General Settings#

💡 Tip: You can change these settings directly in the bgscan application instead of editing the TOML file manually.
Navigate to Settings → General Settings in the main menu to configure these options interactively using the TUI inspector.

Configuration file: settings/general_settings.toml

Global scan limits, probe rate limiting, pipeline execution mode, and buffer sizing.

Quick Reference#

Setting	Default	Description
`status_interval`	`1000`	Progress update interval in milliseconds
`stop_after_found`	`0`	Reserved; not enforced by the current engine
`max_ips_to_test`	`0`	Cap on IPs read from the input source
`pipeline_mode`	`"streaming"`	Execution mode for multi-stage scans
`max_ips_per_stage`	platform-dependent	Channel buffer size between streaming stages
`batch_size`	platform-dependent	Batch size for batch mode
`shuffled`	`true`	Randomize target order
`min_probe_duration`	platform-dependent	Minimum time spent per probe (ms)
`probe_per_sec`	platform-dependent	Sustained probe rate ceiling (token-bucket refill)
`probe_burst`	platform-dependent	Max probes in a burst (token-bucket capacity)

Status Interval#

status_interval = 1000

How often each stage emits a progress snapshot to the UI, in milliseconds. Valid range is 100 ms to 60 s. Lower values give smoother feedback and cost more CPU on very large scans.

Stop After Found#

stop_after_found = 0

Intended to halt a scan after N successful results. The field is loaded, validated, and editable in the inspector, but the scan engine does not act on it yet. Leave it at 0.

Max IPs to Test#

max_ips_to_test = 0

Caps how many IPs are read from the input source. 0 reads everything. Useful when sampling a large list or a wide IPv6 prefix, where scanning to completion is not practical.

Pipeline Mode#

pipeline_mode = "streaming"

Controls how targets flow through a multi-stage scan.

Value	Behavior
`streaming` (alias `parallel`)	All stages run concurrently. Successful IPs move to the next stage through in-memory channels.
`sequential` (alias `simple`)	Stage N finishes and writes its result file. Stage N+1 reads that file as input.
`batch` (alias `pipeline`)	IPs are chunked. Each chunk traverses every stage before the next chunk is read.

Unrecognized or empty values fall back to sequential at parse time, but the config validator rejects anything outside the list above and restores the default first. See Scan Pipeline for the trade-offs.

Max IPs Per Stage#

max_ips_per_stage = 100000

Buffer size of the channel connecting one streaming stage to the next. A larger buffer means a fast stage blocks less often when the following stage is slower, at the cost of memory. Each buffered entry is a single address.

The value is raised to the next stage’s worker count when that count is higher. Only used in streaming mode. Valid range is 1 to 10,000,000.

Batch Size#

batch_size = 5000

Number of IPs per chunk in batch mode. With more than one stage, the effective size is max(batch_size, highest worker count among stages after the first), so a batch is never smaller than the workers that have to drain it.

Shuffle Targets#

shuffled = true

Randomizes target order before scanning. Useful to spread load across subnets, avoid hammering one range, and get a representative sample when combined with max_ips_to_test.

Probe Rate Limit#

probe_per_sec = 500
probe_burst = 100

probe_per_sec and probe_burst form a token-bucket limiter applied to every probe across all stages. probe_per_sec is the sustained ceiling — tokens refilled per second. probe_burst is the bucket capacity — how many probes may fire back-to-back before the limiter blocks. Together they cap outgoing request volume regardless of Workers, which is how you avoid tripping upstream rate limits or DPI.

Both are validated. probe_per_sec ranges 1 to 1,000,000; probe_burst ranges 1 to 10,000. Out-of-range values are rejected, the default is restored, and the correction is logged. The effective default is platform-dependent (see the Quick Reference table).

Minimum Probe Duration#

min_probe_duration = 50

Enforces a floor on how long each probe takes, in milliseconds. After a probe finishes faster than this, the worker sleeps for the remainder before moving to the next target. Used to normalize timing-based side channels and to pace scans that would otherwise complete too fast.

Valid range is 10 ms to 5 s. The effective default is platform-dependent (see the Quick Reference table).

icmp_settings.toml — ICMP scan configuration
tcp_settings.toml — TCP port scan configuration
http_settings.toml — HTTP/HTTPS/HTTP3 probe configuration
dns_settings.toml — DNS scan configuration
xray_settings.toml — Xray outbound validation
writer_settings.toml — Result output configuration