I’m currently under a dynamic IP address, and I have an issue with IP changes:
When changing ip, the vhosts having that ip, at least in the CP UI, change their IPS to the local server address (in my case, 192.168.1.10). I’ve never had really any issue with that since they seem to keep working. Though, I recently discovered some hosts just not responding at all, and when rebuilding them, the NGINX conf file just doesnt include any ip (eg has listen :80, or proxy_pass http://:80 with no Ip), and the restart fails. Its pretty annoying, and i would know if there is a fix / if its a known issue. For a quick fix I’ve thinked about giving all hosts the 0.0.0.0 ip, but i don’t know if its possible.
By default, Nginx listens on all IP addresses assigned to a server when the listen directive is not explicitly configured with a specific IP address. It sounds like you are hosting from a home LAN behind NAT, which means your public IP is never active on your server interface anyway.
This all means that your config file is correct and you shouldn’t alter it. This is generally the case in HestiaCP. If you need to make changes, you would normally be using templates or the settings exposed through the panel itself.
Hey, thanks for the response. Nginx indeed listens to all IPS when the ip isnt specified, but the issue is that the templates dont seem to handle it, if %ip% is empty, the generated config file that was, in the template, listen %ip%:%proxy_port%; becomes listen :80; instead of listen 80;, which causes nginx to crash when rebuilding the domains (same with directives like proxy_pass). In my case I’ll probably just edit the templates to not include the %ip%:part, but I don’t know if its the best thing to do.