Live outage feed

Outages and recoveries from the services you depend on.

Earlier incidents

40 incidents

Back to latest

Apr 25, 2026

  1. ZapierAI by Zapier action Fails with 500 error when a Knowlledge source is added

    The issue causing some AI by Zapier steps to fail with a 500 error when adding a knowledge source has been resolved. If you were affected by this issue, your Zaps with AI by Zapier steps using knowledge sources will now work again. To restore functionality: 1. Open the affected Zap and navigate to the AI by Zapier step 2. Before deleting the step, we recommend copying and saving your prompt to avoid losing any work. 3. Delete the step 4. Re-add the AI by Zapier step 5. Copy your previous prompt into the new step 6. Re-upload your knowledge sources Once you've completed these steps, your Zap will be ready to use. If you are experiencing this issue, you can connect with our support team here: https://zapier.com/app/get-help.

    Resolved

Apr 24, 2026

  1. GitHubPartial degradation for code scanning default setup and for code quality

    On April 20, 2026 between 10:28 UTC and 15:04 UTC GitHub experienced degraded service for code scanning default setup, code quality, and project boards. Repair of affected project boards additionally lasted until April 21, 05:04 UTC <br /><br />During this time, code scanning default setup and code quality analyses were not triggered on newly opened pull requests. Additionally, newly created issues were not appearing on project boards. <br /><br />The cause was a serialization error that prevented proper triggering of code scanning, code quality analyses, and project board updates. <br /><br />We mitigated the issue by deploying a fix, restoring event publishing for code scanning and code quality. For project boards, an additional code change was deployed to update event consumers, followed by a reindex of affected project items. <br /><br />We are working to prevent recurrence by strengthening our schema validations and improving monitoring for drops in publishing on critical hydro topics.

    Resolved

  2. HashiCorpIntermittent connection resets during azure resource management or external provider downloads

    Our Engineering team has mitigated the issue affecting AzureRM-based resource operations, and these systems are now operating normally. We’ve also implemented a workaround to address intermittent errors with other provider downloads, and all systems are currently stable. We are continuing to work with our third-party vendors on a permanent fix and appreciate your patience as we move toward full resolution.

    Resolved

  3. BubbleFull Service Outage

    A component of Bubble's core infrastructure suffered a sudden spike of traffic, and did not autoscale properly. This caused connections to timeout and dropped traffic. The team has adjusted scaling parameters and will continue to monitor.

    Resolved

Apr 23, 2026

  1. SnowflakeGCP - US Central 1 (Iowa): INC0155608

    Snowflake Engineering has completed the postmortem of this service incident. A detailed Root Cause Analysis \(RCA\) is available on the Snowflake Community site: [https://community.snowflake.com/s/article/INC0155608](https://community.snowflake.com/s/article/INC0155608)

    Resolved

  2. ChatGPT / OpenAIElevated errors in Codex

    GPT-5.5 is now available to all paid users in Codex. If you continue to see access issues, please update to the latest Codex version. We’ll share another update if needed.

    Resolved

  3. LeverScheduling Outage for some Microsoft Office 365 Accounts - Investigating

    This incident has been resolved. Incident Duration: April 21, 2026, 1:40 PM– April 23, 2026 11:00 AM PT Impact Summary: During this window, customers using Microsoft Office365 were unable to retrieve interviewer availability or schedule interviews. Existing data was not lost, but new scheduling actions failed until service was restored. Cause: An internal encryption key rotation was completed for several services, but a related dependency used by the calendar service was not updated. As a result, the calendar service was unable to read newly issued Microsoft access tokens, causing scheduling requests to fail for Office365-connected accounts. Mitigation: The calendar service was updated to use the correct dependency compatible with the new encryption keys. Once deployed, scheduling functionality was fully restored and verified in production. An RCA will be provided within 7-10 business days if applicable.

    Resolved

Older incidents