SCALE 21.08-BETA.1 Released!

[Kris Moore]

I'm pleased to officially announce that TrueNAS SCALE 21.08-BETA.1 is now available for download and updating!

This release marks the last BETA before we move towards RC and proper Release status. We very much appreciate all the hard work from the iX engineering team, but also from our fantastic community users who have helped us find and fix issues along the way. SCALE 21.06 has had the most active users of any TrueNAS (or FreeNAS) beta product EVER, and we're ecstatic to see that kind of enthusiasm continue with 21.08 and beyond.

Good luck updating, and as usual please let us know of any issues via our Jira tracker.

Blog Post:

https://www.ixsystems.com/blog/truenas-scale-21-08-delivers-major-smb-enhancements/

Release Notes:

https://www.truenas.com/docs/releasenotes/scale/21.08-beta.1/

Download:

https://www.truenas.com/download-truenas-scale/

 

[sretalla]

please let us know of any issues via our Jira tracker.

Jira doesn't have SCALE 21.08-BETA.1 in the list of "Affects Versions".

I wanted to log an issue that I see on all 3 of my test boxes that the SMB.configure task is running (for a really long time) on all 3 and any attempt to change the SMB config (as it's currently not started/configured on any of them) results in an error about the cluster state and won't save the config change.

[EFAULT] SMB-related changes are not permitted while cluster unhealthy.

 

[anodos]

Jira doesn't have SCALE 21.08-BETA.1 in the list of "Affects Versions".

I wanted to log an issue that I see on all 3 of my test boxes that the SMB.configure task is running (for a really long time) on all 3 and any attempt to change the SMB config (as it's currently not started/configured on any of them) results in an error about the cluster state and won't save the config change.

[EFAULT] SMB-related changes are not permitted while cluster unhealthy.

Is server clustered (i.e. using glusterfs)?

 

[sretalla]

I was playing with Gluster manually from CLI on one of them (actually as part of this: https://www.truenas.com/community/t...tion-i-wish-to-run-autotier.94103/post-651299), but no gluster volumes are actually resident on all.

 

[anodos]

I was playing with Gluster manually from CLI on one of them (actually as part of this: https://www.truenas.com/community/t...tion-i-wish-to-run-autotier.94103/post-651299), but no gluster volumes are actually resident on all.

Okay. You should ensure that the glusterd service is disabled. You can view state via:

Code:

tn23# midclt call service.query '[["service", "=", "glusterd"]]'
[{"id": 26, "service": "glusterd", "enable": false, "state": "STOPPED", "pids": []}]

get `id` then run:

Code:

tn23# midclt call service.update 26 '{"enable": false}'
26

Then reboot server.

 

[csj]

As a side-note, clustering on SCALE is expected to be exclusively managed via TrueCommand for now. The usual disclaimer applies but if you do things via the CLI then expect stuff to be broken. Especially if making manual changes to the glusterd service since that's just 1 of the puzzle pieces in the clustering API

 

[sretalla]

Okay. You should ensure that the glusterd service is disabled. You can view state via:

OK, that stopped the problem.

if you do things via the CLI then expect stuff to be broken

Yes, well understood. (just didn't see the link between SMB and Gluster)

 

[anodos]

Code:

Yes, well understood. (just didn't see the link between SMB and Gluster)

We look at gluster state to see whether clustering is enabled. If clustering is enabled, then SMB server configuration, SMB server state, and directory services state is clustered. If the clustered backend for these is in unhealthy state then service updates are not permitted.

 

[nickhechtel]

somehow i can no longer create Gluster Volumes, first i get a NAS middleware error, after trying again he tells me
NAS API error
"These systems are already part of 2 trusted peer pools"
which is completely right as he already peered the 3 storage nodes. But i cant create a volume anymore.
Truecommand Output is this:
[2021-09-01, 13:56:10 UTC] WARNING: NAS Job Failed: Node2 clusterjob.process_queue 170

[2021-09-01, 13:56:10 UTC] [Got Unhandled Job Event: Node2 170 clusterjob.process_queue FAILED]

any help would be appreciated.

