Webinar Overview
Kick off the new year with a deep dive into what ZFS 2.4 brings to the table.
Allan Jude, Co-founder and Head of Solutions Architecture, and Paul Dagnelie, Principal ZFS Engineer, will walk through the most impactful changes in the new release, from performance gains and smarter observability to reliability improvements that matter for day-to-day operations.
You’ll get practical guidance on when and how to upgrade, what the new features mean for daily workflows, and how 2.4 fits into the broader direction of the ZFS project. If ZFS is part of your 2026 roadmap or already powering your infrastructure, this session will help you start 2026 with confidence.
What You’ll Learn:
- What’s improved in ZFS 2.4: performance, stability, and recovery
- Operational updates admins need to be aware of
- When to upgrade and where ZFS is headed next
Top Questions from the Session—Answered!
🗨️ Is it possible to extract metrics using tools like the Prometheus Node Exporter, specifically for ZFS metrics such as capacity and snapshot usage?
There are ZFS-specific node exporters that expose internal ZFS metrics, including pool and dataset properties, including capacity/fragmentation, memory usage breakdowns, and other runtime metrics. Alternatively, ZFS also provides an InfluxDB plugin that sends pool and dataset statistics to the InfluxDB time-series database.
🗨️ If I have ten NVMe U.2 drives, how should I build my ZFS pool today, and how will best practices change in the near to medium term?
Focus your VDEV layout choices on your workload and failure tolerance, as it won't make a large difference in throughput on high-performance NVMe. Future changes to improve ZFS with NVMe will not require layout changes. Support for newer NVMe designs with large sectors will benefit from the new label design in 2.5, but those devices are not widespread yet.
🗨️ What are the steps to recover a bhyve VM dataset from a snapshot after a compromise or corruption?
Recovery starts with selecting a known-good snapshot and restoring the affected dataset to that state, either by rolling it back or by creating a clone. ZFS snapshots provide a consistent recovery point, allowing VM data to be restored quickly with minimal data loss.
Date: January 22, 2026
Time: 11:00 AM EST
Duration: 50 minutes.
Know someone else who might be interested? Share this entry with them!
Meet the Hosts
Co-Founder and Head of Solutions Architecture at Klara Inc., Allan Jude has been on the team since the beginning. Shepherding an amazing team of developers and sysadmins, he is the technical heart of our team. A core ZFS developer, FreeBSD contributor, and co-author of FreeBSD Mastery: ZFS, Allan is a community go-to person for ZFS and open source through and through. He enjoys spending his time improving ZFS, advancing FreeBSD, and making open source code better.
Learn About Klara
Paul Dagnelie is a Principal Engineer with Klara Inc. With over a decade of experience in ZFS, he has been a major contributor to a number of areas of the OpenZFS codebase, and a repeat presenter at the OpenZFS Summit. Paul enjoys tackling challenging problems, working with with the OpenZFS Community, and creating useful and powerful software that everyone can take advantage of.
Learn About Klara