Okta incident

Latency Issue Affecting Google Imports

Minor Resolved View vendor source →

Okta experienced a minor incident on September 24, 2025 affecting okta.com cell 14 and Core Platform and 1 more component, lasting 36d 8h. The incident has been resolved; the full update timeline is below.

Started
Sep 24, 2025, 04:00 PM UTC
Resolved
Oct 31, 2025, 12:54 AM UTC
Duration
36d 8h
Detected by Pingoru
Sep 24, 2025, 04:00 PM UTC

Affected components

okta.com cell 14Core PlatformThird Party

Update timeline

  1. resolved Sep 24, 2025, 04:00 PM UTC

    On September 24, 2025, customers may have experienced increased latency and repeated retries when running Google imports, which led us to recommend pausing all imports via an admin banner. We have confirmed that the disruption was caused by our third-party provider, and the issue has since been fully resolved. There is no active impact, services are operating normally, and no further action is required from customers. We are collaborating with our provider to complete a comprehensive root cause analysis and will provide it here.

  2. resolved Oct 31, 2025, 12:54 AM UTC

    Timing: Incident start: September 24, 2025, at 09:00AM PDT Incident resolved: September 24, 2025, at 4:30 PM PDT Detection: On September 24, 2025, Okta monitoring detected a spike in latency impacting the Google Dashboard. Impact: Google Workspace API latency issues led to Okta performance degradation in job processing, and all customers using the Google Workspace API may have experienced latency issues. Root Cause: The Google Workspace API experienced a latency spike, causing delays in Okta job processing. Okta's system terminated outbound requests that did not receive a response within five minutes. During the Google Workspace API latency spike, the maximum expected wait time exceeded ten minutes, leading to connection terminations before results were received. This increased delay caused Okta jobs to back up, impacting our customers. Additionally, Okta's automatic retry mechanism for socket timeouts further compounded these delays. Google acknowledged this API issue and deployed a resolution.