So…it’s working now? I haven’t touched anything yet, but I just checked my instance again and it works perfectly fine on desktop now. It always worked through Voyager, so I was able to let people know there was an issue. If it comes back I’ll try some of these suggestions to find a more permanent fix.

Meaning I just wait a bit and retry later?

Upgraded to 0.18.5, which now gives me this error:

lemmy-lemmy-ui-1  | API error: FetchError: request to http://lemmy:8536/api/v3/site?auth=AUTH_KEY failed, reason: connect ECONNREFUSED IP_ADDRESS:8536

I created a gist with my compose file: https://gist.github.com/osiriswrecks/26a875576d3bbcf11923d7715ac15e6e. It should be stripped of all private info. I tried changing the version from 0.17.3 to 0.18.0 and the server returns a gateway error after restarting.

Done. And restarting the lemmy container appears to have helped! Now if I click “forgot password” with my email in the user field it at least shows a “sent a reset email” notification. I don’t see an email yet though, either in inbox or spam, and no error logs from what I can tell but I’ll keep looking.

Ah SHIT. Thanks, doing it now.

I updated my docker-compose.yml above to the full file. Does that help?

No, I didn’t have anything there. The docs say everything but the login and password are required, but because I’m using the relay package I don’t know what those would be to begin with because I’m not using an external SMTP with an account.

!!! THAT DID IT.

Thank you so much I would have never gotten to that solution on my own. Works beautifully now.

I’m almost positive that network is already there but I’m definitely going to check the Lemmy-specific logs. I only knew how to check the general server logs when I last looked at it. If I ever get my kid to bed I’ll take another look.

“No results” never goes away though. I’ve straight up left it and walked away for like 10 min and no change. That’s why I’m starting to wonder if it’s an issue with my instance.

It’s turns out, my issue was that I had nginx on both the host and Docker container, so the host wasn’t forwarding the websocket requests like this. I just removed the host nginx and configured the Docker conf to look at port 80 and it worked!

Not to hijack the conversation, but once a server is open and set up what are the best ways to attract people to it? Just start telling people about it?

So looking at this again now, am I taking that whole block and adding it to the container’s nginx.conf? If so, does that mean I have to change what port it’s currently listening to (because there’s already a rule in the file for port 80)?

There’s a comment in that server rule that says “this is the port inside docker” and a comment immediately after that says “this is facing the public web”, which confuses me.

The one meant for the Docker container or the one on the host?

I might be. See my response here: https://lemmy.world/comment/101425

I think this is where my lack of experience with Docker is showing.

I spun up a DO droplet and installed nginx, Docker CE, and Docker Compose. Then I went through the instructions on the page you linked to and it set it up just fine but when I went to my droplets IP address it wouldn’t connect. I had to add a config file that pointed traffic coming into the droplet on port 80 to redirect to the Docker container instead. Am I overcomplicating it?

I seem to be having a lot of lag at the moment, and my post was created twice so I’m just going to delete the other one and start from here…

So I have this set up per the instructions. My instance is on a Digital Ocean instance, and I’m using nginx on the host to point to localhost:1235, but that’s about all that conf file is doing. Is there something else I need to do?