The 2.16 series release notes contain important changes in this release series.
- MEDIUM: An attacker with direct network access to the server could send a specially crafted sequence of network packets that could cause a kernel panic or slow down the system causing a Denial of Service (DoS). For more information, see the associated CVEs: CVE-2019-11477, CVE-2019-11478, CVE-2019-11479.
- Packages have been updated to the latest security versions.
- Internal API data values exceeded internal buffer sizes and caused access from the Git command-line to fail unconditionally for some users or deploy keys.
- Pre-seeding the initial replica appliance in a HA configuration would result in the failure of the existing primary appliance.
- The "Learn why" link beside the message "Custom sign-in messages are disabled when SAML authentication is enabled" pointed to a nonexistent help article.
- In the GraphQL API, the
suggestedReviewers field returned an error when queried in combination with some other fields (e.g.,
- Pre-receive hooks that printed non UTF-8 characters would fail with an "incompatible character encodings" error message.
- GitHub Enterprise Server incorrectly enforced a version of Backup Utilities that was the same or newer than the precise patch version of GitHub Enterprise Server.
- When pushing a very large number of Git LFS objects to a repository, the returning "Git LFS Integrity Check" warning message was confusing, leading users to think something went wrong.
- On a freshly set up GitHub Enterprise without any users, an attacker could create the first admin user.
- Custom firewall rules aren't maintained during an upgrade.
- svn checkout may timeout while the repository data cache is being built. In most cases, subsequent svn checkout attempts will succeed.
- Git LFS tracked files uploaded through the web interface are incorrectly added directly to the repository.
- Issues cannot be closed if they contain a permalink to a blob in the same repository where the file path is longer than 255 characters.
- Resque workers may not be cleaned up following a configuration run leading to a growing number of stale workers which in turn could lead to high memory consumption.
The GitHub Team