OpenZFS 2.0.0

A major update of the OpenZFS 2.0.0 file system and its maintenance tools has been released. The new version supports Linux kernels starting from 3.10 and FreeBSD starting from version 12.2, and in addition, it now merges the code for both operating systems into a single repository. Among the most significant changes, the developers highlight the following:

• The ability to perform sequential (LBA) rebuilding of a damaged Mirror vDev RAID array has been added. This mechanism is significantly faster than traditional 'healing' recovery. However, it lacks block checksum verification, which means that immediately after its completion, a system integrity check (scrub) is launched.

• L2ARC cache data restoration after system reboot. The cache itself uses a dedicated amount of RAM, avoiding slower hard disk access during frequent data interactions. Now, after a reboot, the L2ARC cache data will be intact.

• Support for compression in the ZStandard format, which provides a compression level comparable to GZIP, but with much higher performance. For convenience, the administrator has the option to choose the compression level to ensure the best balance between performance and disk space savings.

• The ability to select data when transferring with send/receive commands. Now administrators can manually exclude unnecessary or private data before copying a snapshot.

• A number of other, less significant but equally pleasant improvements have been implemented, including the writing of a PAM module for loading folder encryption keys, reorganization of man pages, and updating the documentation, adding a ZFS volume mounting generator for systemd, expanding logging into syslog, improving compatibility with system loaders, and much more.

• New commands and keys have been added to the existing ones, which can be read in more detail in the release notes.

• Optimization of several internal tools for speed and effective use of system resources has been conducted.

Complete list of changes.

Source: linux.org.ru