Last week, the team behind ZFS released zfs-0.8.2, an advanced file system. This release comes with support for 2.6.32 – 5.3 Linux kernels and comes with a list of changes.
What’s new in zfs-0.8.2
- The issue regarding the deadlock condition for scrubbing root pools on kernels has been resolved in this release.
- The team has made QAT related bug fixes.
- Fixes have been made to the zpool subcommands error message and unsupported options.
- zfs-dkms .deb package warning in the prerm script has been fixed.
- zvol_wait script now ignores partially received zvols.
- New service that waits on zvol links have been created.
- In etc/init.d/zfs-functions.in arch warning has been removed.
- Comments have been updated to match code.
- In this release, ZFS_DEV macro is used instead of literals.
- Slog test setup has been made more robust.
- Performance has been improved with the help of dmu_tx_hold_*_by_dnode().
- In this release, default zcmd allocation has been increased to 256K.
- Error text for EINVAL in zfs_receive_one() has been fixed.
Few users on Hacker News seem to be happy about this release and the progress made by the team behind zfs, a user commented on Hacker News, “I contributed a few patches to ZFS on Linux about 8 years ago – at a time when it was still very much in its infancy and panic’d when you looked at it in the wrong way. It’s incredible how far they’ve come. We’re using ZFS on Linux on about 120 servers at work and it’s rock solid. Snapshots are a lifesaver in our day-to-day ops.”
Another user commented, “Always admired ZFS since when it came out. The talks by the creators were so enlightening.”
Few others expected a block-pointer rewrite and background dedupe in this release. One of them commented, “Still no block-pointer rewrite?”
To know more about this news, check out the official post.
Other interesting news in programming