developweekly / blog

Blog

https://developweekly.netlify.app/

0 stars 2 forks source link

Highly Available NFS Server with Pacemaker & DRBD | developweekly #11

Open utterances-bot opened 3 years ago

utterances-bot commented 3 years ago

Highly Available NFS Server with Pacemaker & DRBD | developweekly

https://developweekly.com/docs/pacemaker/

UbHAadmin commented 3 years ago

I am following this guide to setup a 2 node Unbuntu 20.04 NFS cluster to use with kubernetes setup. How have you found the performance and reliability for the docker swarm? What do you prefer the haproxy loadbalancer vs the pacemaker virtual IP method. Have you made any tweaks to this config?

bgatesnccs commented 2 years ago

Using this guide i setup a HA NFS Server using Ubuntu 18.04, The "Soft Failover" using crm node standby alpha works as expected, but a hard failover by unplugging the ethernet, or powering the server down do not work as expected. i created a script that runs off the secondary node for now to run crm node standby alpha if the primary goes down, but its a pretty ugly fix. was wondering if you have experienced this just yet?

bgatesnccs commented 2 years ago

I am also using the VIP from Pacemaker