Losing network connection on server and speedtest fails #1023
-
|
Describe the bug This has been an issue for a while, but I'm unsure when it first happened. It could've started an update, but that I can't confirm. The data logs in the web-UI mention using the path docker-compose.yml# SPEEDTEST-TRACKER
speedtest-tracker:
container_name: speedtest-tracker
environment:
- PUID=$PUID
- PGID=$PGID
- DB_CONNECTION=mysql
- DB_HOST=mariadb
- DB_PORT=3306
- DB_DATABASE=speedtest_tracker
- DB_USERNAME=speedtest
- DB_PASSWORD=$SPEEDTEST_DB_PASSWORD
- MAIL_HOST=$MAIL_HOST
- MAIL_PORT=$MAIL_PORT
- MAIL_USERNAME=$MAIL_USERNAME
- MAIL_PASSWORD=$SPEEDTEST_MAIL_PASSWORD
- MAIL_FROM_ADDRESS=$MAIL_USERNAME
volumes:
- $DATA/speedtest-tracker:/config
labels:
- traefik.enable=true
- traefik.http.services.speedtest-tracker.loadbalancer.server.port=443
- traefik.http.services.speedtest-tracker.loadbalancer.server.scheme=https
- traefik.http.routers.speedtest-tracker.rule=Host(`speedtest.$DOMAIN`)
image: ghcr.io/alexjustesen/speedtest-tracker:latest
restart: unless-stopped
depends_on:
- mariadb
networks:
- default-networkTo Reproduce Environment (please complete the following information):
Screenshots Logs Docker container logs |
Beta Was this translation helpful? Give feedback.
Replies: 7 comments 3 replies
-
|
So It's interesting how your network is dropping connection when a speedtest is run... Can you try the following steps and let me know the outcome? Also looking for logs that don't reference the download threshold from above.
|
Beta Was this translation helpful? Give feedback.
-
|
I've disabled the threshold. CLI commandWhen I tried running the CLI command. After running the command it seems to work fine for the first couple of percent (the first time it stopped at 4% and the 2nd at 8%): But then it suddenly stops, I temporarily lose connection to the server, and the CLI output says Manual run from webI got the same message as before when manually running a speedtest from the web-UI. The docker container logs doesn't mention anything about the run now that the threshold is disabled. The only things in the container logs are the requests for the admin panel from my browser, so no errors there. |
Beta Was this translation helpful? Give feedback.
-
This is good at least from my app's stand point this means the error isn't there. Are you able to run speedtests in the browser without your connection dying? |
Beta Was this translation helpful? Give feedback.
-
|
Yeah, the network test via speedtest.net works just fine, with expected speeds. I think I'll try with another docker network config, but if you have any suggestions I'll be happy to test anything. |
Beta Was this translation helpful? Give feedback.
-
|
Yeah this is a new one, after changing the docker network I'd also try searching the logs on your host machine and your router to see if any clues might be there. I'm going to transfer this issue to a discussion so please keep the updates coming on your findings and I'll open a new issue if we can target something concrete. |
Beta Was this translation helpful? Give feedback.
-
|
I wanted to chime in and say I'm having the same problem as NorskNoobing.
|
Beta Was this translation helpful? Give feedback.
-
|
I'm having the same issue, any update on this topic? |
Beta Was this translation helpful? Give feedback.
-
|
I just wanted to update this topic, because a couple of days ago I re-tested the container, and it's working! Maybe the issue I had has been patched in the later versions of the image? It could be unrelated, but the only change/networking change I've made is to the Proxmox host that the VM with the docker containers ran on. On the Proxmox host I removed the following lines from the Here's my full working docker-compose.yml# SPEEDTEST-TRACKER
speedtest-tracker:
container_name: speedtest-tracker
image: lscr.io/linuxserver/speedtest-tracker:latest
environment:
- APP_KEY=$SPEEDTEST_APP_KEY
- DISPLAY_TIMEZONE=$TZ
- RESULTS_POLLING=60s
- SPEEDTEST_SCHEDULE=*/30 * * * * #Run at every 30th minute
- PUID=$PUID
- PGID=$PGID
- DB_CONNECTION=mysql
- DB_HOST=mariadb
- DB_PORT=3306
- DB_DATABASE=speedtest_tracker
- DB_USERNAME=speedtest
- DB_PASSWORD=$SPEEDTEST_DB_PASSWORD
- MAIL_HOST=$MAIL_HOST
- MAIL_PORT=$MAIL_PORT
- MAIL_USERNAME=$MAIL_USERNAME
- MAIL_PASSWORD=$SPEEDTEST_MAIL_PASSWORD
- MAIL_FROM_ADDRESS=$MAIL_USERNAME
volumes:
- $DATA/speedtest-tracker:/config
labels:
- traefik.enable=true
- traefik.http.services.speedtest-tracker.loadbalancer.server.port=443
- traefik.http.services.speedtest-tracker.loadbalancer.server.scheme=https
- traefik.http.routers.speedtest-tracker.rule=Host(`speedtest.$DOMAIN`)
- homepage.group=Networking
- homepage.name=Speedtest Tracker
- homepage.icon=sh-speedtest-tracker.svg
- homepage.href=https://speedtest.$DOMAIN
- homepage.description=
depends_on:
- mariadb
restart: unless-stopped
networks:
- default-network |
Beta Was this translation helpful? Give feedback.
-
|
I am having this same issue for 4 days now. My Speedtest containers are connected to the network via macvlan if that helps? |
Beta Was this translation helpful? Give feedback.
-
|
What does your composefile look like? You can try to take my settings, or maybe change the networking mode to host? |
Beta Was this translation helpful? Give feedback.
I just wanted to update this topic, because a couple of days ago I re-tested the container, and it's working! Maybe the issue I had has been patched in the later versions of the image? It could be unrelated, but the only change/networking change I've made is to the Proxmox host that the VM with the docker containers ran on.
On the Proxmox host I removed the following lines from the
/etc/network/interfaces, because I don't use VLANs, and it has drastically improved the network stability of my homelab. Some seemingly random network issues that happened when rebooting have also been fixed by this. Here are the two lines I removed from thevmbr0network interface, then I ransudo systemctl re…