Day 67 building in public ⏩ Casual Friday. The agent read the CloudNativePG documentation for me, discovered that it supports full data import from another database instance during bootstrap, set up base backups and WAL archiving, configured automatic uploads to S3-compatible storage, and prepared all the GitOps scripts. So the actual migration became a breeze. Life is good! 😄

Day 66 building in public ⏩ The last few days were very busy, so I needed to slow down a bit. Today's update: I decided to migrate my database once again (hopefully for the last time) after discovering what seems to be a better way to manage it - via a Kubernetes operator, specifically CloudNativePG. The main reason is its excellent backup and restore capabilities. It handles the entire database lifecycle on autopilot and allows point-in-time recovery with minimal effort. Backups shouldn't just be created - they should be easy to restore as well. Today I also spent some time tinkering with Shelly circuitry and automating a few routines at home. It was genuinely fun and a pleasure to work with them.

商用K8s環境に妥協のない堅牢性を。「EDB CloudNativePG Cluster 1.29.0」リリース！【10分でわかる動画音声解説付】 GitOpsやIaCを推進しながら、ミッションクリティカルなDB基盤を維持する最前線の仕組みを、ぜひご確認ください！ ▼ 詳細はこちら edbjapan.com/cnpg-cluster-1-… #Kubernetes #CNPG

【オンデマンド配信】CNCFプロジェクト「CloudNativePG」に学ぶ、Kubernetes上の次世代DB運用戦略 K8s上でのPostgres運用のデファクトスタンダード、CNPGの強みや自動化の仕組みを事例を基に、デモを交えて徹底解説。 ▼ 詳細・無料視聴はこちらから edbjapan.com/webinar-cnpg-jp… #K8s #Postgres #EDB

we'll do that at some point 👍

Awesome. Would love to see a comparison/write-up on technical details with Multigres compared to @CloudNativePg

🚨 106 vulnérabilités critiques aujourd'hui CloudNativePG (9.9), OpenStack Mistral (9.9), Progress LoadMaster (9.6), Cisco Unified CM (8.6), TigerVNC (8.5), snapd Ubuntu (7.8) Et des dizaines d'applications métiers turques exposées. Article CRL : cybersearchlab.com/blog/read… #CRL

⚡ CNPG speed: Switchover completes in under 10 seconds. Faster PostgreSQL high availability than dedicated solutions - your DBA will be impressed! Running databases on K8s? 🤔 #Kubernetes #CloudNativePG #PostgreSQL #HighAvailability

🐘 CloudNativePG: Production-grade PostgreSQL on Kubernetes with automated failover, streaming replication, and backup management - all as K8s resources! #Kubernetes #PostgreSQL #Database #CloudNative

CloudNativePG is the Kubernetes operator that covers the entire lifecycle of a highly available PostgreSQL database cluster with a primary/standby architecture, using native streaming replication ➜ ku.bz/n6gpgcYtf

How to design #CloudNativePG for production failure by using plugin-based backups, WAL archiving, point-in-time recovery, snapshots, and PgBouncer so recovery is treated as the real operational priority. dylanmarkdacosta.medium.com/… #Kubernetes #PostgreSQL #CloudNative

Warning: A critical privilege escalation #vulnerability in #CloudNativePG allows an authenticated attacker to gain superuser privileges and execute arbitrary OS commands inside the primary pod. #CVE-2026-44477 CVSS(4.0): 9.4. #Patch #Patch #Patch

Last 3 days on atlas-platform - Total Chaos Observability. Atlas now has a full GitOps-managed observability stack running locally on Kubernetes. Completed this week: • Installed kube-prometheus-stack • Added Loki Promtail for centralized log aggregation • Added ServiceMonitors for workloads, PostgreSQL, PgBouncer, and ArgoCD • Defined SLIs SLOs for availability and latency • Implemented multi-window burn-rate alerts (Google SRE pattern) • Built Grafana dashboards for SLOs golden signals • Configured AlertManager routing • Wrote ADR-004 (Loki vs ELK decision) • Added SLO breach response runbook • Ran an availability GameDay to validate alerts and recovery paths Current Atlas stack: Kubernetes ArgoCD CloudNativePG PgBouncer Prometheus Loki Grafana AlertManager GitHub: github.com/prashant-zo/atlas… #Kubernetes #DevOps #Observability #Prometheus #Grafana #Loki #SRE #PlatformEngineering #GitOps #CloudNative #OpenSource #LearningInPublic

Yep 👀 Running PostgreSQL via CloudNativePG with PgBouncer pooler managed through ArgoCD.

