Handling multiple persistant volumes for a Rethinkdb Docker Swarm -
i'm using rethinkdb
across cloud servers manually joining each server @ setup. i'm interested in moving on swarm
approach make scaling , failover easier. current approach cumbersome scale.
in current manual approach, create local folder on each server rdb
, mount volume store data. however, using swarm
means i'd need handle volumes more dynamically. each container need distinct volume keep data separate in case of errors.
any recommendations on how handle scenario? lot of tutorials i've seen far mention flocker manage persistent storage, can't see being handled dynamically.
currently struggling situation this. i've created temporary fix glusterfs.
what install glusterfs on docker nodes , mount folders. way data exists on nodes. less ideal if have lot of writes. slow because of way gluster treats data replication prevent data loss. solid, have issues speed.
in case suggest looking flocker. flocker volume plugin migrates data when container moves host. haven't had experience it, in case concept of flocker renders useless, need data in multiple containers on multiple hosts (read only) gluster came play
Comments
Post a Comment