The volume of post-deployment steps happening in deploy.sh will cause the production site to be down for an unforgivable amount of time. Sending locks to memcache to mitigate a menu rebuild stampede allowed the site to remain 'up' while the deployment finished.
The volume of post-deployment steps happening in deploy.sh will cause the production site to be down for an unforgivable amount of time. Sending locks to memcache to mitigate a menu rebuild stampede allowed the site to remain 'up' while the deployment finished.