Simple definition
Ceph is a distributed storage platform that can provide block, file, and object storage from the same cluster.
Technical definition
Ceph relies on RADOS and daemons such as MON, MGR, and OSD. Data is distributed across OSDs according to cluster placement rules; services such as RBD, CephFS, and RGW then expose block, file, and object use cases respectively.
How it works / role
When a client writes data, Ceph determines its placement in the cluster and applies the configured replication or protection mechanism. Monitor, OSD, placement-group, network, and capacity health directly affect availability and performance.
What is it used for?
Build resilient distributed storage for virtual machines, files, backups, or object services without depending on a single storage array.
Practical example
A Proxmox cluster uses Ceph RBD for VM disks and keeps data available through failures that are tolerated by the configured replication policy.
Common issues
- An OSD is down or the cluster reports HEALTH_WARN/HEALTH_ERR
- Capacity is too high or unevenly distributed across nodes
- Network or storage latency reduces performance
- Placement, replication, or public/cluster network design is inadequate
Key takeaway: Ceph troubleshooting should correlate cluster health, capacity, networking, and OSD state rather than focusing on a single node.