Troubleshoot Ceph

When to use this skill

Capacity exhaustion: Ceph throttles at nearfull (85%), blocks backfill at
Stuck PGs: PGs unable to reach active+clean due to failed OSDs, map inconsistencies,
Slow requests: OSD operations exceeding osd_op_complaint_time (default 30s). Causes
MON quorum loss: Without majority agreement, cluster cannot accept map updates. Client
OSD flapping: OSDs repeatedly marking up/down due to heartbeat timeouts, network
Recovery storm: Multiple OSD failures trigger massive data movement that competes with
Any time the user reports a Ceph service behaving outside its expected envelope (elevated errors, latency, saturation, resource exhaustion, or unexpected restarts).
An on-call engineer is paging on a Netdata alert tied to a Ceph instance and wants a structured triage path.