We are pleased to release TrueNAS CORE 13.3-BETA2!
Early Release Software
Early releases of a major version are intended for testing and feedback purposes only. Do not use early-release software for critical tasks.
This early release includes improvements and fixes for issues discovered after the release of 13.3-BETA1.
Notable changes:
Fixed replication for SSH+NETCAT connections (NAS-128866).
Allow creation of more than 4 billion files per dataset (NAS-129166).
I think we need to set some proper expectations here. 13.0-U5+ was “the most stable release ever”, but only after 5+ dot releases to make it so by quashing all the edge cases like this which broke in the transition from FreeBSD 12 → 13.
In case nobody remembers recent history, it was pretty common on 11/12/13 for it to take 6+ months before a release had all the weird corner cases quashed. 13.3 is no different, and it will be starting over in many ways since we have synced with upstream FreeBSD, OpenZFS and an entirely updated ports tree. This is why we run Beta / RC cycles to try and catch as many of these “fun” changes as we can.
That said, the Core / Bhyve VM functionality is indeed not supported by us directly. We just rely on upstream to improve and (hopefully) not break things. If anybody wants to go investigate and propose a patch on it for CORE, they are welcome to. We will gladly review incoming patches.
Appreciate the response. Right now I’m just waiting to see if other users are seeing the same issue, or if my use case is a special one. If it is, then I’ll happily do a reinstall on fresh hardware to see if that resolves the issue.
@pmh@winnielinnie If you have a minute, spin up a linux VM and see if you’re seeing the same issue.
But some of the time was due to the version 10 release, so having extra developer time was appreciated by many end users like myself. No need for speed as accuracy in my book is more important. And while I may not agree with everything iX does, I still appreciate them and the product in use.
Excellent, I will have to wait until the current replication task has finished as the current CORE replication task cannot deal with interruptions - it either completes or it fails the replication and restarts each dataset / snapshot transfer anew. Given my current transfer speed and the size of the dataset, that’ll be a while.
I take it that SCALE has the ability to keep track of what was successfully transferred and picks up where it left off. Would be great to backport the same feature into CORE.
@kris, it would make a lot sense to have the BETA2 show up with version info in the System → Upgrade UI window. The current window only shows
For Virtualization support - we don’t recommend running TrueNAS in a VM. This is not a recommended/supported/validated config. Baremetal should be used.
Just to reiterate, CORE is in maintenance mode. The code-base for it and SCALE have diverged quite significantly at this point and we’re not going to be back-porting any feature functionality like that going forward.
Because both releases are in the BETA train, they will show up in that single train option. I.E. a train can have multiple releases in it, BETA{1|2|3}. When we roll an RC1 later that would end up in its own train.
"As noted in the release notes the Virtual Machines features are untested and provided without support to the TrueNAS Community."
and this:
“That said, the Core / Bhyve VM functionality is indeed not supported by us directly. We just rely on upstream to improve and (hopefully) not break things. If anybody wants to go investigate and propose a patch on it for CORE, they are welcome to. We will gladly review incoming patches.”
Like @victor, I’m left wondering about the future of VM support in 13.3. AFAIK, a problem about mouse control (275760 – bhyve: xhci not working on linux guests) in graphical VMs has been sorted upstream months ago, but for some reason the fix is not in 13.3 beta2. Why is that?
A separate issue concerns awkward UEFI installs of some VMs because TrueNAS CORE has not used “BHYVE_UEFI_VARS.fd” with “bootrom,/usr/local/share/uefi-firmware/BHYVE_UEFI.fd”. Clearly that would require modification of the vm.py code, so depending on the definition of “maintenance mode” may never happen.
Error: NextCloud had a failure Exception: RuntimeError Message: pkg error: - py39-fail2ban : Refusing to fetch artifact and run post_install.sh! Partial plugin destroyed
I’ve been running the Beta without issue for the last 18 days. But temper that with the fact that I only use it as a NAS, only one Plex jail which I don’t really use anymore, so I should blow that away. I’ve rebooted (not due to problems) and reboots appear fine, nothing whacky. I wish the SHELL was back but I understand why it went away, because either it couldn’t be fixed or prohibitive to be fixed. But I use SSH so not a big deal. I just looked at the bug reports to see what others may have been going through, that is probably a good place to look, found 32 tickets, 30 closed and 2 open.
. Letting it run until it breaks or 13.3 RC comes out.