We've also made some improvements with DNS to help with random fetch failed errors in Sites.

We'll close this incident now, but continue to monitor for issues.

We identified several issues causing instability with Functions and Sites, resulting in 500 events from executions. We've released a number of fixes, and now see a significant reduction in these events. We're continuing to monitor the system closely.

We've added additional fixes and improvements and are seeing further declines in new errors for functions. We are continuing to actively identify any lingering failures and are continuing to monitor.

Errors have decreased but we are still continuing to make improvements for any more small edge cases that may cause functions or sites to fail.

We've improved our logging earlier, helping us point out an edge case. We've deployed a fix and are continuing to monitor and find areas for improvement.

Our team is currently investigating the issue.