An issue was discovered in net/ipv6/ip6mr.c in the Linux kernel before 4.11. By setting a specific socket option, an attacker can control a pointer in kernel land and cause an inetcsklistenstop general protection fault, or potentially execute arbitrary code under certain circumstances. The issue can be triggered as root (e.g., inside a default LXC container or with the CAPNETADMIN capability) or after namespace unsharing. This occurs because sktype and protocol are not checked in the appropriate part of the ip6mroute* functions. NOTE: this affects Linux distributions that use 4.9.x longterm kernels before 4.9.187.
"https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2017-18509.json"
[
{
"events": [
{
"introduced": "0"
},
{
"fixed": "3.16.72"
}
]
},
{
"events": [
{
"introduced": "3.17"
},
{
"fixed": "4.4.187"
}
]
},
{
"events": [
{
"introduced": "4.5"
},
{
"fixed": "4.9.187"
}
]
},
{
"events": [
{
"introduced": "4.10"
},
{
"fixed": "4.11"
}
]
},
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "16.04"
}
]
},
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "8.0"
}
]
},
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "9.0"
}
]
},
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "10.0"
}
]
}
]