TrueNAS SCALE Announcement and Nightly Image Downloads

[anodos]

thx for that quick answer

which SBC is that?
they are running as a cluster or single node?

Right now we don't have support for clustered services. I'm using a Supermicro A2SDi-8C-HLN4F. For my testing / dev purposes this is sufficient. This is personal use, not something that has been through our official QE / hardware processes. It works well and I'm happy with it, but it may not be a perfect fit for everyone out there (limited number of sata ports, too many gigabit NICs, etc). I'm sure there are much more opinionated people that can give recommendations.

If you want something certified to work with FN / TN, then you can look at the official products iXsystems sells (FreeNAS mini and FreeNAS certified).

 

[PackElend]

If you want something certified to work with FN / TN, then you can look at the official products iXsystems sells.

nope, it will be for personal use, just curious
thx :)

 

[ornias]

Right now we don't have support for clustered services. I'm using a Supermicro A2SDi-8C-HLN4F. For my testing / dev purposes this is sufficient. This is personal use, not something that has been through our official QE / hardware processes. It works well and I'm happy with it, but it may not be a perfect fit for everyone out there (limited number of sata ports, too many gigabit NICs, etc). I'm sure there are much more opinionated people that can give recommendations.

Thats a fine professional grade board though and the 8c atoms aren't the worst either :)

If you want something certified to work with FN / TN, then you can look at the official products iXsystems sells (FreeNAS mini and FreeNAS certified).

If one wants something a little cheaper: one could get the same board IX uses from supermicro and the case is also available on the open market. That being said: It won't be MUCH cheaper, just a bit.

 

[brando56894]

The Avotons (8Core Atoms) definitely aren't a slouch when it comes to power, the first NAS I built (running FreeNAS 9.3) was a C2750 until I hit that lovely watchdog bug and the board died. I remember being one of the first handful to experience it, about like 2 months before it was posted here and affected everyone with a Mini.

 

[antsh]

Quick question. I see that NFSv4 ACLs are still to do. Is this the reason the permissions on a pool cannot be changed yet?

 

[anodos]

Quick question. I see that NFSv4 ACLs are still to do. Is this the reason the permissions on a pool cannot be changed yet?

Yup. Still a lot of work to do there.

 

[ornias]

@anodos Is that also the reason I couldn't get NFS to work right as discussed on the slack?

 

[anodos]

@anodos Is that also the reason I couldn't get NFS to work right as discussed on the slack?

I just got back from vacation. It will probably be a day or two before I get up to speed on things. Possibly could cause issues for NFS. You can test by simply using chmod / chown from CLI.

 

[j0hnnyclaymore]

What version of Debian is TrueNAS Scale based on?

 

[Kris Moore]

What version of Debian is TrueNAS Scale based on?

Debian 11 / Bullseye (Unstable) right now. Linux Kernel 5.9.0.

 

[Patrick M. Hausen]

VMware ESXi 7 for some reason insists it is Debian 4 and I should set the OS type accordingly. Any idea?

 

[Kris Moore]

VMware ESXi 7 for some reason insists it is Debian 4 and I should set the OS type accordingly. Any idea?

Not sure how VMware identifies a particular Debian version TBH. However you can try setting it to a newer Debian version there if it has 10/11.

 

[Patrick M. Hausen]

Of course. I did set it to Debian 11 - which leads to continues warning messages from ESXi about a supposed mismatch until you set it to Debian 4 ...
Whatever that setting changes, I wonder.

Edit: looks like this:

 

[inman.turbo]

Of course. I did set it to Debian 11 - which leads to continues warning messages from ESXi about a supposed mismatch until you set it to Debian 4 ...
Whatever that setting changes, I wonder.

Edit: looks like this:
View attachment 43109

This will change later on. VMWare used to do this for CentOs8 and say it needed to be changed to CentOs6. It tends to get mixed up on some of the newer releases. It is not likely to change anything, as this is just relating mostly to how power control is handled, which tends to be the same anyway over many many releases.

 

[msgonz]

I need some help, I installed Scales but I don't see any option to change the update train to nightly. It's currently stuck on prerelease.

 

[beagle]

I need some help, I installed Scales but I don't see any option to change the update train to nightly. It's currently stuck on prerelease.

According to morganL's post on another thread:

My understanding is that the NIGHTly train issue is temporarily not available due to a technical issue. I expect it will be available again next week, but will report back if not.

 

[Satspares]

I need some help, I installed Scales but I don't see any option to change the update train to nightly. It's currently stuck on prerelease.

The latest manual update file would be here....

https://update.freenas.org/scale/TrueNAS-SCALE-Angelfish-Nightlies/

Regards.

 

[msgonz]

The latest manual update file would be here....

https://update.freenas.org/scale/TrueNAS-SCALE-Angelfish-Nightlies/

Regards.

Thank you

 

[usergiven]

Late to the Truenas Scale party, looks super cool. If you don't mind, a functionality question: From what I gather, in Scale I could cluster two hosts running scale and theoretically have all the data in a pool on one host and all the compute (docker, etc) on another but in it's functionality it would act like all one big box? Secondly, across those two hosts I also could have a span of two pools that functioned as one, each holding their own unique data but presented as one pool. Or I could decide to replicate the exact same data across both hosts for high availability but those still would be presented as one big box?

I'm not sure if I'm just speaking bananas but wanted to doublecheck with you. Thanks!

 

[morganL]

Late to the Truenas Scale party, looks super cool. If you don't mind, a functionality question: From what I gather, in Scale I could cluster two hosts running scale and theoretically have all the data in a pool on one host and all the compute (docker, etc) on another but in it's functionality it would act like all one big box? Secondly, across those two hosts I also could have a span of two pools that functioned as one, each holding their own unique data but presented as one pool. Or I could decide to replicate the exact same data across both hosts for high availability but those still would be presented as one big box?

Yes....
Data and compute can be on different nodes
Data can be striped across nodes into a larger pool... (no HA in this case)
Data can be mirrored across nodes, but a third node is needed to arbitrate in case of failure (split brain)
Planning to make this functionality easy to use on Feb release. 21.02