This is a weird one. I've been trying to resolve it for about a week and decided it was time to post to see if anyone had any advice of things to try. I'll try to summarize the steps I've taken and results seen, as well as the general setup.
Despite identical configuration, now I'm seeing 1-2 hours to connect to even a single seed, even on public trackers, such as the Ubuntu ISO (this is my test item). At most, I will get 1-2 connections and most trackers will time out
If it downloads at all, it's at 0.25mbit, but often it won't even get that high
qBit randomly reports firewalling issues, though these tend to go away after about an hour
Port configuration is identical to the original configuration
New system took over the place of the old, including IPs, etc.
The host system can access the internet with no issues and gets 700-800mbit speedtests. Issue appears to be isolated to docker containers
Tested a docker container without gluetun. Speedtest = 770mbit
Tested a docker container with gluetun, but not a torrent. Speedtest = 1mbit
Redid qbittorrent container setup from scratch. No change.
Tested libtorrentv1 version of qbit by various recommendations. No change.
Manually set the interface for qbit. No change.
Verified firewall isn't somehow blocking (it shouldn't, because VPN), and it shows no blocked/dropped packets.
Tested a different torrent solution (transmission and deluge). Speed = 0.25mbit, various connection issues.
Changed IPs
Changed ports
Set "pre-allocate space". No change.
Drive speed is fine (it's a fast SSD).
Tested in TCP/TCP&UDP/UDP modes. No change.
Disabled UFW for giggles. No change.
Given the above, it looks like a gluetun configuration problem.
Redid gluetun container setup from scratch. No change.
Tested a workstation with the same VPN settings, but Windows version. Speeds are 50-100mbit. qBit works there.
Tested old docker setup (again, which is identical) and it works fine.
SERVER_HOSTNAMES=<tested with various servers in diff countries, all with P2P>
cap_add:
NET_ADMIN
My qBit container config:
qbittorrent:
image: linuxserver/qbittorrent:libtorrentv1 #tested this with both "latest" and "libtorrentv1"
container_name: qbittorrent
depends_on:
gluetun:
condition: service_healthy
network_mode: service:gluetun
environment:
I'm assuming I'm missing something obvious here, but I'm not seeing it. I'd believed moving from a guest inside a guest scenario to a physical host just for docker would speed things up, but for this, it's been a nightmare. Open to suggestions!
This is a weird one. I've been trying to resolve it for about a week and decided it was time to post to see if anyone had any advice of things to try. I'll try to summarize the steps I've taken and results seen, as well as the general setup.
[Original configuration]
[New configuration]
Given the above, it looks like a gluetun configuration problem.
My gluetun container config: gluetun: image: qmcgaw/gluetun:latest container_name: gluetun network_mode: bridge restart: always devices:
My qBit container config: qbittorrent: image: linuxserver/qbittorrent:libtorrentv1 #tested this with both "latest" and "libtorrentv1" container_name: qbittorrent depends_on: gluetun: condition: service_healthy network_mode: service:gluetun environment:
I'm assuming I'm missing something obvious here, but I'm not seeing it. I'd believed moving from a guest inside a guest scenario to a physical host just for docker would speed things up, but for this, it's been a nightmare. Open to suggestions!