request for robots.txt crashes server

[acozine]

This weekend the press.curationexperts.com server got a a request from a webcrawler for press.curationexperts.com/robots.txt. Instead of returning the contents of public/robots.txt, the server interpreted the request as seekeing a subdomain (press) callled "robots" and crashed. I think this will go away when #107 is done, and the press home pages look like umich,press.curationexperts.com instead of press.curationexperts.com/umich.

Gist of the production log is here: https://gist.github.com/acozine/b22f24ff5caccb36e8ab9f30976c5c93.

[acozine]

Problem fixed by an extension of the ProxyPass directives in /etc/apache2/sites-available/heliotrope.conf. Once I added specific directives for the files in public/, the errors went away. Full VirtualHost definition for use with Puma is:

<VirtualHost *:80>
  ServerName 127.0.0.1
  # !!! Be sure to point DocumentRoot to 'public'!
  DocumentRoot /opt/heliotrope/current/public

  ProxyPass /assets !
  ProxyPass /favicon.ico !
  ProxyPass /robots.txt !
  ProxyPass /system !
  ProxyPassMatch ^/(404|422|500).html$ !

  ProxyPass / http://127.0.0.1:3030/
  ProxyPassReverse / http://127.0.0.1:3030/

  # allows forward slashes for IIIF
  AllowEncodedSlashes NoDecode
  <Directory /opt/heliotrope/current/public>
        # MultiViews must be turned off.
        Options -MultiViews
        # Uncomment this if you're on Apache >= 2.4:
        Require all granted
  </Directory>
</VirtualHost>

where Puma is running on port 3030.