Previous incidents

The issue affecting conversation processing has been resolved. Our team is continuing investigate the RCA.

The issue affecting conversation processing has been resolved.
Impact Window: 13:55 PST – 15:02 PST
We’ll continue to monitor and will provide a preliminary post-incident review upon request within 48hrs.

The issue affecting OTP email delivery has been resolved by our auth vendor and we are seeing full recovery across our customers. We’ll continue to monitor and will provide a preliminary post-incident review upon request within 48hrs.

We are responding to an elevated rate of tool errors. We've rolled back and have seen error rates return to normal, but are continuing to monitor to confirm recovery.

This issue has now been resolved and we have fully recovered.

This issue has now been resolved.
Impact Window: 12:35pm – 1:25pm ET

Cause: An outage with a feature flag provider caused temporary impact on some features, which included disabling Duet. We'll continue to monitor and will provide a preliminary post-incident review within 48hrs. We will follow up with customers who may have experienced any additional impact. Thank you for your patience and understanding.

Errors have subsided as of 1:27 PT. For affected customers, we will follow up with a root cause analysis and description of impact.

We have implemented a mitigation to enable jobs processing. Scheduled jobs are now executing as expected, and any delays to asynchronously processed tasks should clear shortly. Conversations across all channels were unaffected throughout the incident.

The service disruption affecting chat and voice channels has been resolved — all channels are now operating normally. We identified the issue, deployed a mitigation, and confirmed that latency and error rates returned to normal levels as of 11:05am PDT.

We apologize for the disruption, and for impacted customers, we will follow up with a root cause analysis (RCA)

Starting at approximately 8:15am PT this morning, a platform update caused the chat widget to render in a manner that may have obscured its content for a subset of customers. We have a fix in place as of 8:57am PT and are verifying recovery across affected customers. Widgets are rendering normally again, and we're continuing to monitor to confirm full resolution.