New webserver not running - manual intervention needed?

[mdcfe]

It seems like uWSGI still isn't being exposed by Nginx, despite the #515 hotfix, and somehow the old server.py is running on port 80 even though startup.d/70-webserver.sh should be killing it. In addition, neither Nginx nor uWSGI seem to be logging to logs/ as should have happened in as of #515 and #488 respectively.

Is it possible to check whether:

• etc/nginx/chaos_uwsgi has been linked to /etc/nginx/sites-enabled,
• Nginx has been reloaded and isn't erroring,
• The Nginx logs have been linked to logs/,
• uWSGI is actually running (seems to be, according to stderr), and
• chaos_server has been killed correctly?

(Not sure who to ping - is @amoffat the host?)

[amoffat]

• linked etc/nginx/chaos_uwsgi to /etc/nginx/sites-enabled
• linked the nginx log files to log/
• killed the chaos_server and brought it back up
• reloaded nginx

I'm not sure how to test the new webserver, can anyone check that the expected behavior is now happening?

Side note, I'll try to work on our full docker deployments today, this way we can manage the server's dependencies from the Dockerfile and have fewer problems like these

[mdcfe]

/vote close Seems to be working - the status code page rendering endpoint should show the code given with the ?status parameter.

[Moulick]

What ever someone just did, its back. Like this very instant it come up

[andrewda]

@Moulick It keeps going down then coming back up. It's down again now.

[chaosbot]

/vote close Seems to be working - the status code page rendering endpoint should show the code given with the ?status parameter.

Command Ran

[chaosbot]

:no_entry: The issue has been closed after a vote.