azacan

Happened to me following same upgrade path and using Firefox. I got it solved, as mentioned here, by rebooting the system (no need of safe mode) and doing the first log in with a different browser, not Firefox. I tried Brave (based on Chrome) and everything went fine.

Hi, Any idea on how to solve this or even if it is possible? Thanks!

Hi, I have been trying to setup something I am not sure is possible to do with my current setup and swag. Basically is to reverse proxy http only services on my unraid machine from a domain like photoprism.lan to its containerIP and port (2342) I have swag running on unraid 6.9.1 host and listening on ports 80 and 443, those are port forwarded from my router for external access. I can successfully access my desired services running on https behind the subdomain certs I have generated for nextcloud and bitwarden: nextcloud.mydomain.com and bitwarden.mydomain.com. Everything works fine also internally: I have two entries on PiHole internal DNS server that resolves nextcloud.mydomain.com and bitwarden.mydomain.com to the local unraid IP where swag nginx is listening. Now I am trying to make use of the nginx reverse proxy on swag to locally access a new service on my unraid, in this case photoprism. The thing is that photoprism gui is running on port 2342 and is running over http. I would like to access photoprism with a domain (different from my external one used for nextcloud and bitwarden) and without needing to write the port each time, for example with http://photoprism.lan and no port (I have added a dns entry on the pihole to resolve photoprism.lan to the unraid IP where swag nginx is listening) but I have not find a way to configure a proxy-conf in nginx that proxies this domain to the right IP and port. What I have tried, among many other things is to put a file (local-servers.conf) inside proxy-confs folder of ngingx with: server { listen 80; server_name photoprism.*; location / { include /config/nginx/proxy.conf; resolver 127.0.0.11 valid=30s; set $upstream_app photoprism; set $upstream_port 2342; set $upstream_proto http; proxy_pass $upstream_proto://$upstream_app:$upstream_port; } } I have tried also with server_name photoprism.lan* Although the internal docker dns works ok resolving the container name, I have tried also setting the proxy_pass with the final docker IP and port with no luck. When I try to go to http://photoprism.lan I got redirected to a https://photoprism.lan/ and see the default nginx webpage: Welcome to our server The website is currently being setup under this address. For help and support, please contact: [email protected] Is this because by default only https is being configured to be proxied? Any way of allowing http for internal lan without compromising security? My certs are subdomains, as stated above, like nextcloud.mydomain.com but the photoprism is not in the same domain but photoprism.lan, does this cause the failure? Thanks!