IBM I Access AS400 connection in ADF

Question

I want to connect IBM I Access AS400 connection for data source in ADF

Answer 1

Hi @Godasi, Kumar R L D P,

it sounds like you’re trying to pull data from an IBM i (AS/400) system into Azure Data Factory. ADF doesn’t have a built-in “AS400” connector, but you have two solid options:

1. Use the DB2 for i (Db2 on IBM i) connector
2. Use the generic ODBC connector via a Self-hosted Integration Runtime (SHIR)

Here’s how you can get either approach working:

Spin up a Self-hosted Integration Runtime • Install SHIR on a Windows box that can reach your IBM i server. • Make sure this box can ping/telnet to your AS/400’s Db2 port (usually 446).

Option A – Db2 connector (recommended if you’re directly querying Db2 tables) • Install the IBM Data Server Driver Package or IBM i Access ODBC driver on the SHIR machine. • In ADF, go to Manage → Linked services → New → select “IBM Db2” connector. • Fill in: – Server name (your IBM i host) – Database name (your Db2 schema/library) – Port (typically 446) – Authentication: Basic (store user/pass in Key Vault if you like) • Test the connection and save.

Option B – Generic ODBC connector • Install & configure an ODBC DSN on the SHIR machine pointing to your IBM i database. • In ADF, create a new Linked Service → choose “ODBC”. • Select your SHIR, enter the DSN name, and supply credentials. • Test connection.

Build your Copy or Data Flow activity using the linked service you created.

If you run into errors, double-check: • Network/firewall rules between SHIR and AS/400 • That your Db2 driver/DSN is 64-bit and matches SHIR’s bitness • Credentials & permissions in Db2 • SHIR service account has access to the DSN (if using ODBC)

Hope that helps you get connected! If you hit specific errors, let us know what you see in the ADF Monitor tab or in the SHIR logs and we can troubleshoot further.

Reference List

• Troubleshooting IBM DB2 Connection Issues from Azure Data Factory https://learn.microsoft.com/azure/data-factory/connector-db2-troubleshoot
• Microsoft Connector for DB2 https://docs.microsoft.com/connectors/db2/
• Self-Hosted Integration Runtime Troubleshoot Guide https://learn.microsoft.com/azure/data-factory/self-hosted-integration-runtime-troubleshoot-guide?tabs=data-factory
• Azure Key Vault documentation https://learn.microsoft.com/azure/key-vault/ 

Answer 2

Azure Data Factory doesn’t have a native IBM i (AS/400) connector, but IBM i data can be accessed from Azure services using IBM-focused connectivity components and then exposed to ADF through supported interfaces.

From the provided information, Microsoft Host Integration Server and related components can connect to IBM i file systems and applications over SNA LU6.2 (APPC) or TCP/IP, and can surface that data to consumer tools such as BizTalk Server and Visual Studio via the Data Provider for Host Files and the Data Source Wizard. In an Azure integration scenario, the typical pattern is:

1. Use Host Integration Server / Data Provider for Host Files to connect to IBM i
   • Configure a data source to IBM i using the Data Source Wizard (Host Files):
     • In Data Source, select IBM i file system as the platform when targeting IBM i.
     • In TCP/IP Network Connection, specify:
       • Address or alias: IP or DNS name of the IBM i.
       • Port: TCP port for IBM i file system (default 446 for IBM i file system).
       • Optionally Certificate Common Name to enable SSL/TLS.
   • Alternatively, SNA LU6.2 APPC can be used instead of TCP/IP if that is how the IBM i is exposed.
2. Ensure IBM i communication parameters are correctly configured
   • For SNA / APPC connectivity, configure Host Integration Server parameters to match IBM i network attributes, for example:
     • Network Name (Local Node) ↔ IBM i RMTNETID.
     • Control Point Name (Local Node) ↔ IBM i RMTCPNAME.
     • Remote Network Name / Control Point Name ↔ IBM i RMTNETID and CP Name.
   • Use the IBM i communication parameter mapping table to align:
     • Mode: use QPCSUPP mode for IBM i connectivity.
     • Remote APPC LU, Local LU, and default LUs for users/groups.
3. Define IBM i access properties
   • In Definition properties for IBM i:
     • Select the appropriate IBM i Remote LU Alias and Local LU Alias.
     • Set Mode to QPCSUPP.
     • Provide IBM i User Name and IBM i Password.
   • Optionally restrict access by IP using the IP Address List:
     • Add specific IP Address and Subnet Mask entries for client workstations or services that will connect.
4. Expose IBM i data to ADF through an intermediate, supported source
   • Once Host Integration Server / Data Provider is configured and can read IBM i host files or data, surface that data via:
     • SQL Server (for example, by loading IBM i data into SQL tables), or
     • Another ADF-supported store.
   • Then, in Azure Data Factory, create a linked service to that intermediate store (for example, SQL Server) and use copy or mapping data flows to move or transform the data.
5. Alternative Azure-native integration pattern
   • For programmatic integration with IBM i applications (COBOL/RPG), Azure Logic Apps Standard includes an IBM i Program Call built-in connector action (Call an IBM i Program). This can be used to call IBM i programs over TCP/IP with parameters such as:
     • Server Name, Port Number, User Name, Password, Use TLS, Validate Server certificate, and Code Page.
   • Logic Apps workflows can then write the results into an ADF-supported data store, which ADF can consume.

In summary, to connect IBM i / AS400 as a data source for ADF, configure IBM i connectivity using Host Integration Server (or Logic Apps’ IBM i connector) to reach IBM i over TCP/IP or SNA, then land the data into a supported store (such as SQL Server) that ADF can connect to via a standard linked service.

---

References:

• Data Source Wizard (Host Files)
• Definition properties for IBM i
• IBM i connections
• Table of Parameters for IBM i Communication
• Integrate COBOL and RPG programs on IBM midrange systems with Standard workflows in Azure Logic Apps