We understand that our services are critical to the business of many of our clients. That's why we do everything we can to make sure that everything operates as smoothly as possible. One of the most important elements in this is proper monitoring.
We use New Relic to monitor our applications platforms. Apart from the expected backend monitoring, we also measure the user's experience. Both what you experience as an admin and what your subscribers experience when interacting with the system.
The following parts of our system are monitored through New Relic:
- Backend performance
Real User Monitoring
This keeps track of different DOM events on every page view. This is done in order to monitor network delays as well as user experiences, e.g. time spent interpreting and rendering
- Server monitoring
- SQL performance
A series of thresholds have been set for each element we monitor. As soon as something differs from the norm, our operations staff is notified.
We use Pingdom to monitor the availability of the systems that communicate outwards. Through Pingdom, these are monitored from various different, geographically dispersed, places. This means that we get alerted quickly if one of our systems suddenly stops responding.
The following services are monitored through Pingdom:
- The administration system (subscribing and unsubscribing)
- CDN (Content Delivery Network)
- SMTP cluster
These data are gathered in an uptime report, which summarises uptime across a three month period.