For example, CVE-2017-14491 has a priority level of "high".
Is there any formal method to determine the priority of this vulnerability (unique to Canonical)? Or is this priority based on the disclosure of other vendors?
For example, CVE-2017-14491 has a priority level of "high".
Is there any formal method to determine the priority of this vulnerability (unique to Canonical)? Or is this priority based on the disclosure of other vendors?
The Ubuntu Security Team's wiki page on Bug Triage says that priority levels are described in ubuntu-cve-tracker's README, where you can see this:
Ubuntu Priorities
These are very similar to the Debian priorities, but with some differences. Priorities can be roughly mapped as:
negligible Something that is technically a security problem, but is only theoretical in nature, requires a very special situation, has almost no install base, or does no real damage. These tend not to get backport from upstreams, and will likely not be included in security updates unless there is an easy fix and some other issue causes an update.
low Something that is a security problem, but is hard to exploit due to environment, requires a user-assisted attack, a small install base, or does very little damage. These tend to be included in security updates only when higher priority issues require an update, or if many low priority issues have built up.
medium Something is a real security problem, and is exploitable for many people. Includes network daemon denial of service attacks, cross-site scripting, and gaining user privileges. Updates should be made soon for this priority of issue.
high A real problem, exploitable for many people in a default installation. Includes serious remote denial of services, local root privilege escalations, or data loss.
critical A world-burning problem, exploitable for nearly all people in a default installation of Ubuntu. Includes remote root privilege escalations, or massive data loss.
CVE-2017-14491 has high priority because dnsmasq is installed by default and the bug allows both denial-of-service and code execution remotely and is not particularly hard to exploit.