A user, pablostarter, is having trouble installing Discourse following the official instructions. After trying to access the site, the server doesn’t respond, and they found an error message in the nginx log related to a missing SSL certificate file. pfaffman suggests that the issue might be due to Let’s Encrypt and asks for the app.yml file. pfaffman also mentions that the domain name might not be resolving to the droplet, which could be preventing Let’s Encrypt from working. However, pablostarter discovers that the issue was due to a monitoring process on another server, which was pointing to the subdomain IP. After stopping the monitoring process and updating the A record, pablostarter is able to successfully install Discourse.
I’m trying to install Discourse into a VPS following the provided instructions but after doing everything they say, step by step when I try to visit the site, the server doesn’t respond.
I spent the last to day trying to find out what is the problem and until now the only piece of information that I found and I think it’s useful is the next error message form the nginx log(shared/standalone/log/var-log/nginx/error.log):
[emerg] 2928#2928: BIO_new_file(“/shared/ssl/startups.startuc3m.com.cer”) failed (SSL: error:02001002:system library:fopen:No such file or directory:fopen(‘/shared/ssl/startups.startuc3m.com.cer’,‘r’) error:2006D080:BIO routines:BIO_new_file:no such file)
Does the domain name that you used resolve to your droplet? If not lets encrypt cannot work.
You might disable (comment out with a # at the start of the line) the two lines that have to do with letsemcrypt, or start again and don’t define the let’s encrypt email address.
Ok! I found the problem, I had changed the A record so the subdomain was pointing to my server, what I forgot is that I had another server monitoring that subdomain ip and changing it to point to it.
I just stoped the monitoring process, changed the A record again, ./launcher rebuild app and now it’s working lika a charm.