High Availability Containers with VIP (Virtual IP)

on YouTube ↗15m

" summary

Three containers (web1, web2, web3) run CentOS 8 with nginx and keepalived, creating a virtual IP 172.17.0.200 bound to web1 by priority 200/100; a health-check script check_nginx.sh runs every 2s and kills keepalived if nginx stays down so the VIP fails over to web2. The demo shows curl and ping tests during restarts and observes seamless failover with no packet loss and successful failback when web1 recovers.

AI-written summary · about

" chapters (5)

1. 0:00Start
2. 0:15Architecture
3. 4:00Nginx changes
4. 5:11check_nginx.sh
5. 7:45testing

" related talks

• 33mCloud Native Load Balancing From ScratchFangyuan Li · KubeCon + CloudNativeCon China 2019
• 33mCloud Native Load Balancing From ScratchFangyuan Li · KubeCon + CloudNativeCon + Open Source Summit China 2019
• 36mCan on-prem be also a cloud Networking made easyMateusz Kowalski · DevConf.CZ 2023
• 37mUsing Istio to Manage the Cross-Regional & Cross-Cluster...Xi Ning Wang, Xiaozhong Liu · KubeCon + CloudNativeCon China 2019
• 06mHighly available DNS recursion with PowerDNSDave Kempe · linux.conf.au 2020
• 18mDeveloping a Custom Load Balancer to serve traffic at scaleSandeep Bhat · IndiaFOSS 3.0
• 20mEmpowering container-based NFVi with VPP on Arm ServersSong Zhu · Linaro Connect Hong Kong 2018
• 46mMicroservices and containers networkingLuca Relandini · Codemotion Milan 2017