Kangaroot open shift best practices - straight from the battlefield
0 likes883 views
This document discusses best practices for Day 2 operations on OpenShift infrastructure from experts with 20 years of experience in Linux and open source. It provides recommendations around designing highly available etcd clusters, implementing federated Prometheus monitoring across multiple clusters using Prometheus or Thanos, centralized logging with ElasticStack, persistent storage options, container registry considerations, backup solutions using Minio and Velero, application deployments with GitOps, and secrets storage with Vault. The company also provides 24/7 support for customers.
Ad
More Related Content
What's hot (20)
Similar to Kangaroot open shift best practices - straight from the battlefield (20)
![[오픈소스컨설팅] 프로메테우스 모니터링 살펴보고 구성하기](https://cdn.slidesharecdn.com/ss_thumbnails/oscprometheus-190422050231-thumbnail.jpg?width=560&fit=bounds)
Ad
More from Kangaroot (20)
Ad
Recently uploaded (20)
Kangaroot open shift best practices - straight from the battlefield
- 1. LET US BUILD YOUR OPEN SOURCE INFRASTRUCTURE
- 2. Designing OpenShift for Day 2 operations : best practices, straight from the battlefield Peter Dens – Rien Valkenaers
- 3. BELGIUM’S TOP LINUX AND OPEN SOURCE SERVICE PROVIDER VENDOR INDEPENDENT 20 YEARS OF EXPERIENCE IN OPEN SOURCE BROAD SET OF PARTNERS ABOUT US
- 4. ABOUT US
- 5. INTRODUCTION Day 2 operations ? < Mon – Tue – Wed – Thu – Fri – Sat – Sun >
- 6. INTRODUCTION Day 2 operations ? < Mon – Tue – Wed – Thu – Fri – Sat – Sun > ETCD design
- 7. INTRODUCTION Day 2 operations ? < Mon – Tue – Wed – Thu – Fri – Sat – Sun > Prometheus Federated Monitoring
- 8. INTRODUCTION Day 2 operations ? < Mon – Tue – Wed – Thu – Fri – Sat – Sun > Centralized logging with Elastic
- 9. INTRODUCTION Day 2 operations ? < Mon – Tue – Wed – Thu – Fri – Sat – Sun > Persistent storage for your applications
- 10. INTRODUCTION Day 2 operations ? < Mon – Tue – Wed – Thu – Fri – Sat – Sun > Container registry considerations
- 11. INTRODUCTION Day 2 operations ? < Mon – Tue – Wed – Thu – Fri – Sat – Sun > Backup
- 12. INTRODUCTION Day 2 operations ? < Mon – Tue – Wed – Thu – Fri – Sat – Sun > Application deploys and secrets storage
- 13. INTRODUCTION Day 2 operations ? < Mon – Tue – Wed – Thu – Fri – Sat – Sun > Yes, We work Weekends too ;)
- 14. ETCD DESIGN What is etcd ? • Persistence • Leadership Election Availability • Clustering with 5 instances (N-1/2) • Snapshots & Recovery in etcd v3
- 15. Monitoring Prometheus • Openshift has prometheus built in for monitoring • Multiple clusters ? • Offsite storage
- 16. Monitoring Prometheus Federation Pro : • Easy Setup • Central place for configuring alerts • No persistent storage needed in cluster Con : • Not horizontally scalable • Introduces Lag
- 17. Monitoring Prometheus Thanos Pro : • Global Query View • Unlimited Retention • Scalable Con : • Need persistent storage inside cluster • Need to configure alerts in every cluster
- 18. Logging • How to keep logs when container disappears • How to search logs across multiple instances • How to aggregate openshift infra, supporting infra and application logs è Elastic Stack
- 19. Persistent Storage • Databases : PostgreSQL, MAriaDB, MongoDB • Minio • Red Hat Openshift Container storage • Rook.io
- 20. Registry & scanning Built in registry is OK for small scale deployments • What if you have multiple clusters ? • Need LDAP auth ? • High Availability / Performance • Security Scans ? • Quay
- 21. Backup • Minio + std backup tool • Velero + minio
- 22. App Deploys Cluster up and running – now how do I deploy ? • Gitops ! Ansible or Helm • Store your application details in Git • Automate with CI/CD • Webhooks
- 23. Secrets storage How do I keep my secrets encrypted / out of SCM history ? • Hashicorp Vault !
- 24. Elastic Stack Container Registry Prom. Red Hat Quay VMware Harbor Prom. + Alertmanager Mattermost Email Phone OPS Git DEPLOY scrape BACKUP
- 25. Q & A