Stratus has active HTTP monitoring for your sites if they meet the following criteria:
- Your domain exists in the URLs section of your Stratus environment panel.
- The domain is on a production Stratus instance i.e. not a dev instance. Development domains are not monitored.
- A valid SSL is applied that covers your domain.
- We have been granted GA access to validate session usage. Un-validated domains are not monitored.
We also monitor the root domain, and accept redirects to avoid duplicate alerts. www.example.com is not directly monitored, but example.com would be and if redirected to www the last redirect status code is recorded.
If you have additional domains/pages you want monitored, there are innumerable 3rd party services out there for you.
Rather than monitoring for every possible failure, our monitoring is focused on identifying system-side issues within Stratus. We only monitor the following status codes:
404s,401,403,503 etc are all ignored. Customers frequently restrict their sites and make changes that cause error responses. Though they are not really causes for alarm. 50x errors are almost always an issue with the code-base, configuration, or the Stratus system. Focusing on those errors reduces alert noise and spam.
The errors are also triple checked before an alert is fired. We exclude certain conditions if detected:
- The site is suspended (sends a 504)
- Services were recently restarted (customer action causing a 504)
- A domain was removed from the URLs list by a customer
What happens when an alert fires ?
When an alert meets our conditions, the following sequence occurs:
- A ticket opens to the environment owner's account and any authorized users are cc'ed on the ticket.
- Our internal bot sends a Slack message available to our Support team.
- A support tech will try to verify the issue and attempt to resolve it, with an RCA follow up. The RCA will include recommend actions to avoid future outages.