Zūm Rails incident

Zum Rails Portal / API - Partial Outage

Resolved after 2 hours 18 minutes, affecting API and Portal, on October 1, 2026.

resolved Oct 01, 2026, 07:07 PM UTC

On October 1, between approximately 12:14 PM and 1:20 PM EDT, some Zūm Rails API requests were slow or failing. The issue started during a routine software update that coincided with a period of unusually high traffic, which left our API servers temporarily unable to handle incoming requests. Webhooks, and routine jobs such as EFT kept running throughout, from initial check ups no data was found to be lost. Our team restored full service by 1:20 PM EDT. We're now improving how updates are rolled out, to avoid the issue from happening again. We apologize for the disruption and will keep you updated as needed. Thank you for your understanding

Started
Oct 01, 2026, 04:49 PM UTC
Resolved
Oct 01, 2026, 07:07 PM UTC
Duration
2h 18m
Detected by Pingoru
Oct 01, 2026, 04:49 PM UTC

Affected components

APIPortal

Update timeline

  1. investigating Oct 01, 2026, 04:49 PM UTC

    Please be advised we are investigating an issue which has caused outages in both API and Production Portal. We will provide updates here as they arrive. Thank you for your understanding

  2. identified Oct 01, 2026, 05:57 PM UTC

    Please be advised that we've successfully identified the issue and are working on a fix. Merchants should start seeing some successful connections, but we will update when we have a 100% fix. Thank you for your understanding

  3. monitoring Oct 01, 2026, 06:08 PM UTC

    Please be advised that the fix has been implemented and we are monitoring the system closely. We recommended logging out of the portal for refresh and try again. As we review, please contact support if you're still experiencing issues. Thank you for your understanding

  4. resolved Oct 01, 2026, 07:07 PM UTC

    On October 1, between approximately 12:14 PM and 1:20 PM EDT, some Zūm Rails API requests were slow or failing. The issue started during a routine software update that coincided with a period of unusually high traffic, which left our API servers temporarily unable to handle incoming requests. Webhooks, and routine jobs such as EFT kept running throughout, from initial check ups no data was found to be lost. Our team restored full service by 1:20 PM EDT. We're now improving how updates are rolled out, to avoid the issue from happening again. We apologize for the disruption and will keep you updated as needed. Thank you for your understanding

Your product runs on more than Zūm Rails

Add the vendors you depend on and get one message when any of them reports trouble. Free for 3 vendors, no card.

Add your stack →