Close

Rebalance Still In Progress

A project log for Raspberry Pi Ceph Cluster

A Raspberry Pi Ceph Cluster using 2TB USB drives.

Robert RouquetteRobert Rouquette 11/13/2020 at 02:462 Comments

The OSD rebalance is still running, but it did free up enough space that I was able to finish migrating my media files.  The total raw space currently in use is 66% of 54 TiB.

Discussions

Robert Rouquette wrote 01/25/2021 at 21:07 point

The OSD daemons typically need around 1.5 GB of RAM per TB of disk to remain responsive under all conditions.  The RPi cpu load (actual processing time, not iowait) is fairly low even during recovery ops, and the quad core ARM processor seems to be more than adequate.

  Are you sure? yes | no

Diomede wrote 11/15/2020 at 20:27 point

Hi Robert, I came across your project because I was thinking of implementing pretty much the same thing but I want to add Kubernetes on each node. Looking at the component you've used I've noticed that the RPis with the HD attached are all RPi 4/8GB is that because the 4GB ones cannot handle the load? How heavy is the ceph load for the RPi?

Anyway, really good job! I love that cabinet :D

  Are you sure? yes | no