Overview Of Azure On-Premise Data Gateway

In this article, we will explore an overview of one of service in Azure named as on-premise data gateway. This article will help you to understand what on-premise data gateway is, how to install and configure it with step by step approach. Let’s gear up.


If you are a novice to Azure cloud and want to learn what a logic app is, then go through my previous articles.

  1. An Overview of Cloud And Azure Cloud
  2. Pillars of Cloud And Azure Services

In this article, we will explore the overview of one of the services, named On-premise Data Gateway, in Azure.


This article will help you to understand what on-premise data gateway is, and in next articles, we will go through the steps to help you on how to install and configure it with a step by step approach. Let’s gear up.


  • Basic knowledge of Azure Services


  • When you migrate/redesign/develop your application to the cloud, then many times, you need to connect to the resources which are present in on-premise servers.
  • On-premise data gateway works with the following Azure resources.

    • Logic Apps
    • Power BI
    • Microsoft Flow
    • Azure Analysis Services


  • The on-premise data gateway service comes under the integration services category in Azure.
  • This gateway is nothing but a bridge between on-premise resources and Azure resource.

  • If we need to connect with on-premise resources; e.g., SQL or FTP data source etc., from Azure Services like Logic Apps, Power BI, Microsoft Flow, and Azure Analysis Services, then on-premise data gateway is one of the efficient and accurate solutions to make this integration possible.

  • It helps us to communicate between two environments securely by encrypting the data and credentials without modifying the firewall settings.
  • This gateway supports only Logic Apps, Power BI, Microsoft Flow, and Azure Analysis Services.

  • Transmission of data between these two environments happens through data compression and transports encryption.

How it works

  • Send a request from Azure resources to Azure service bus through the connector.
  • Connector builds a query with credentials and sends to Azure Service Bus (fully managed enterprise integration message broker service).
  • Azure service bus keeps requests in the queue for processing.
  • On-premise data gateway polls and receives the pending requests from Azure service bus.

    The gateway creates an outbound connection to Azure Service Bus. It communicates on outbound ports: TCP 443 (default), 5671, 5672, 9350 through 9354. We can also able to force to use HTTPs but that lead to reducing the performance.

  • The gateway decrypts the credentials and connects to the data sources with those credentials and sends a request to execute it.
  • On-premise data source executes the query and the results send back to the gateway
  • Finally, on-premise data gateway sends the results to the cloud service.
  • The below image by Microsoft shows detailed data flow of on-premise data gateway

Azure OnPremise Data Gateway 

What about troubleshooting?
    • On-premise data gateway maintains logs to help developers
    • Gateway service logs – C:\Users\PBIEgwService\AppData\Local\Microsoft\On-premises data gateway\<yyyyymmdd>.<Number>.log
    • Configuration logs - C:\Users\<username>\AppData\Local\Microsoft\On-premises data gateway\GatewayConfigurator.log
    • Event logs – We can get it under Application and Services Logs.

      • Open the event viewer.
      • Expand ‘event viewer (local)’
      • Go to ‘applications and services logs’
      • Select ‘on-premises data gateway service’

Availability and Disaster Recovery

    • For communication to Azure service from on-premise data gateway, the gateway should be in a running state.
    • When we install the gateway in the on-premise machine then we need to use one key called recovery key.
    • By using the recovery key, we can restore or move the gateway to other machines.
    • Recovery key also helps to migrate or recover the earlier settings of the gateway.


  1. A single gateway can connect to multiple data sources.
  2. It is not compulsory to install this on-premise gateway connector on the machine where the data source is present.
  3. To minimize latency, you can install the gateway as close as possible to your data source, or on the same computer, assuming that you have permissions.
  4. We can install it anywhere within your on-premise network, but destination data source must be accessible from a machine where the gateway is installed.
  5. Azure Application gateway and on-premise gateway are different things. Don’t be confused.


In this part 1, we have learned an overview of the on-premise gateway and its features.
In next article we will learn to execute the steps to install and configure it on on-premise machine.