June 2020 Summaries

We recently released the first version of our OpsGenie integration, allowing users to leverage Checkly with the popular alerting and incident management platform more easily. Our plan is to add additional work to enhance the experience. We use public dashboards to monitor checks running against production and staging environments, setting up alerts via email, Slack, and OpsGenie for check priority. To integrate OpsGenie using webhooks, users can create an alert on failure and automatically close it when the check recovers. If Checkly goes down, we continuously monitor lower-level components like queue and notification services to prevent issues. We use Amazon CloudWatch alerts and AppOptics to handle spikes in incoming messages and ensure timely results. Our backend systems are separated into multiple services, allowing for independent deployment, operation, and functional decoupling.

Between June 5th, 2020 and June 12th, 2020, checks using async IIFE syntax were incorrectly marked as passed due to changes in security measures for Browser checks. The issue affected 18 active checks and was caused by a change that altered the way promises are handled in the runner. After a customer reported an outage not detected by their tests, the issue was identified and resolved quickly with a fix implemented just hours after it was reported. The fix ensured that the node process exits after executing all promises, rather than exiting immediately when the execution block finishes.