A comprehensive guide to DB2 for i SQL stored procedures and user-defined functions in 2026: CREATE PROCEDURE syntax, IN/OUT parameters, cursor loops, calling from RPG and Node.js, scalar and table UDFs, and external stored procedures backed by ILE RPG programs.
A practical guide to using AI and large language models to document legacy IBM i RPG programs in 2026: extracting source code from IBM i, calling GPT-4o and IBM Granite via API, prompt engineering for RPG documentation, building a batch documentation pipeline, and creating a semantic search knowledge base over your entire IBM i codebase.
A complete guide to IBM i system startup and shutdown procedures in 2026: understanding the IPL sequence, configuring QSTRUP, starting TCP/IP and subsystems with STRTCP and STRSBS, performing controlled shutdowns with ENDSBS and PWRDWNSYS, scheduling maintenance IPLs, and diagnosing common startup failures.
HTTPAPI by Scott Klement is the standard open source library for making outbound HTTP calls from IBM i RPG programs. This post covers installing HTTPAPI and YAJL, making GET and POST requests, parsing JSON responses, implementing the OAuth 2.0 client credentials flow from ILE RPG, and managing SSL certificates on IBM i.
DB2 for i triggers fire automatically on INSERT, UPDATE, and DELETE, enforcing rules that application code cannot bypass. This post covers SQL and external trigger types, BEFORE and AFTER timing, transition variables, writing audit tables, calling RPG programs from external triggers using ADDPFTRG, and practical audit and validation patterns.
A complete guide to RPGUnit for ILE RPG in 2026: install the framework, write test cases and test suites for RPG service programs, run tests with RURUN, parse TAP output, and integrate automated RPG testing into a Bob or Jenkins CI pipeline.
A practical IBM i application modernisation roadmap for 2026: compare six strategies from screen transformation and API wrapping to selective rewrite and full migration, apply the strangler fig pattern, assess your RPG estate, and build the business case for phased modernisation.
Vector search and text embeddings bring semantic search to IBM i in 2026 — converting DB2 for i text fields into float arrays via OpenAI or watsonx.ai, storing vectors as VARBINARY or JSON in DB2, computing cosine similarity in SQL, and building RAG pipelines that ground LLM answers in real IBM i business data. This post covers the full embedding pipeline from generation to retrieval.
DB2 for i replication in 2026 is built on IBM i journals — the same journal receivers that underpin HA tools and Kafka CDC also power Q-Replication and MIMIX. This post explains journal-based logical replication, the Q-Replication Capture and Apply architecture, what MIMIX adds for HA and DR, and how to design a replication strategy that matches your latency targets and data distribution needs.
A hands-on guide to connecting IBM i with Apache Kafka in 2026 — running kafkajs and confluent-kafka-python producers in PASE, journal-based change data capture using QSYS2.DISPLAY_JOURNAL, Kafka Connect via IBM MQ bridge, consuming Kafka messages to write back into DB2 for i, and practical event-driven architecture patterns for IBM i integration.