Mastodon Meta Discussion Board

Services "mastodon-syskiq" and "mastodon-web" won't start?

TechnicalSuwako 2017-10-01 20:06:58 UTC #1

I’ve been on the instance “mstdn.jp” for my personal account (because it was the first instance I heard about in the media), and I’ll continue being there.
But now I want to set up an instance for just my brand accounts that are still stuck on Twitter.

But while “mastodon-streaming” is running fine, the other 2 services are not.
Furthermore, I can’t find any more hints (not even logs!) than:

Oct 02 02:33:49 SERVER-NAME systemd[1]: Started mastodon-sidekiq.
Oct 02 02:33:49 SERVER-NAME systemd[1]: mastodon-sidekiq.service: Main process exited, code=exited, status=203/EXEC
Oct 02 02:33:49 SERVER-NAME systemd[1]: mastodon-sidekiq.service: Unit entered failed state.
Oct 02 02:33:49 SERVER-NAME systemd[1]: mastodon-sidekiq.service: Failed with result 'exit-code'.
Oct 02 02:33:49 SERVER-NAME systemd[1]: mastodon-sidekiq.service: Service hold-off time over, scheduling restart.
Oct 02 02:33:49 SERVER-NAME systemd[1]: Stopped mastodon-sidekiq.
Oct 02 02:33:49 SERVER-NAME systemd[1]: mastodon-sidekiq.service: Start request repeated too quickly.
Oct 02 02:33:49 SERVER-NAME systemd[1]: Failed to start mastodon-sidekiq.

And:

Oct 02 02:33:49 SERVER-NAME systemd[1]: Started mastodon-web.
Oct 02 02:33:49 SERVER-NAME systemd[1]: mastodon-web.service: Main process exited, code=exited, status=203/EXEC
Oct 02 02:33:49 SERVER-NAME systemd[1]: mastodon-web.service: Unit entered failed state.
Oct 02 02:33:49 SERVER-NAME systemd[1]: mastodon-web.service: Failed with result 'exit-code'.
Oct 02 02:33:49 SERVER-NAME systemd[1]: mastodon-web.service: Service hold-off time over, scheduling restart.
Oct 02 02:33:49 SERVER-NAME systemd[1]: Stopped mastodon-web.
Oct 02 02:33:49 SERVER-NAME systemd[1]: mastodon-web.service: Start request repeated too quickly.
Oct 02 02:33:49 SERVER-NAME systemd[1]: Failed to start mastodon-web.

So, how do I get those to work?
Or at least, more information about why those fail to start.

TechnicalSuwako 2017-10-03 21:00:28 UTC #2

Oh, is anyone here yet?

saper 2017-10-03 22:26:59 UTC #3

anything in the actual logs? can you try to start puma or sidekiq by hand? Is the database access working?

TechnicalSuwako 2017-10-03 22:47:17 UTC #4

Sidekiq and Puma were not installed yet it seems.
I installed sidekiq though apt, but puma seems not to be installable through apt.
I figured out I had to install it through Ruby, I did, but how do I actually start it now?

I assume it does.

saper 2017-10-03 23:04:57 UTC #5

Do not install components by hand. You will introduce a hard to debug mess. Sidekiq and puma should live in your bundler directory together with the rest of mastodon.

Never assume anything. Test.

saper 2017-10-03 23:07:34 UTC #6

documentation/Production-guide.md at master · tootsuite/documentation · GitHub there you will find the contents of systemd unit files. systemd should be logging the output of the failing commands (journalctl --unit=xxx), if you cannot find it - try running ExecStart commands as the mastodon user - but make sure all environment variables are set!