TrueNAS SCALE: The voyage begins with version 20.10

[usern]

EXT4 boot devices would be preferred for my use case since the bare metal server hosting service does not have PXE booted remote rescue tools for ZFS yet. At the moment I just use SMB server running on Ubuntu with SSH management, but TrueNAS web management would be way nicer.

 

[horizonbrave]

Please could you define the "industry-standard servers" which will support HCI (hyper convergence)?
Maaaaaaany of us won't be able to afford, economically and logistically (non-US residents), a R-Series system but as you know already run Truenas on server grade hardware.
Thank you

PS some of us "cheap pockets" wouldn't mind paying for a subscription model, so that we can at least give a "small" economical contribute back ;)

 

[morganL]

Please could you define the "industry-standard servers" which will support HCI (hyper convergence)?
Maaaaaaany of us won't be able to afford, economically and logistically (non-US residents), a R-Series system but as you know already run Truenas on server grade hardware.
Thank you

PS some of us "cheap pockets" wouldn't mind paying for a subscription model, so that we can at least give a "small" economical contribute back ;)

The same hardware that runs TrueNAS CORE will run TrueNAS SCALE for free. The primary difference is that multinode clusters will need a high bandwidth interconnect network. 10Gbe will work, but 100Gbe is better.

 

[Mlovelace]

The same hardware that runs TrueNAS CORE will run TrueNAS SCALE for free. The primary difference is that multinode clusters will need a high bandwidth interconnect network. 10Gbe will work, but 100Gbe is better.

When I read multinode clusters with high bandwidth interconnect, I start thinking iWARP or RoCE. Any plans to include Debian's RDMA-core so users can leverage it?

 

[morganL]

When I read multinode clusters with high bandwidth interconnect, I start thinking iWARP or RoCE. Any plans to include Debian's RDMA-core so users can leverage it?

If someone needs it, we rarely refuse a purchase order or a developer contribution....

 

[stele77]

Just came to say: Thanks for the nice christmas present! December release should be my sweetspot to dip my toes in the water.. :)

 

[Kris Moore]

Just came to say: Thanks for the nice christmas present! December release should be my sweetspot to dip my toes in the water.. :)

Great! We should have some good stuff available for Plugins / Containers by then, which I know more than a few of you are eager to use.

 

[Stefan1970]

Scale-Out ZFS: Capacity & Performance

So when I run out of physical storage space in my chassis, I could add a new server to increase my storage pool. Similair to Ceph?

 

[Kris Moore]

Yes, if you grow beyond a single system, you'll be able to cluster and add 1-2+ systems to the pool to grow further.

 

[Geek Baba]

The same hardware that runs TrueNAS CORE will run TrueNAS SCALE for free. The primary difference is that multinode clusters will need a high bandwidth interconnect network. 10Gbe will work, but 100Gbe is better.

Hi @morganL will it require the 100Gbe switches to scale out or point to point will work with adding say 2 NIC - e.g. Mellanox CX455A 100GbE NIC HPE InfiniBand

 

[Patrick M. Hausen]

As long as you stick to two nodes you can always drop the switch - true for every clustering solution - the nodes won't notice if there is a switch inbetween or just a cable/fibre.

 

[Geek Baba]

As long as you stick to two nodes you can always drop the switch - true for every clustering solution - the nodes won't notice if there is a switch inbetween or just a cable/fibre.

Thanks! Are any recommendation for 100Gbe network cards or switches that work with scale?

 

[morganL]

The speed of the network needed depends on the speed of the cluster. It's possible to network with 10Gbe, but 100Gbe is preferred for performance. We've tested with Chelsio and Mellanox, but don't expect any major problems with other NICs. Linux has strong driver support.

 

[Geek Baba]

The speed of the network needed depends on the speed of the cluster. It's possible to network with 10Gbe, but 100Gbe is preferred for performance. We've tested with Chelsio and Mellanox, but don't expect any major problems with other NICs. Linux has strong driver support.

Thanks! Thats what i was hoping for

 

[adrianwi]

Anyone running it in a bhyve VM, just to play with?

