Open traviscrist opened 6 years ago
traviscrist
commented
6 years ago 
Miningman111
commented
6 years ago First, we started by following along with the kungfunerd on youtube.
We also cross-referenced two codius installation articles. https://docs.google.com/document/d/1WWQsEed4dfigZff2fJTFoLb9mVWhpkOs3ao7CUn56MM/edit and https://medium.com/codius/how-to-run-your-own-codius-host-42e13afe1fb2
There were some tips that people shared with us, such as waiting AT LEAST 30 minutes before entering second key for DNS record, and wait ~48 hours for your domain to be listed/working. Initially, we had to ask hostgater to direct our domain to our static IP, even though we thought we had taken care of it.
Our instance became unresponsive, so we wiped it and started over. In Google cloud, after we created our second instance, our OLD static IP was given to us after we requested to set it to static.
First step, before installing anything else, was to update CentOS 7. [yum check-update] [yum upgrade] [yum clean all]
Then we followed the installation guide on medium, cross referencing with the second article containing a walk-through with tips.
We had to do the certbot again. (Regarding the two acme records: after creating the first, WAIT FOR 30 MINUTES!!! At least 30! This is very important, as we assume you'd rather not waste your time trying to troubleshoot)
Assuming that you have followed every step that pertains to your particular installation type, including permanently opening ports and ensuring the XRP was taken out of your wallet for the payment channel auth, you'll want to enable and reload/(stop/start) every service pertaining to Codiusd, and then finally enable and reload codiusd.
Our reasoning: any changes made in the conf files would require the program to be restarted for the new parameters to be understood/executed. After moneyd, nginx, and hyperd have been reloaded, we can reload codiusd.
I've been saying reload vs restart because in the warning for restart, it states that you may lose files or they may become corrupted?
After codiusd reloaded, we checked the status of each program using [systemctl status (app name (codiusd, moneyd-xrp, hyperd, nginx)) -l] running only one at a time. Everything looked good.
We pinged our server/info
BOOM! Self-test true and running a contract.
Haven't ran one since..but that's beside the point! Hopefully this has helped you! Remember...sometimes, you just have to start a fresh instance.
traviscrist
commented
6 years ago @Miningman111 this is all really great info! I think it would be great to get this added to the wiki for others to use, can you look into making a PR to add this info to the wiki, the medium tutorial on setting up a Codius host is at https://github.com/codius/codius-wiki/wiki/HostTutorial which you could update if you feel that is the right place.