Netscan
Netscan is a fast TCP port scanner for IPv4 and IPv6 networks. Netscan is written in Go and is intended for normal users to run on Linux, Windows or Mac systems. The goal of netscan is to find open server ports quickly for service tracking and later testing and deep analysis using tools such as Nmap and Nessus. Netscan results are intended to be loaded into a relational database.
Time to scan 1 million hosts for 45 common ports (using 32K semaphore)
$ time ./netscan 172.txt ports.txt > 172.results.txt
real 44m3.894s
user 24m15.180s
sys 48m20.012s
To get or build netscan
$ go get github.com/62726164/netscan
$ go build netscan.goTo run netscan
$ netscan ips.txt ports.txt > results.txtNotes
- To scan large networks for many dozens of ports, consider increasing the number of open files for the user who runs netscan and also increasing the size of the counting semaphore in the netscan code. Experiment to find a suitable number on your scanner systems for your particular networks. Here's an example on Linux systems in the file /etc/security/limits.conf:
user_name soft nofile 150000
user_name hard nofile 150000- The file ips.txt must be a plain text file with one IP address or hostname per line. It must contain one or more entries. Netscan does not support CIDR notation. See the gen_ips.py script (in this repository) as an example of how to generate a list of IPs based on a CIDR prefix. Here's an example ips.txt file:
192.168.1.54
2001:468:c80:c111:0:401a:d2f8:de6c
hostname.your.domain- The file ports.txt must be a plain text file with one TCP port number per line. It must contain at least one port number. Typically, ports.txt would look something like this:
21
22
23
80
443
445
3389How to scan IPv6 networks efficiently
Add AAAA DNS records for all of your IPv6 servers and allow your netscan scan engines to do zone transfers. This limits the scan to IPv6 enabled servers that are in the DNS.
You may also consider using a network management tool, such as netdisco, as input to netscan.