DevOps Taiwan - Monitoring Tools 大亂鬥 - Prometheus

Transcript

1. Monitor Tools 大亂鬥 Prometheus 篇 Adam Chen
2. Agenda ▪ Prometheus ▪ Percona Monitoring and Management ▪ Demo 2
3. Hello! I am Adam Chen Networking / DevOps / DBA ▪ AWSUG TW Board Member ▪ Taipei HashiCorp User Group Staff You can find me at adaam 3
4. About Prometheus 普羅米修斯
6. 6 豆知識： Prometheus 其實是源自希臘文，在希臘神話中，是提坦神族的神明之一， 名字的意思是「先見之明」 (forethought)
7. What is Prometheus? ▪ Monitor System ▪ Timeseries Database ▪ Pull based ▪ Alert System (with AlertManager) ▪ Metrics Collection & Storage ▪ Dashboarding / Graphing / Trending ▪ Focus on OS monitoring 7
8. 8 Source: https://www.slideshare.net/KasperNissen1/monitoring-with-prometheus
9. 9 Short-Live Job Long-Live Job Grafana Alert Manager
11. Pull v.s. Push ▪ Less configuration ▪ Horizontal monitoring ▪ Easy for scale ▪ More flexible ▪ Simple HA 11
12. Data Model (Label) 12 <Metric Name>{label1=value1, label2=value2, ...}
13. Prometheus NOT do ▪ Raw log / Event collection ▪ Durable long-term storage (can use remote storage) ▪ Automatic horizontal scaling 13
14. Percona Monitoring and Management
15. What is PMM ▪ Open-source platform for managing and monitoring MySQL and MongoDB performance. ▪ Based on Prometheus, Consul, Grafana, Application exporters 15
16. Why PMM ? 16 ▪ Easy to configure (Compare to Prometheus Config) ▪ 3 Step: Install, config, use ▪ Pre-installed Grafana Template
18. Demo Use Percona Monitoring and Management (PMM)
19. PMM Architecture 19
21. 21 Thanks! Any questions? You can find me at ▪ adaam ▪ Adam dot cws@gmail dot com