Integration Service Customers Facing intermittent failures in US

Update timeline

1. identified Jul 07, 2026, 11:45 AM UTC

   Integration Service Customers Facing intermittent failures during connection creation and activities execution in US and Europe regions

2. identified Jul 07, 2026, 01:04 PM UTC

   The issue causing degraded performance affecting Integration Service in the US and Europe regions has been identified, and mitigation efforts are currently underway.

3. identified Jul 07, 2026, 02:30 PM UTC

   Fix has been applyed we are currently monitoring the service to ensure stability

4. monitoring Jul 07, 2026, 04:09 PM UTC

   Fix has been applyed we are currently monitoring the service to ensure stability

5. monitoring Jul 07, 2026, 05:36 PM UTC

   Fix has been applyed we are currently monitoring the service to ensure stability

6. identified Jul 07, 2026, 07:04 PM UTC

   we are seeing some intermittent failures. We are still investigating further on this.

7. identified Jul 07, 2026, 07:17 PM UTC

   Issue resolved in Us region. we are seeing some intermittent failures in Europe region. We are investigating further on this.

8. identified Jul 07, 2026, 07:18 PM UTC

   Issue resolved in Us region. we are seeing some intermittent failures in Europe region. We are investigating further on this.

9. identified Jul 07, 2026, 09:13 PM UTC

   We are working to deploy the fix

10. identified Jul 07, 2026, 10:37 PM UTC

    We are working to deploy the fix

11. identified Jul 08, 2026, 12:03 AM UTC

    We are working to deploy the fix

12. identified Jul 08, 2026, 05:45 AM UTC

    We are working to deploy the fix

13. identified Jul 08, 2026, 04:50 PM UTC

    Issue is now resolved in Europe. We are seeing some errors in US and we are continuing to on the fix.

14. monitoring Jul 08, 2026, 05:50 PM UTC

    Issue is mitigated now. we are now monitoring the issue.

15. resolved Jul 08, 2026, 05:58 PM UTC

    Issue is resolved now.

16. postmortem Jul 15, 2026, 07:15 PM UTC

    ## Customer impact Between July 7, 2026 at 10:26 am UTC and July 8, 2026 at 5:58 pm UTC, a subset of customers experienced intermittent failures when creating Integration Service connections and when running activities that depend on those connections. Customers encountered intermittent HTTP 502 errors and connection reset failures during connection creation and execution. The impact affected customers using Integration Service in the United States and Europe regions. The incident lasted approximately 31 hours and 32 minutes \(This was not a complete outage. Requests continued to succeed intermittently, and many failed requests were successful upon retry.\). The U.S. region initially recovered before intermittent failures continued in the Europe region, and residual errors were later observed again in the U.S. region. ## Root cause The root cause was exhaustion of available outbound connection capacity on shared infrastructure. This capacity is shared across multiple UiPath services, and a high aggregate volume of concurrent outbound connections exhausted the available capacity. Once the limit was reached, new outbound connections could not be opened reliably. As a result, the Integration Service components responsible for connection creation and activity execution intermittently failed to establish new outbound connections, which surfaced to customers as intermittent 502 responses and connection reset errors. The condition was not caused by any single service; it resulted from the combined outbound connection volume of multiple services sharing the same infrastructure. ## Detection The issue was detected through automated monitoring and customer reports on July 7, 2026 at 10:26 am UTC. The incident was declared customer-impacting within approximately one minute of the initial alert. Customer support cases were also reported later that morning, confirming that the failures were visible to affected users. ## Response At 11:23 am UTC on July 7, our engineering team identified that a high volume of external connections on shared infrastructure was blocking new outbound connections once capacity was reached. At 11:39 am UTC, additional capacity for the affected Integration Service component in the Europe region was requested. By 12:26 pm UTC, connection reset errors had decreased, and the team continued monitoring. At 12:49 pm UTC, two overloaded infrastructure components were identified for remediation to reduce the immediate impact. At 2:30 pm UTC, the public status page was updated to indicate that a fix had been applied and that the service was being monitored. At 4:09 pm UTC, the incident was marked as mitigated. During continued monitoring, additional infrastructure capacity was added at approximately 5:48 pm UTC. At 7:17 pm UTC, the U.S. region was reported as resolved, while intermittent failures continued in the Europe region. The incident was reopened for continued investigation. On July 8, 2026, a connection pooling change was deployed to the U.S. region at 3:09 pm UTC to improve connection efficiency; however, errors continued afterward. Full recovery was achieved by remediating the affected infrastructure to restore outbound connection capacity. Europe was reported resolved at 4:50 pm UTC. Remaining U.S. errors then subsided. The incident was marked mitigated at 5:50 pm UTC, and full recovery was confirmed when the incident was marked resolved at 5:58 pm UTC. ## Follow-up The following actions are planned or recommended based on the investigation findings: ### Additional Recommendations 1. Improve connection efficiency \(connection reuse\) to reduce the number of outbound connections created under load, with careful validation before rollout which would prevent incidents like these in future 2. Improve detection and mitigation mechanisms so that outbound connection capacity pressure is identified and remediated before it causes customer-facing connection creation or activity execution failures.