Connection pool poisoning can escalate from row-level brownouts to complete system outages, emphasizing the need for proper PgBouncer configuration and monitoring.
About:
Artistic and resourceful professional with expertise in data platforms, cloud computing, and software engineering, valuing excellence and diversity.
Website:
Specializations:
Interests:
Outgoing Links:
Subscribe to RSS:
The article discusses the misconceptions surrounding Postgres database-level synchronous replication, clarifying that it is inherently asynchronous. It explains that while a client waits for a COMMIT to confirm replication, the pr...
The article discusses the author's follow-up experiments using Jepsen to demonstrate data loss in asynchronous replication configurations compared to synchronous replication. It details the methodology, including forced pod deleti...
The blog post announces a special event featuring Nikolay Samokhvalov, founder of PostgresAI, who will speak at the Seattle Postgres User Group. It discusses the importance of synchronous replication in PostgreSQL to prevent data ...
Adjusting the Postgres client_connection_check_interval setting can enhance transaction performance and connection management, preventing system outages during high loads.
This blog post discusses the results of running Jepsen tests on CloudNativePG, focusing on data durability and availability configurations. It compares different modes of operation (Max Performance, Max Availability, and Max Prote...
The author reflects on their experiences at KubeCon, focusing on improving Postgres deployment on Kubernetes by optimizing the use of the Linux page cache and reducing out-of-memory (OOM) kills. They discuss the potential for cost...
The author discusses resources for learning PostgreSQL, emphasizing the value of community engagement through platforms like Slack. They delve into the nuances of sanitizing SQL, particularly regarding sensitive data, and propose ...
The article discusses the importance of data safety in database management, emphasizing the need to prioritize requirements such as security, durability, and cost-effectiveness. It highlights the balance between keeping data safe ...
The post critiques the tech community's complicity in irresponsible AI use and calls for accountability and personal responsibility in the face of bullying in open-source projects.
The post discusses the CPU utilization differences between two AWS instances, Graviton2 and Graviton4, running identical workloads in a CloudNativePG playground. The Graviton2 instance shows 40% CPU utilization while Graviton4 sho...
The PASS Data Community Summit 2025 successfully concluded, continuing its legacy of attracting database professionals. The event has evolved to include a Postgres track, with volunteers from the Seattle Postgres User Group managi...
Bohan Zhang from OpenAI will speak at the Seattle Postgres User Group on November 6, discussing OpenAI's use of Postgres, including challenges, lessons from outages, and scaling strategies. The group seeks locations for meetups an...
Automated emails are spam and should be banned in open source communities to maintain trust, while AI tools can assist users responsibly.
The blog post discusses a recent meeting of a platform team that focused on the internal architecture of Postgres replication. It highlights key resources shared during the discussion, including a conference talk by Alexander Kuku...
The collation torture test results for Debian are uploaded. The test did not pick up any changes in en_US sort order for Bullseye or Bookworm. Buster has glibc 2.28 so it shows lots of changes. The postgres wiki had claimed that J...
The text discusses the recent patch committed by Christoph Berg, which closed Debian bug 927182, addressing the significant size increase of adding LLVM as a requirement to support Postgres JIT functionality. The patch makes LLVM ...
The text describes the process of testing loadBalancerSourceRanges with CloudNativePG on Azure Kubernetes. The author shares the steps taken, including creating an AKS cluster, installing CloudNativePG, creating client VMs, and de...
0Postgres Per-Connection Statistics
2025-01-09 •
The author discusses the wish list of observability-related things for community/open-source Postgres, including wait event counters, comprehensive tracking of CPU time, and per-connection statistics. They also highlight the recen...
0Challenges of Postgres Containers
2024-12-31 •
The text discusses the challenges of migrating enterprise workloads to Postgres containers, the similarities and differences between Postgres and other databases, the growing adoption of containers for Postgres, and the security, ...
0Kubernetes Requests and Limits for Postgres
2024-09-23 •
The text discusses the use of Kubernetes to run Postgres, the challenges and considerations involved in managing consolidation with requests and limits, and the impact of memory pressure on consolidation. It also highlights the ne...
The text discusses the four major components of being a good benchmark engineer, and the author's plans to attend a Postgres benchmarking meetup in Portland. The author also shares their thoughts on a recent podcast episode about ...
The text discusses the default sort order in Db2, SQL Server, Oracle, and Postgres. It explains the best choice of default DB collation, the quirks of different SQL databases, and the collation behavior in different countries. The...
The text discusses the major developments in Postgres extension discovery and distribution. It highlights the importance of extensions in extending the capabilities of the PostgreSQL core database server and the decentralized natu...