Inactive Timeout Question

[ryancasler]

I want to start by saying thanks so much for this! It has made it super easy for those of us with "more than the average Joe but nowhere even close to expert level" knowledge to set-up. And all of the documentation is amazing and helped me work through some issues when I was first implementing about 6 months ago. Thank you so much!

I have run into a bit of an issue and I've hit a bit of a wall trying to figure out 1) if it needs to be addressed and 2) how to go about figuring out what needs to be addressed.

I recently moved my configuration to a new PC and it is running on a Ubuntu server virtualbox machine under a Windows 10 host. My VPN provider is CyberGhost VPN. I don't believe I was experiencing this same issue on my old setup, which was running with a previous version of the docker image. My docker run command is as follows:

docker run --cap-add=NET_ADMIN -d\
              -v /data/downloads/:/data \
              -v /data/vpn/config/:/etc/openvpn/custom/ \
              -v /data/vpn/scripts:/scripts \
              -e DISABLE_PORT_UPDATER=true \
              -e OPENVPN_PROVIDER=CUSTOM \
              -e OPENVPN_USERNAME=REDACTED\
              -e OPENVPN_PASSWORD=REDACTED\
              -e LOCAL_NETWORK=192.168.1.0/24 \
              -e TRANSMISSION_RATIO_LIMIT=0 \
              -e TRANSMISSION_RATIO_LIMIT_ENABLED=true \
              -e TRANSMISSION_SCRIPT_TORRENT_DONE_ENABLED=true  \
              -e TRANSMISSION_SCRIPT_TORRENT_DONE_FILENAME=/scripts/complete.sh \
              --restart=always \
              --name transmission \
              -p 9091:9091 \
              -e TZ=America/New_York \
              -e "OPENVPN_OPTS=--inactive 3600 --ping 10 --ping-exit 60" \
              haugene/transmission-openvpn

It appears that about every 20 minutes or so, the connection to my VPN is timing out due to inactivity and rebooting the container. Here's what I can see in the container's logs:

Sun Oct 10 11:18:29 2021 [toronto-rack401.nodes.gen4.ninja] Inactivity timeout (--ping-exit), exiting
Sun Oct 10 11:18:29 2021 /etc/openvpn/tunnelDown.sh tun0 1500 1574 10.11.5.127 255.255.255.0 init
Sending kill signal to transmission-daemon
Successfuly closed transmission-daemon
Sun Oct 10 11:18:30 2021 /sbin/ip route del 154.3.42.136/32
Sun Oct 10 11:18:30 2021 /sbin/ip route del 0.0.0.0/1
Sun Oct 10 11:18:30 2021 /sbin/ip route del 128.0.0.0/1
Sun Oct 10 11:18:30 2021 Closing TUN/TAP interface
Sun Oct 10 11:18:30 2021 /sbin/ip addr del dev tun0 10.11.5.127/24
Sun Oct 10 11:18:30 2021 SIGTERM[soft,ping-exit] received, process exiting
Starting container with revision: ae45b1f22178ddee089d9ea10d4732dbde2d8ef2

It appears that the connection is closing due to inactivity but the strange part is that I am getting this even after downloading a 1.3 gig torrent inside of that period of time. A torrent completed at 00:09 then at 00:17 I got the inactivity error and it restarted the container. Do I have the "OPENVPN_OPTS" line defined incorrectly? Is there something I should be doing or running to keep the connection alive? Or is it okay that the container reboots like this due to inactivity? Thanks in advance for your help!!

[pkishino]

first of all, from a technical point of view there is no problem with the reboots per se, but it should not be occurring. This might be due to issues with the provider predominantly..
mind sharing the .ovpn config ?
also can you share the log from startup?
thanks

[pkishino]

add following env variable:
VPN_CONFIG_SOURCE=custom
that will avoid trying to download from the vpn repo..unecessary..should updated this to be automatic..
anyway, you can remove "OPENVPN_OPTS=--inactive 3600 --ping 10 --ping-exit 60" as those are added to the config automatically as well..
apart from that I would try a different server to check if same occurs

[ryancasler]

Thanks. I'll give that a try. I'll also set up another config to try a different server too.

[ryancasler]

I tried with a different server and I'm still getting the restarts but more inconsistent than before. Like you said, I don't think anything is wrong so I'll just keep an eye on it.

If you don't mind, I did have one other question. When I trash data from the transmission webui, all of the files get deleted but the folders that were created remain. So, I have to periodically go in and clean up these empty folders. Not a big deal at all but is there a permission change that I could make to those folders that would fix that? I checked and the user that i ran the container under can delete folders from there without sudo and inside the container transmission is running as root....so what's the trick I'm missing? Thanks so much for your time....you don't know how much I appreciate it.

[pkishino]

Hmm, strange.
Check the UID and group of the user outside container and you could set the
same inside via the env variables PUID and PGID to try.

[CreatoX]

@pkishino I had the same problem with the same VPN (CyberGhost).

I think it has something to do with using UDP or TCP connection. Yesterday I went from TCP to an UDP configuration and I do not have any problems with a docker restart. Before, using TCP, it was every x minutes. (very random)

Here's my config (for what it's worth)

client
remote 87-1-nl.cg-dialup.net 443
dev tun 
proto udp
auth-user-pass _<path to login credentials file>_

resolv-retry infinite
redirect-gateway def1
persist-key
persist-tun
nobind
cipher AES-256-CBC
ncp-disable
auth SHA256
script-security 2
remote-cert-tls server
route-delay 5
verb 4

ca /etc/openvpn/custom/ca.crt

cert /etc/openvpn/custom/client.crt

key /etc/openvpn/custom/client.key

ping 5
ping-exit 60
ping-timer-rem
explicit-exit-notify 2