Product Updates

What is Resilient Event Ingestion?

Resilient Event Ingestion is a feature for our event ingestion pipeline that prevents permanent data loss during authentication outages. When LaunchDarkly's authentication dependencies become unavailable, such as in the case of the October 20th AWS incident, Resilient Event Ingestion automatically writes raw SDK events to S3 instead of dropping them. Once services recover, we can replay these events and make them available.

This is a key part of our post-October 20th resilience improvements — ensuring that even during major provider outages, customer events are preserved and eventually processed.

How does it work?

Resilient Event Ingestion has two key features:

• Automatic failover: Any request that would have returned a 5xx error to the client is instead stored in S3 and acknowledged with a 202. No manual intervention is required to ensure preservation of data
• Replaying data: once authentication is recovered, we can replay the stored data to make it available in the LaunchDarkly app

Availability & Who's it for

This is an infrastructure-level improvement for all LaunchDarkly customers across all regions. No customer action is required.

What's new?

As a planned addition to the recent sequential testing release, we have made an important update to the current Guardian stats setup and config. Guarded releases now use absolute difference (not relative difference) when measuring regressions and displaying charts. Additionally, with our new improved stats approach (and reduced false positives) we no longer require customers to set manual metric thresholds and they have been removed from the configuration.

Why does it matter?

Charts won't show wide variability during the early stages (esp. when one value is zero).
Go use Guarded Releases! They're smoother than ever.

Docs

What's new?

It's now possible for Datadog customers that are not using OpenTelemetry to send observability data to power Guarded Releases, Experiments, and more.

Docs

You can now route user feedback directly to Slack using the LaunchDarkly Slack bot. When a user submits feedback on a feature flag, your team gets a real-time notification in the Slack channel where they already work complete with sentiment signals, feedback content, and the exact flag variation the user saw.

What’s new

• Route feedback to any Slack channel your team uses
• See sentiment, feedback content, and flag variation context right in the notification
• Set it up in seconds from the Feedback tab on any feature flag

Docs

A new Service Map dashboard is now available as part of the default Observability dashboards. This built-in view automatically discovers and visualizes relationships between services, helping you understand how traffic and errors move through your system.

The dashboard aggregates service-level request volume and error counts so you can assess dependencies and blast radius without building custom dashboards.

With service maps, you can:

• Identify upstream and downstream dependencies
• See which services generate the most traffic
• Pinpoint where errors originate and how they propagate

Docs