Skip to main content
liuzhilong62

liuzhilong62

PostgreSQL DBA & Blogger

PostgreSQL DBA. Writing about database internals, production cases, and source code analysis.

80 articles · 8 categories · updating

Featured

Case studies, internals, source code & annual reviews

More articles

PostgreSQL Operations Experience 2025

·5942 words·28 mins
A 2025 summary of PostgreSQL operations experience covering CPU, memory, IO, WAL, autovacuum, locks, partitioned tables, connection pools, statistics, and other core operational topics with practical tips and troubleshooting

Recent

My Blog is Live

·508 words·3 mins
From Jekyll to Hugo, from GitHub Pages to a custom domain, the journey of launching a blog with 76 articles and reflections along the way.

DBA, Writing, Learning and the Future in the AI Era

·5058 words·24 mins
A deep reflection on the essence of DBA operations and the limitations of AI Ops in the AI era, exploring non-technical barriers such as consensus, responsibility, and security, as well as changes in writing and learning

PostgreSQL Operations Experience 2025

·5942 words·28 mins
A 2025 summary of PostgreSQL operations experience covering CPU, memory, IO, WAL, autovacuum, locks, partitioned tables, connection pools, statistics, and other core operational topics with practical tips and troubleshooting

Case: Partition Data UPDATE Failure on 2026-01-01

·1367 words·7 mins
Analysis of UPDATE failure on new partitions without primary keys: partitioned tables under logical replication publications lack replica identity; PG11 publication mechanism combined with partition creation workflow results in new partitions missing both primary keys and indexes

Paper Deep Read: Anarchy in the Database

·1423 words·7 mins
A deep reading of 'Anarchy in the Database': a compatibility assessment of 400+ PostgreSQL extensions reveals 16.8% have mutual incompatibility issues; analyzing extension architectures across 6 databases

Case Study: Row Locks and LWLock LockManager

·2063 words·10 mins
Analyzing how high-concurrency updates on the same row cause massive row locks and LWLock LockManager waits, verifying through benchmarks that row locks bypassing the fastpath mechanism is the root cause of increased LWLock contention