Docker/Kubernetes setup doesn’t work as expected

We’re using WP 4.9 + PHP 7.1 in docker on top of Kubernetes. We’re doing CI/CD with Jenkins. We have images dir and ‘wp-supercache’ dirs in EFS, and they are mounted automatically in the containers via NFS. We also have have database cache plugins enable.

We saw many posts about people sharing wp-content via NFS/EFS in its entirity, but they are not versioning the containers properly. If all content is inside EFS, the containers works like ‘dumb’ instances of wordpress, so you don’t have control of what is new at content level, and most important: you don’t have control in a rollback situation.

So, in our setup, we’re experiencing problems like:
– Session logout: I log into wp-admin. Then I click in some link (then ELB points traffic to a different container), then my session goes down (because it’s a different container).
– Mis-configurations: we enable a plugin, then few seconds after, it’s disable (because ELB pointed to a different container).

What is the best way to control these situations in a high-availability environment (many instances of wordpress) as Kubernetes?

• This topic was modified 6 years, 10 months ago by fwynyk.
• This topic was modified 6 years, 10 months ago by fwynyk.