID] -f

Once it downloads the file list, stop all the files from downloading

transmission-remote -t [ID] -G all

Then resume 1 file at a time by passing its ID

transmission-remote -t [ID] -g [file ID]

Noted every time you lxc-stop & lxc-start a certain container, it creates a new virtual eth device.  Eventually ifconfig & brctl show fill with dead virtual eth devices.  These dead devices cause networking in the container to fail with TLS errors, dropped connections, while DNS & ping continue to work. 

The leading theory is these containers start processes which aren't killed by lxc-stop.  kill -9 instead of lxc-stop doesn't release the devices. 

meld shows an interesting evolution of the process table during the lifecycle of the container.

The container creates a bunch of nfsd & kworker which aren't visible inside the container.  These are left behind after lxc-stop or kill -9

killall -9 nfsd gets rid of all the dangling eth devices & processes.

The nfsds are all created outside of the container by lxc-start & bound to the virtual eth.

For some reason, the offending container has a bunch of rpc programs running inside the container & these cause nfsd to run outside the container.  There's no reason to use nfsd in a container since the filesystem is a subdirectory on the host.  A quick disabling of RPC solves the dangling network interfaces.

mv /usr/sbin/rpc.idmapd /usr/sbin/rpc.idmapd.bak

mv /usr/sbin/rpc.mountd /usr/sbin/rpc.mountd.bak

mv /usr/sbin/rpcbind /usr/sbin/rpcbind.bak

That fixed the connection errors.  As in all things, containers introduce many new problems in exchange for solving old problems with virtual machines.  The hope is the new problems are less debilitating than the old problems but there can never be a perfect solution.

The goog spits out many hits for running an X11 program in a container by creating a new container from scratch.  Buried in this list was 1 hit for reconfiguring an existing container.

https://blog.simos.info/running-x11-software-in-lxd-containers/

The general idea is to add 1 line to your /var/lib/lxc/*/config file

lxc.mount.entry = /tmp/.X11-unix tmp/.X11-unix none bind,optional,create=dir,ro

Then restart the container.

This creates a path which redirects the X server sockets of display :0 to the X server sockets of the host display.