nodetool bootstrap¶

Manages the bootstrap process for a node joining the cluster.

Synopsis¶

nodetool [connection_options] bootstrap resume

Description¶

nodetool bootstrap manages the bootstrap process when a new node joins a Cassandra cluster. Bootstrap is the process by which a new node receives data from existing nodes to take ownership of its assigned token ranges.

The resume subcommand allows resuming a previously interrupted bootstrap operation.

Bootstrap Context

Bootstrap typically happens automatically when a new node starts. This command is primarily used to resume interrupted bootstrap operations.

Subcommands¶

Subcommand	Description
`resume`	Resume a previously interrupted bootstrap process

Examples¶

Resume Interrupted Bootstrap¶

nodetool bootstrap resume

When to Use¶

After Bootstrap Failure¶

If bootstrap was interrupted due to network issues or node failure:

# Check current bootstrap status
nodetool netstats

# Resume bootstrap
nodetool bootstrap resume

After Node Restart During Bootstrap¶

When a node restarts before completing bootstrap:

# Node has restarted, resume bootstrap
nodetool bootstrap resume

# Monitor progress
nodetool netstats

Bootstrap Process¶

Normal Bootstrap Flow¶

1. New node starts and contacts seed nodes
2. Node receives cluster topology via gossip
3. Token assignment determined
4. Streaming begins from existing nodes
5. Node receives data for its token ranges
6. Bootstrap completes, node becomes UN (Up/Normal)

Interrupted Bootstrap¶

If bootstrap is interrupted:

Node status shows as UJ (Up/Joining) or similar
Data streaming incomplete
nodetool bootstrap resume continues from where it stopped

Monitoring Bootstrap¶

Check Status¶

# View bootstrap/streaming status
nodetool netstats

# Check node state
nodetool status

Monitor Progress¶

# Watch streaming progress
watch -n 5 'nodetool netstats | head -30'

Troubleshooting¶

Bootstrap Won't Resume¶

# Check node state
nodetool info | grep "Gossip"

# Check for errors
tail -100 /var/log/cassandra/system.log | grep -i bootstrap

Bootstrap Taking Too Long¶

# Check streaming throughput
nodetool getstreamthroughput

# Increase if needed
nodetool setstreamthroughput 200

Bootstrap Fails Repeatedly¶

# Check disk space
df -h /var/lib/cassandra

# Check network connectivity to seed nodes
nodetool status

# Review logs for specific errors
grep -i "bootstrap\|stream" /var/log/cassandra/system.log | tail -50

Best Practices¶

Bootstrap Guidelines

Monitor progress - Watch nodetool netstats during bootstrap
Ensure resources - Sufficient disk space and network bandwidth
One at a time - Bootstrap one node at a time
Check completion - Verify node reaches UN status

Bootstrap Considerations

Bootstrap can take hours for large datasets
Network bandwidth is critical
Other cluster operations may be slower during bootstrap
Don't add multiple nodes simultaneously

Command	Relationship
netstats	Monitor streaming progress
status	Check cluster state
rebuild	Rebuild from another datacenter
join	Join the ring