Traefik "Enable dashboard": dokploy traefik container gone, all services domains down. [with solution]

[GabriielEger]

To Reproduce

1. Web server > Traefik > Enable dashboard P.S: You've got to have some service(s) using port 8080 when enabling to encounter the problem

Current vs. Expected behavior

I didn't know what I was expecting to happen, because I was just trying to fix another problem with my DBs, but just by clicking "Enable dashboard", now I lost about 4 hours. So I'm just trying to tell about how I fixed it in order to help someone not spend as time as I did.~

Provide environment information

Ubuntu 20.04
Dokploy Version v0.25.6
VPS from Hostinger

Which area(s) are affected? (Select all that apply)

Traefik

Are you deploying the applications where Dokploy is installed or on a remote server?

Same server where Dokploy is installed

Additional context

When I pressed the button to enable dashboard, after a few seconds, nothing happened, and the option was still available "Enable dashboard". But then, suddenly, after reloading the page, I couldn't access dokploy anymore. And here I spend a lot of time until a tried to access via :3000, and then I could access dokploy again (I'm no dev expert - just trying to make things work).

After I could access the dokploy, I couldnt do much (the option "Enable dashboard" was still available, my services domains weren't working at all and neither docker traefik container was running).

But then, after doing a lot of steps/debugs, I tried to update dokploy - nothing happened. Then I removed all my containers (everything), removed personalized networks and tried to reinstall dokploy using the initial code: curl -sSL https://dokploy.com/install.sh | sh

And then, traefik was reinstalled and all my projects/services were perfectly there, no backup needed (I did copied all my composes and .envs, but whatever...).

So, all of this could've been gone really, really different if when pressing "Enable dashboard", some popup informed me "There are a few services running using port 8080 and traefik uses 8080 too" in order to me reevaluate what I'd do, or helping me set another port for traefik (still don't know if it is something easy).

That is it. Spent my afternoon trying to fix it. And I did. So that's a win, I guess. Hope this little story can help someone passing through the same problem, or even help someone come up with something to prevent this from happening.

Thanks./

Will you send a PR to fix it?

No

[harrisyn]

Got into this exact situation a couple of hours ago. I think there have to be global restrictions on some ports to prevent running any containers mapping host ports to any of the Dokploy exposed ports.

My solution was to log into the server via ssh, grep for services running on port 8080 and stopped them, restarting the dokploy_traefik service worked after that and restored service.

[entrptaher]

Wow, I had a real production server with this exact issue just a few weeks ago. I faced this issue even yesterday. Basically the configations on various containers seems to be changed and traefik ends up not finding the right container anymore even though the app is running.

+1