How to set up redundant Fess containers

(from github.com/kujiy)
Hi, thank you for sharing your excellent project. We’ve used the docker image over a year. It’s literally excellent.

We’d like to make our Fess container redundant. What is the best solution? It will be nice that we can build a master-slave or multi-master system. Reducing duplicated crawl is better also. Does anybody give us a hint?

(from github.com/marevol)

  1. Setup/Configure Elasticsearch cluster with plugins (multiple nodes)
  2. Setup multiple Fess instance connected to the above Elasticsearch cluster
  3. Set a node name to scheduler.target.name in fess_config.properties
  4. Edit Scheduled job with Target (Job is executed if Target matches scheduler.target.name)

(from github.com/kujiy)
Hi @marevol ,
Thank you for answering. I should have asked at docker-fess project. Does the Target feature mean Fess has already supported the elastic search cluster? That’s awesome. I think it’s hard to apply the instruction to the docker entrypoint at this time. Hense we’ll just try it manually. I may ask you again during the challenge. Thanks.

(from github.com/marevol)
Fess supports Target feature.