I am building a basic WAF and I have access to multiple CIDR ranges of bad visitors.
My implementation looks like this:
$badIPRanges = [
'1.2.220.142/31',
'1.2.220.144/29',
'1.2.220.152/30',
];
function isBlackListed(string $ip, array $ipRanges): bool {
foreach ($ipRanges as $ipRange) {
if (ipInRange($ip, $ipRange)) {
return true;
}
}
return false;
}
function ipInRange(string $ip, string $ipRange): bool {
[$net, $mask] = explode('/', $ipRange);
$ip_net = ip2long($net);
$ip_mask = ~((1 << (32 - $mask)) - 1);
$ip_ip = ip2long($ip);
$ip_ip_net = $ip_ip & $ip_mask;
return ($ip_ip_net == $ip_net);
}
var_dump(isBlackListed('1.2.220.145', $badIPRanges) ? 'Yes' : 'No');
I expected 1.2.220.145
not to be in the multiple CIDR ranges, but apparently not. I am getting Yes
back from the above test code.
Am I doing something wrong? or is this valid? The original implementation was taken from here.
Using https://tehnoblog.org/ip-tools/ip-address-in-cidr-range/ I verified that 1.2.220.145
is in the 1.2.220.144/29
CIDR block, so does this mean the CIDR block 1.2.220.142/31
is redundant in my $badIPRanges
list?
Unless you've deeply internalized the format, you can seldom look at the dotted quad notation and have any idea what the subnetting is for anything other than a multiple of 8.
The binary representation of those networks is:
1.2.220.142 31 00000001 00000010 11011100 10001110
1.2.220.144 29 00000001 00000010 11011100 10010000
1.2.220.152 30 00000001 00000010 11011100 10011000
1.2.220.142/31
is wholly different from the other two, and none overlap.