Robust process management on PostgreSQL
Developed by Microsoft, pg_durable is a library designed to manage durable execution processes on PostgreSQL. Written in Rust, the tool enables complex workflows to run within the database in a fault-tolerant and persistent manner.
Updates
- September 12, 2026: Stars 2,800 → 2,811, latest release v0.2.8 (September 11, 2026).
- September 2, 2026: Stars 2,781 → 2,800, latest release v0.2.7 (September 1, 2026).
- August 24, 2026: Stars 2,716 → 2,781, latest release v0.2.6 (August 24, 2026).
- August 2, 2026: Stars 1,580 → 2,716, latest release v0.2.5 (July 30, 2026).
What you get
- It manages workflows within the database in a fault-tolerant and persistent manner.
- In case of crash or interruption, it continues operations from the last checkpoint.
- It runs directly on PostgreSQL without requiring additional infrastructure.
Installation
CREATE EXTENSION pg_durable;Running it
SELECT df.start(
'SELECT id FROM documents WHERE processed = false LIMIT 100' |=> 'batch'
~> 'UPDATE documents SET processed = true WHERE id = ANY($batch)'
);If you don't write code
I want to create a workflow using the pg_durable plugin on PostgreSQL. How should I configure the df.start() function to manage a fault-tolerant and persistent process within the database? How can I create a structure that processes data and can continue from where it left off in case of error, using the ~> and |=> operators that connect SQL steps? Please explain this process with examples using SQL commands.
Related dictionary terms
Links
TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-06-08: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. This page was machine-translated from the Turkish original · the Turkish version prevails.