Fabric Pipeline copy activities (source: on-premises MariaDB via on-premises data gateway) are failing.

Question

Test Connection on the Copy Activity returns:

{"errorCode":41,"parameters":[],"activityId":"asdsdx-asdv32144-xasf2-43-asd","message":"An internal error has occurred. Please retry the operation at a later time. If the issue persists, please contact Microsoft support for further assistance.","stackTrace":"Microsoft.PowerBI.DataMovement.Pipeline.Diagnostics.GatewayClientErrorResponseException;Received error payload from gateway service with ID 347106: FabricIntegrationRuntimeServiceHost is not started..;"}

Separately, running the pipeline returns:

"The FabricPipelineWorker.exe or Fabricworker.exe on the On-premises data gateway is unable to establish a connection with the Fabric Pipelines services. Please verify your local network settings..."

Environment: single on-premises data gateway, no proxy configured on this network.

Timeline: issue appeared around July 25–26, 2026, coinciding with a Windows update installed on the gateway server around that date (July 14, 2026 Windows security/.NET cumulative updates, which we believe reached our server via WSUS around the 25th–26th).

What we've already tried:

• Confirmed no proxy is used on this network — ruled out proxy misconfiguration in FabricPipelineWorker.exe.config.
• Ran the built-in Network ports test in the gateway app — 100% success on all standard ports/endpoints.
• Uninstalled and reinstalled the on-premises data gateway to the latest available version — issue persists.
• Confirmed FabricIntegrationRuntimeServiceHost.exe and its config file exist on disk under FabricIntegrationRuntime\5.0\Shared, but this component does not appear as a registered service in Windows services.msc, and we cannot confirm whether/how it's supposed to run.

Answer 1

Hello Oleksii Roiko,

Welcome to the Microsoft Q&A and thank you for posting your questions here.

I understand that your Fabric Pipeline copy activities (source: on-premises MariaDB via on-premises data gateway) are failing.

The failure is occurring inside the Fabric Pipeline runtime on the on-premises data gateway, before the database connection is reached. That's not your MariaDB. Since the key errors are: FabricIntegrationRuntimeServi*eHost is not startedand FabricPipelineWorker.exe or Fbricworker.exe ... is unable to esablish a connection with the Fabri* Pipelines services.

What you can do to fix is to:

• Explicitly allow outbound HTTPS from the gateway server to: *.frontend.clouddatahub.net TCP 43
• Ensure TLS/SSL inspection or endpoint security is not blocking Fabric Pipeline runtime traffic.
• Restart the supported on-premises data gateway service: ne* stop PBIEgwService net start PBIE*wService
• Retest the Fabric Pipeline Copy Activity.
• If the issue persists, export gateway logs and check Event Viewer for blocked startup, TLS, policy, or certificate errors involving FabricPipelineWorker.exe, Fabricworker.exe, or FabricIntegrationRuntimeServiceHost.exe.

After allowing *.frontend.clouddatahub.net:443 and restarting the gateway service, the Fabric Pipeline runtime should be able to connect to Fabric services. If the same error continues after these validations, then this becomes a Microsoft Fabric Support/Product Group escalation with gateway logs, Event Viewer logs, activity ID, gateway version, and exact UTC failure time.

Use the below resource links for more reading and action/steps:

• https://learn.microsoft.com/en-us/data-integration/gateway/service-gateway-communication
• https://learn.microsoft.com/en-us/data-integration/gateway/service-gateway-tshoot
• https://learn.microsoft.com/en-us/data-integration/gateway/service-gateway-restart
• https://learn.microsoft.com/en-us/fabric/data-factory/how-to-access-on-premises-data

I hope this is helpful. Please! Do not hesitate to let me know if you have any other questions, steps or clarifications.

---

Please do not close the thread by upvoting and accepting the answer if any part of it is helpful.

Answer 2

Hi everyone, reinstalling the gateway fixed it

Answer 3

The errors occur in the gateway’s Fabric pipeline runtime before MariaDB is contacted: the service host is not running and the worker cannot reach Fabric. Since the port test passed and the gateway reinstall failed, do not change the MariaDB connector or manually register the executable as a Windows service; Guidance does not prescribe that repair. Restart the data gateway service and server, then explicitly allow outbound HTTPS from the gateway machine to *.frontend.clouddatahub.net and the gateway FQDNs. Reproduce, export logs from the gateway app’s Diagnostics page, and check Event Viewer > Applications and Services Logs > On-premises data gateway service for process-start, TLS, or policy failures. If endpoint security blocked the pipeline binaries after the Windows update, allow them and retest. Otherwise open a Microsoft Fabric support case with the activity ID, failure time, gateway version, and exported logs; do not roll back the update without matching log evidence.