"Back in the day" to get alerted for any Amazon related health issues required a combination for "screen scraping" the AWS Status Health Dashboard and polling the AWS API for health related issues. The "screen scraping" was constantly prone to issues and required continual maintenance. Everytime Amazon made a change to their webpage the status alerting would have issues. The AWS API was only mildly better, but still required you to poll serveral different APIs for status. Want to know if EC2 has an issue? Poll the EC2 API. Want to know if RDS has an issue? Poll the RDS API. AWS has over 100 services and each one requires a different check. This is certainly not a method that scales well.
We are happy to announce a major improvement in how we handle Amazon alerting. Now we are leveraging Amazon's new Personal Health Dashboard to get events. Those events include:
- Major outages with AWS in general
- Specific issues with AWS infrastructure that affects only you, e.g. an EC2 instance has failed, direct connect outage, etc.
- Scheduled maintenance events, e.g. an EC2 instance reboot, VPN upgrade, etc.
- General notifications made by AWS for your account, e.g. an upgrade to REDIS, a new version of storage gateway, etc.
Our customers also provided us feedback that the alert volume can be too high at times. We've taken that feedback and made some adjustments to how we do alerting. Here are the major changes:
- Major issues that effect your account will be sent every 30 mins.
- Scheduled maintenance alerts will be sent only once per day
- General notifications on your account will be sent only once per day for a maximum of 4 times per notification.
All of these new features are available immediately to our Managed Services customers. No additional fees or charges apply. Enjoy!
Comments