⚙️Open Source Projects: CloudNativePG

Just landed: Learn Kubernetes weekly 185! My top picks: 🏥 Designing for Failure with CloudNativePG 🏠 Homelab Kubernetes Cluster with $0 Cloud Costs 🔐 From ACM to ALB on EKS Auto Mode ⚡ Zeropod: Scale to Zero Read it here: kube.today/issues/185

Todays I focused on database resilience inside Kubernetes. This is major Progress on - Atlas Platform Added PgBouncer connection pooling in front of PostgreSQL on Atlas and tested failover handling under real cluster conditions. What shipped: • Integrated PgBouncer using CloudNativePG Pooler CRDs • Added pooled write traffic routing for the application layer • Updated workloads to connect through the pooler instead of directly to Postgres • Validated Kubernetes reconciliation through GitOps with Argo CD • Ran failover testing by deleting the active primary pod • Verified automatic primary promotion standby recovery • Confirmed application traffic continued through the pool during recovery Atlas keeps moving closer to real production infrastructure: GitOps → HA PostgreSQL → WAL archiving → connection pooling → failover testing GitHub: github.com/prashant-zo/atlas… #Kubernetes #PostgreSQL #PgBouncer #CloudNativePG #ArgoCD #GitOps #DevOps #PlatformEngineering #CloudNative #LearningInPublic

Todays I focused on one of the most critical parts of running databases on Kubernetes: -- data durability and recovery. -- High availability keeps services running. Durability protects the data itself. On Atlas, I integrated PostgreSQL WAL archiving with CloudNativePG and MinIO to build a recovery pipeline directly into the GitOps workflow. Completed: • Configured continuous PostgreSQL WAL archiving to MinIO using S3-compatible storage • Added backup configuration inside CloudNativePG Cluster CRDs • Verified end-to-end WAL shipping from PostgreSQL → object storage • Tested archive generation and validation directly from the cluster • Extended ArgoCD drift handling for database CRDs • Validated reconciliation and rollout behavior across environments • Added recovery-focused documentation and operational notes A big takeaway from this work: Running databases on Kubernetes starts making sense when recovery is automated and infrastructure becomes reproducible. That layer of backup recovery changes the system completely. Building in public: github.com/prashant-zo/atlas… #Kubernetes #PostgreSQL #CloudNativePG #MinIO #DevOps #PlatformEngineering #GitOps #CloudNative #OpenSource #LearningInPublic

Todays I moved the platform deeper into production-grade infrastructure. The focus was database reliability inside Kubernetes. Replaced the temporary Postgres setup with CloudNativePG and started building a highly available database layer directly into the GitOps workflow. Completed: • Installed CloudNativePG operator through Argo CD • Integrated Helm Kustomize inside the platform GitOps pipeline • Added PostgreSQL cluster management through Kubernetes CRDs • Built high-availability database architecture (1 primary replicas) • Verified operator lifecycle through ArgoCD sync reconciliation • Debugged and resolved namespace manifest rendering issues during deployment • Added incident documentation for platform recovery and architecture decisions What I’m learning while building Atlas: Kubernetes gets very interesting when infrastructure starts managing infrastructure. Operators, CRDs, GitOps reconciliation, failure recovery—this is where platform engineering starts feeling real. Building in public: github.com/prashant-zo/atlas… #Kubernetes #DevOps #PlatformEngineering #CloudNative #GitOps #ArgoCD #PostgreSQL #CloudNativePG #OpenSource #LearningInPublic