The ceph command‑line utility is the primary way to administer a Ceph cluster. It lets you query the cluster state, view detailed health information, and manage monitors (MONs), OSDs and placement groups (PGs).

Checking cluster status

ceph status

• Shows a concise overview of the cluster: number of MONs, OSDs, and any warnings or errors.
• Example output: cluster 1.2x/1.2x health OK followed by a list of services.

Viewing detailed health information

ceph health detail

• Provides a deeper health report, including specific OSD or PG issues.
• Useful when ceph status reports a warning but you need to know which component is affected.

Managing monitors (MONs)

# List all monitors
ceph mon stat

# Add a new monitor (replace  and 
)
ceph mon add  


# Remove a monitor
ceph mon remove 

# Force a monitor election (rarely needed)
ceph mon force-election

• Monitors store the cluster map; keep at least three for quorum in production.

Managing OSDs (Object Storage Daemons)

# List OSD health and usage
ceph osd stat
ceph osd tree

# Check OSD usage details
ceph osd df

# Bring an OSD up or down
ceph osd down 
ceph osd out 

# Remove an OSD (after marking it out)
ceph osd out 
ceph osd purge  --yes-i-really-mean-it

Managing placement groups (PGs)

# List PGs and their states
ceph pg dump

# Show detailed PG information
ceph pg dump pgs | grep 

# Repair a PG that is in a bad state
ceph pg repair 

Common operational commands

Frequently used flags

• -i – read JSON input from a file.
• --cluster-id – specify a non‑default cluster ID.
• --pool – limit the command to a specific pool.
• --format – change output format for scripting.

When to check next

If a command returns an error or unexpected output, verify:

1. The Ceph configuration files (/etc/ceph/*.conf) are correct.
2. All monitors are reachable (ceph mon stat).
3. OSD health (ceph osd stat) and PG health (ceph pg dump).

These checks will help isolate whether the issue lies with the monitor layer, OSD layer, or the pool configuration.