ERROR 500 - Internal Server Error for CMS after fresh auto install ispconfig

Hi

I have a few websites including joomla running in ispconfig + apache, and I am trying to migrate them to a new server with ispconfig + ngnix.

I use the auto install script to install ispconfig 3 with ngnix on ubuntu 18.04, and everything runs fine.
then I created all the sites in ispconfig (not copy the ispconfig data), and copy all the websites to the new server.

1. somehow all sites got 504 gateway time-out error.
2. to see if this is new server setting issue, then I installed a new copy of joomla to the ngnix server, the installation process is fine. I can login to backend, and I can open the frontend. but if I click any link like lost password, then I got ERROR 404 - Not Found! (I fixed the internal server 500 error by changing the files permission)

how do I find out what is causing the 504 gateway time-out error and ERROR 404 - Not Found!

error log from site:
2020/11/15 20:13:38 [error] 16052#16052: *458 openat() "/var/www/mysite.com/web/j4/index.php/component/ajax" failed (20: Not a directory), client: 60.246.xx.xx, server: mysite.com, request: "GET /j4/index.php/component/ajax?format=json HTTP/1.1", host: "www.mysite.com", referrer: "http://www.mysite.com/j4/"
2020/11/15 20:27:38 [error] 16052#16052: *503 openat() "/var/www/mysite.com/web/j4/index.php/component/ajax" failed (20: Not a directory), client: 60.246.xx.xx, server: mysite.com, request: "GET /j4/index.php/component/ajax?format=json HTTP/1.1", host: "www.mysite.com", referrer: "http://www.mysite.com/j4/"
2020/11/15 20:31:14 [error] 22559#22559: *10 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 60.246.xx.xx, server: mysite.com, request: "GET / HTTP/1.1", upstream: "fastcgi://unix:/var/lib/php7.2-fpm/web2.sock", host: "www.mysite.com"


from ngnix error log below, I only saw below during that time.

2020/11/15 20:00:57 [crit] 16052#16052: *438 SSL_do_handshake() failed (SSL: error:14209102:SSL routines:tls_early_post_process_client_hello:unsupported protocol) while SSL handshaking, client: 80.82.xx.xx, server: 0.0.0.0:443
2020/11/15 20:01:02 [error] 16052#16052: *442 open() "/usr/share/nginx/html/robots.txt" failed (2: No such file or directory), client: 80.82.xx.xx, server: , request: "GET /robots.txt HTTP/1.1", host: "8.210.xx.xx"
2020/11/15 20:01:03 [error] 16052#16052: *443 open() "/usr/share/nginx/html/sitemap.xml" failed (2: No such file or directory), client: 80.82.xx.xx, server: , request: "GET /sitemap.xml HTTP/1.1", host: "8.210.xx.xx"
2020/11/15 20:01:04 [error] 16052#16052: *444 open() "/usr/share/nginx/html/.well-known/security.txt" failed (2: No such file or directory), client: 80.82.xx.xx, server: , request: "GET /.well-known/security.txt HTTP/1.1", host: "8.210.xx.xx"


thanks

 

Which version of ISPConfig 3?

How?

Which files and to which permission? Usually it makes things worse to change the file and directory permissions that ISPConfig creates.

Is that a copied site or the new site you created on tne new server?
Does the new server have the same IP-number as the old? Have you changed name service or changed domain names of the websites? You can test tne new server without altering DNS: https://www.faqforge.com/linux/cont...ess-a-namebased-website-without-a-dns-record/ If the website has url with domain name, you may need to edit /etc/hosts on the webserver while doing this testing.
Is there a reason to switch from apache to nginx?