Swag stopped resolving reverse proxy

LSIO Discussion Container Support
,
1

Hi,
I’m really lost, hopefully you will be able to help me.

I’m running swag docker image on my home server (ubuntu).
my setup was simple, certbot set up with zerossl and plex + overseerr (both lsio images) set up as subdomains plex.* and overseerr.* using sample configuration. I followed docs and everything worked fine.
today neither addresses are working.
I checked both plex and overseerr using IP address and they are both up and running.
I updated ubuntu and all docker images.
I checked if my domain DNS records are up to date, they are pointing to the correct IP address.
I checked if my router is correctly routing ports 80 and 443 to my server.
My zerossl certificate was about to expire, I renewed it just to be sure.
I found out timezone on my server was wrong, I updated it but no change.
I checked swag logs (docker logs swag), everything looked ok, there was something about wrong version dates of proxy.conf, I deleted it, restarted container, log is clean now.
I looked into nginx logs. I found multiple errors, not sure if any are relevant.
there are multiple SSL: error:141CF06C:SSL routines:tls_parse_ctos_key_share:bad key share but after some googling this looks like non issue.
then there is this:

[error] 6000#6000: *322764 FastCGI sent in stderr: "Primary script unknown" while reading response header from upstream, client: 20.83.148.119, server: _, request: "GET //xmlrpc.php?rsd HTTP/1.1", upstream: "fastcgi://127.0.0.1:9000", host: "<MY IP REDACTED>"
[error] 6000#6000: *333203 FastCGI sent in stderr: "Primary script unknown" while reading response header from upstream, client: 87.251.85.227, server: _, request: "GET /roots.php HTTP/1.1", upstream: "fastcgi://127.0.0.1:9000", host: "<MY DOMAIN REDACTED>", referrer: "https://<MY DOMAIN REDACTED>/roots.php"
[error] 6000#6000: *333233 FastCGI sent in stderr: "Primary script unknown" while reading response header from upstream, client: 180.149.125.169, server: _, request: "GET /stalker_portal/server/tools/auth_simple.php HTTP/1.1", upstream: "fastcgi://127.0.0.1:9000", host: "<MY IP REDACTED>", referrer: "http://<MY IP REDACTED>/stalker_portal/server/tools/auth_simple.php"

I honestly have no idea what are they, if this is the issue or how to fix this. but these errors appear multiple times and some are few days older and I’m pretty sure overseerr worked 2-3 days ago.

Any ideas what could be the issue?

2

are you relying on hairpin nat? does nslookup resolve to your external ip? did your ISP recently update your router? please share full docker logs and your compose snippet.

also share the proxy-conf for both plex and overseerr. Finally, please verify you can access both of those directly (ie, without swag)

3

Hi driz!

not sure what hairpin nat is, I guess I don’t rely on that then.
yes, nslookup resolves my external IP. there were no updates from ISP and I’m able to access both plex and overseerr directly using IP address.

I apologize, but I’m not sure what you mean by full docker logs, so I attached logs for swag, plex and overseerr.
here is everything: 32.4 KB file on MEGA

Thank you for help!

4

Hi driz,

I just found out the issue had nothing to do with swag, it it something with my pi-hole server not resolving my own domain.

sorry to inconvenience you during holidays. this post can be closed now.

Regards,
Oliver