Bad Gateway with traefik and rallly

[fibis]

We are using rallly together with Traefik as reverse proxy. I have upgraded rallly from version 2 to 3.

We always worked with a "wrong" setup, where we didn't add the scheme to the public base, so ´NEXT_PUBLIC_BASE_URL=example.cominstead ofNEXT_PUBLIC_BASE_URL=https://example.com`. This was working great with version two, but in version 3, we're now getting an ERR_INVALID_URL-Error. So I switched to the correct NEXT_PUBLIC_BASE_URL and now we are getting a Bad Gateway from Traefik. Directly accessing the application is still possible.

Is there something happening, that is causing this Bad Gateway error, when changing this env?

[lukevella]

I've never used traefik so I don't know anything about it but it sounds like it's not forwarding all the request headers. In nginx you would use proxy_pass to tell it to forward all headers. Not sure what the equivalent in traefik is but hope that helps.

[fibis]

@lukevella that was my first idea. But I only change with NEXT_PUBLIC_BASE_URL the behavior of rallly, not of the load balancer. That is the part, that I am not understand.

[corvec]

I'm also using Traefik and when updating from an earlier version of v3* I also started getting Bad Gateway errors. I reverted to my old version and that solved the issue. I'm not sure if you're experiencing the same issue, but just in case:

Today I noticed #947 and tried out the fix suggested there by @scottwallacesh. I added hostname: 0.0.0.0 and switched to the lukevella/rallly:3.4.2 image and that solved the Bad Gateway error I'd been seeing. Might be worth a try for you, too.

* - I'm unsure which version exactly. I pulled it on September 20th from the lukevella/rallly:3 tag, but the image isn't tagged anymore in dockerhub. The ID was sha256:0f1dfe32acfd81fd85dfc8c28031be629a6731a4f78280cd11a7d0216f5e3df0

EDIT: My docker-compose config is here

[fibis]

@corvec thanks, adding the hostname to the docker-compose file fixed the problem.