Redundant Off-Network Monitoring Reports
Although we're very trustworthy people, we feel obligated to provide our customers with completely impartial and untainted server uptime reports. You can view detailed statistics for any of our many servers below.
These uptime statistics are provided for customer reference. We also run our own internal monitoring system which polls all services on all systems every 2 minutes (more accurate and faster to respond).
| Monitoring Interval: | Two Minutes |
| Service Monitored: | HTTP (Port 80) |
| Monitoring Locations: | San Diego, CA New York City, NY |
Caveats and Possible Discrepancies
Scheduled Maintenance and Upgrades - These statistics do not account for scheduled maintenance outages. Every host needs to do maintenance to keep their fleet up to date, secure and running smoothly. Although we do this work during off-peak hours (at night), it still counts as "downtime" in the eyes of this monitoring service.
Webserver-only Monitoring - Only the webserver (Apache, port 80) is monitored. If the webserver is down but other services (email, FTP, mySQL) are up, the entire server reports down.
Sampling Error - The reported uptime may vary slightly from the server's actual uptime, as the monitoring interval is set at 2 minutes. For instance, if a server goes down right before a service check, and then we reboot it and it's back online within 1 minute, the monitor will catch in on the next check (which is in 2 minutes) and estimate a 2 minute downtime when in reality there was only 1 minute of downtime.
Webserver Configuration Error - The third party monitoring system counts 404 and 500 errors as negative responses, when they are, in fact, evidence that the webserver is up and running. On multiple occasions there has been a mis-configuration with our monitoring target script which caused the remote system to think that a server was down when it really wasn't. Because the measurements are taken by a third-party company, we can not modify them to compensate for these types of errors.