High Availability GlusterFS DRBD Ceph et al

Discussion in 'HOWTO-Related Questions' started by Chrys, Oct 30, 2019.

  1. Chrys

    Chrys Member

    Hi guys.
    After successful installation of ISPConfig on several hardware servers (for cold redundancy) I am at a stage of implementing storage high availability for a specific part of the server (which uses raid1), on several other servers. Emails and documents drafted by people are critical so there cannot be any downtime nor loss. So basically the suggested set up would entail a mail and file server all being kept under one machine under raid 1 whose storage will be replicated to a second machine which can take over in time of a psu failure or other downtime issue. I know this can be set up in the form of a full live HA system with automated kick in should one server go down, but extra administration will defeat the purpose so all we need for now is only storage data replication of emails and files to the second server.
    The question is which of the above applications GlusterFS is more appropriate and most reliable under these circumstances. Reliability is the most important aspect of it all. Raid 1 will remain on each server for extra availability, so two or more raid 1 machines with only replicated storage over the network is what we need.
    Last edited: Oct 30, 2019

Share This Page