What Building Multiple Scalable DC/OS Deployments Taught Me About Running Stateful Services on DC/OS

" summary

Multiple scalable DC/OS deployments expose concrete limits and fixes: local volumes being pinned to hosts and “one task per volume” force either Portworx-style cloud‑native storage or manual workarounds; dynamic provisioning, rolling upgrades, and replica placement across failure domains get solved with automation that rebuilds clusters from code and with testing in a production‑like environment. Failure injection with Torpedo and Chaos Monkey, monitoring via Prometheus and Grafana, automated snapshots to S3, and handling a 100 GB growth or losing one versus three nodes all prove out and reduce outage time.

AI-written summary · about

" related talks

• 01h 35mBuilding Your First Stateful DC/OS ServiceBen Wood · MesosCon North America 2017
• 10m⚡ Lightning Talk: Navigating the Complexities of Stateful Workloads in KubernetesPadmarajan Narayanan · Data on Kubernetes Day Europe 2025
• 20mLightning Talk: Multi-Datacenter Strategies for Data ServicesJulian Fischer · Cloud Foundry Summit Europe 2017
• 35mDataStax Enterprise on DC/OS - Yes, It’s Possible; Customer Case StudiesRavi Yadav, Kathryn Erickson · MesosCon North America 2017
• 48mContainer: check! Aber wohin mit Persistent Data, Big Data oder Fast Data?!Johannes Unterstein · code.talks 2017
• 34mData Management in Kubernetes – Backup, DR, HAMiguel Rojas · stackconf 2022
• 33mSponsored Session: Open Source Cloud Native Storage Solutions for Stateful Applications from ...Rahul Sharma · ContainerCon @ Open Source Summit North America 2022
• 22mRunning distributed stateful system on Kubernetes - Beyond StatefulsetsBabu Srithar · Kubernetes Community Days Bengaluru 2023