I installed on a DO droplet and I didn’t daemonize caddy. I just started the server with nohup caddy & then disown the process. It ran for quite some time like that, but went to sleep at some point and stopped serving the site. What should I do to avoid that happening?
The droplet went to sleep or the process? (Either way, I do the same thing and have never heard of that problem or had it myself.)
When I used to run Caddy by itself I would just open up byobu and run Caddy directly. Worked perfectly for me.
The process went to sleep. I don’t think anything went on with the droplet. When I tried to load the site this morning from having put it up yesterday, it wouldn’t load. I started with confirming that caddy was still running and this was the ps aux output: Screen Shot 2016-06-14 at 1....
I’m not quite sure, but the error I got from the browser was that the server was unresponsive, either that or “not found”.
Now I’m kind of bummed that I killed it[the process] and restarted. It hasn’t happened again at this point and the site has been back up since 6am CST-US.
I’ve never heard of Linux putting a server process to sleep on its own, that’s so weird.
If you ever find out more about it, or can reproduce the behavior, I’d be interested in learning more about it.
Thanks for letting us know! If anyone else is reading this and has experienced the same problem please chime in.
Could it be related to systemd? Systemd v230 kills background processes after user logs out, breaks screen, tmux | Hacker News
1 Like