Rate Limiting

Added in version 2.0.0

Limit Number of Requests per Second

Version 2.0.0 added the ability to limit the number of requests per second.

Interaction with Auto-Tune

As of version 2.13.1, --rate-limit can be used together with --auto-tune. When both flags are provided:

  • --rate-limit serves as a hard cap on the maximum request rate
  • --auto-tune will dynamically adjust the rate limit downward when errors occur
  • Auto-tune adjustments will never exceed the value specified by --rate-limit
  • When auto-tune attempts to remove the rate limit (after successful recovery), it will instead reset to the --rate-limit cap rather than removing it entirely

This combination is useful when you want adaptive rate limiting with a guaranteed maximum to ensure you never exceed a specific request rate.

Examples

Basic Rate Limiting

Limit number of requests per second, per directory, to 100 (requests per second will increase by 100 for each active directory found during recursion)

./feroxbuster -u http://localhost --rate-limit 100

Limit number of requests per second to 100 to the target as a whole (only one directory at a time will be scanned, thus limiting the number of requests per second overall)

./feroxbuster -u http://localhost --rate-limit 100 --scan-limit 1

Capped Auto-Tune

Combine auto-tune with a hard cap of 50 requests per second:

./feroxbuster -u http://localhost --auto-tune --rate-limit 50

With this configuration:

  • Auto-tune will start at 50 req/s (the cap)
  • If errors occur, it will reduce the rate (e.g., to 25, then 12, etc.)
  • As errors decrease, it will increase the rate back up
  • It will never exceed 50 req/s, even during recovery

Demonstration

rate-limit