Disk Space Exhaustion from Snapshot Accumulation
criticalOld snapshots from backups or schema changes accumulate on disk, silently consuming storage until reaching 90%+ capacity. This blocks compaction and eventually causes node failures with disk-full errors.
Monitor cassandra_storage_load and cassandra_sstable_live_disk_space_used trending toward disk capacity limits (>90% full). Cross-check with cassandra_compaction_tasks_pending suddenly spiking or stalling. Verify with system-level disk usage commands (df -h) showing unexpected consumption in data directories.
Run 'nodetool clearsnapshot' to remove old snapshots immediately. Review snapshot retention policies and automate cleanup. Check commit log directory size and move commit logs to separate drives if possible. Use 'nodetool compact' to reclaim space from deleted data. If disk space cannot be recovered, add storage capacity or scale cluster horizontally before failure.