In case anyone is interested in trying, the answer appears to be YES it does

 

[Geek Baba]

After a very successful development cycle with 200+ trial users, the first version of TrueNAS SCALE is now available. As our initial community post and blog on SCALE indicated, TrueNAS SCALE is defined by its acronym:

Scale-Out ZFS: Capacity & Performance
Converged compute and storage
Active-Active reliability
Linux containers; Docker, K8s, and KVM
Easy Setup & Management

TrueNAS starts from the TrueNAS 12.0 base which includes OpenZFS 2.0, all the storage services, the middleware to coordinate these, and the web UI to present a user-oriented view of the system. This base has been tested by hundreds of thousands of users over the last few years. The new capabilities being added are based on Debian Linux and these define the new opportunities for SCALE:

KVM Virtualization: Mature Hypervisor with good reliability, Guest OS support, and Enterprise features.

Kubernetes: Applications can be single (docker) containers or “pods” of containers.

Scale-out ZFS: SCALE will enable datasets to be defined as ZFS datasets or cluster datasets which span multiple nodes and ZFS pools. Cluster datasets will have a variety of redundancy properties and still support ZFS snapshots.

Unlike other Hyperconverged Infrastructure solutions, TrueNAS SCALE will have deployment benefits as a single node, a dual-node “HA” system, or as a cluster of multiple nodes. Start with a single node system and in the future, you will be able to scale-out.

The high level release plan follows this process.

​

Typical Use Case	TrueNAS SCALE Tag	Target Date
Developers	NIGHTLY	June 2020
Testers	NIGHTLY	August 2020
Enthusiasts (1 Node)	ALPHA	October 2020
Enthusiasts (Cluster)	BETA	December 2020
Home Users	RC	Q1 2021
Business Users	RELEASE	Q2 2021

“ANGELFISH”
Angelfish is the codename for the feature set described by a set of feature groups. Each “feature group” is described as either PREVIEW, ALPHA, BETA, RC, or RELEASE quality. This first version is called TrueNAS SCALE 20.10 (Angelfish) and is described as follows:

​

Feature Groups	Quality
Debian Linux	RELEASE
Open ZFS 2.0	RC
TrueNAS Middleware and REST API	BETA
TrueNAS WebUI (new features)	ALPHA
NFS, SMB, AFP, iSCSI, S3, WebDAV	BETA
KVM Virtualization	BETA
Kubernetes and Docker Containers	PREVIEW
Cluster Datasets	PREVIEW

Users should read the release notes to confirm support for their particular use case. KVM has little testing by this community at this point but is widely used elsewhere. Kubernetes is based on stable, released code (k3s), but the WebUI and Middleware are PREVIEW quality.

Cluster datasets require some additional TrueCommand features (expected in TrueCommand 2.0 - November) to provide an easy-to-use WebUI to manage them. In the meantime, the CLI and APIs can be tested, and this feature group also has PREVIEW status.

The UI, while similar to TrueNAS CORE, has also been improved with some new UX enhancements across the networking and settings sub-sections. Further UX improvements are expected to arrive later this year.

We appreciate the community feedback and bug reports and hope to get all those features to RELEASE quality faster.

A special thanks also goes to the large number of community members who joined the development and test team. We’ve really enjoyed your contributions and teamwork and it has greatly contributed to the accelerated development process. We look forward to a continued engagement and a few drinks at some point (physically or virtually).

Is TrueNAS SCALE for Users or Developers?

Out of the gate, TrueNAS SCALE is primarily for developers and bug hunters and can be downloaded here. For Linux developers, there are many opportunities to contribute to the Open Source TrueNAS SCALE project. We have made it a very well coordinated and managed environment to develop the best Open Hyperconverged Infrastructure. For more information, see this Community post.

This TrueNAS SCALE 20.10 version is also intended for tech-savvy enthusiasts who have a single node, a backup plan, and a willingness to resolve any issues they find. The 20.12 version to be released in December is intended for multi-node expansion.

In 2021, TrueNAS SCALE is expected to get to full RELEASE quality for a clustered system. It will be able to run on High Availability (HA) X-Series and M-Series systems, as well as industry-standard servers.

If you have any additional questions or need advice on a new project, please email us at info@iXsystems.com. We are standing by to help.

Hi @Kris Moore - is 20.12 still on track to be released this month...

 

[morganL]

Yes, 20.12 will be available before Christmas..

 

[perriko]

ya, ya
but what developers do not inspire/copy a logical manner if is works well
you must stay in 1 foot (on the left of course) and scratch in the left ear with RIGHT hand from 3 to 3 second to work something
for that is ALPHA, BETA and other things. TO tell developers what we think about they works

What??!

 

[bar1]

can this be installed on a normal flash drive/sd card or is not not recommended?

 

[Ericloewe]

It's as recommended as for any vaguely modern version of FreeNAS/TrueNAS (i.e. not recommended).