Zeotap incident

Audience Refresh Pipeline | P1 | Google Cloud Infrastructure Issue

Major Resolved View vendor source →

Zeotap experienced a major incident on May 17, 2026 affecting Audience Publisher, lasting 2h 28m. The incident has been resolved; the full update timeline is below.

Started
May 17, 2026, 01:55 PM UTC
Resolved
May 17, 2026, 04:23 PM UTC
Duration
2h 28m
Detected by Pingoru
May 17, 2026, 01:55 PM UTC

Affected components

Audience Publisher

Update timeline

  1. investigating May 17, 2026, 11:17 AM UTC

    2026-05-17 06:00 UTC We are aware of an issue affecting Audience Segment Refresh and Export for some customers. Segments that were scheduled to refresh or were manually triggered after 06:00 UTC today may be delayed or show a failed status. Our team is actively investigating. We have identified this as an issue with our cloud infrastructure provider, who has confirmed they are working on a fix. No customer data has been lost. All segment definitions and configurations remain intact. Next update: 12:00 UTC

  2. investigating May 17, 2026, 11:18 AM UTC

    ⏺ Update — 2026-05-17 11:30 UTC We have taken remediation steps on our side and our cloud provider (Google Cloud) has confirmed an infrastructure issue affecting a subset of compute instances in our EU region. Their team is actively working on resolution. Audience Refresh and Export jobs triggered during this window may show as failed. Once resolved, we will re-trigger affected jobs and no action will be required from customers. Next update: 12:00 UTC or sooner if status changes.

  3. investigating May 17, 2026, 12:53 PM UTC

    We are still yet to hear from the GCP team. We are actively looking at the issue.

  4. identified May 17, 2026, 01:02 PM UTC

    We have received mitigation steps from GCP. We are working on it.

  5. monitoring May 17, 2026, 01:55 PM UTC

    A fix has been implemented from google and we are monitoring the results. The audience is processing the pending refreshes

  6. resolved May 17, 2026, 04:23 PM UTC

    This incident has been resolved.