Stop restarting apache2 every hour

root's crontab has carried "20 * * * * systemctl restart apache2" since
before Anubis, unmanaged. It guards nothing (children settle around 25 MB
and never leak) and it causes the worker exhaustion it looks like a cure
for: 57 of the 91 AH00484 events in the last two weeks fall within eight
minutes of it, when the 63,000 requests an hour hitting the front door all
reconnect at once against StartServers 2.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

Author: hsbt

recipes/apache2.rb

@@ -76,3 +76,13 @@
     end
   end
 end
+
+# root's crontab carried "20 * * * * systemctl restart apache2" from before
+# Anubis, outside this repository. It fixed nothing: children settle at 25 MB
+# and never leak. It caused harm instead. 57 of the 91 AH00484 (MaxRequestWorkers
+# exhausted) events in the last two weeks landed within eight minutes of it,
+# because the 63,000 requests an hour arriving at the front door all reconnect
+# at once against StartServers 2.
+file '/var/spool/cron/crontabs/root' do
+  action :delete
+end