Downtime & Poor Performance
During the month of August, we moved all managed websites over to new Virtual Machines. The reason for this move is that some websites were experiencing downtime. We attributed the downtime to running out of disk space on each instance, which were set at 10GB.
When we looked at the space being taken up, it was due to excessive logging by MySQL and Google Cloud related software. The configuration that created this scenario was from our using a Google Cloud Marketplace click-to-install solution for WordPress.
Server Updates & Maintenance
The solution was to build our own LAMP servers from scratch so that we were in control of every aspect of the server and ensure optimal operation and performance. We started by using Ubuntu as the operating system, replacing Debian, both of which are Linux operating systems. We then installed and configured Apache, MySQL, PHP, and WordPress manually, and migrated the sites to their new homes. We also reset the counters for UptimeRobot.
All managed websites now have 30GB of storage and the new configuration resulted in freeing up an average of 5GB of disk space, as well as eliminating the intermittent downtime and performance issues.