 

[overeat]

Question: After upgrade I get this note:

New ZFS version or feature flags are available for pool IronWolf_3TB. Upgrading pools is a one-time process that can prevent rolling the system back to an earlier.

Is there a command I need to run after upgrading? Or is this just a weird way of saying that's a new version of TrueNas out and maybe you should upgrade. (I don't recall if this message was here before I upgraded or not, so I'm kind of confused).

PS. This fixed a lot of issues I had setting up SMB before this new version, so thanks for the hardwork.

 

[jgreco]

No, it means that new ZFS feature flags are available. You may upgrade your pool to use them, but once you do so, you may not roll back to an earlier version of TrueNAS. It is a one-way process. You do not have to do it now, you can wait a month, or even a year. If there is any possibility whatsoever that you might want to go back to an older version, you should not do it.

 

[anodos]

somehow i can no longer create Gluster Volumes, first i get a NAS middleware error, after trying again he tells me
NAS API error
"These systems are already part of 2 trusted peer pools"
which is completely right as he already peered the 3 storage nodes. But i cant create a volume anymore.
Truecommand Output is this:
[2021-09-01, 13:56:10 UTC] WARNING: NAS Job Failed: Node2 clusterjob.process_queue 170

[2021-09-01, 13:56:10 UTC] [Got Unhandled Job Event: Node2 170 clusterjob.process_queue FAILED]

any help would be appreciated.

Please file a jira ticket for this at `jira.ixsystems.com`.

 

[phin]

Did we lose the option to add manually added docker images to the Host Network? It doesnt seem to be an option and has broke my current plex install.

 

[zamana]

We'll need a way to apply mass operations in some screens. That's the one of them:

DELETE one by one is out of my plans...

THANKS for new beta!

 

[morganL]

We'll need a way to apply mass operations in some screens. That's the one of them:
DELETE one by one is out of my plans...
THANKS for new beta!

THanks... could you make a suggestion via report a bug. Please identify the specific screens that impact you and you might have the same issue again.

 

[07stuntar1]

As a side-note, clustering on SCALE is expected to be exclusively managed via TrueCommand for now. The usual disclaimer applies but if you do things via the CLI then expect stuff to be broken. Especially if making manual changes to the glusterd service since that's just 1 of the puzzle pieces in the clustering API

Is SMB share available on TrueCommand Nightly? I don't see an option to create a share of the gluster volume.

 

[anodos]

Is SMB share available on TrueCommand Nightly? I don't see an option to create a share of the gluster volume.

No it has not been implemented yet in TrueCommand. This is still WIP.

 

[overeat]

No, it means that new ZFS feature flags are available. You may upgrade your pool to use them, but once you do so, you may not roll back to an earlier version of TrueNAS. It is a one-way process. You do not have to do it now, you can wait a month, or even a year. If there is any possibility whatsoever that you might want to go back to an older version, you should not do it.

Thanks for the reply. To be honest, as a complete newb to TrueNas & ZFS (this week) but with lots of experience with Linux, I'm still confused by the response. Consider me the typical dumb users. Since I have no frame of reference with that any of this means, I'm still unsure even after your reply if upgrading from BETA-0.6 to BETA-0.8 performed this or not.

There's no "What to do after upgrade" documentation, so I'm at a wall there as well. At least for TrueNas SCALE. I know it's new and will come with time.

A little more googling to get to the bottom of this, I found some older FreeNas documentation which explains "Upgrading a ZFS pool". I'm posting this mostly for Google and others in the future.

2. Installing and Upgrading — FreeNAS®11.2-U3 User Guide Table of Contents

www.ixsystems.com

My suggestion would be to add a link to appropriate documentation in the alert to hand hold users and avoid idiots like me from wasting community resources.

 

[morganL]

Thanks for the reply. To be honest, as a complete newb to TrueNas & ZFS (this week) but with lots of experience with Linux, I'm still confused by the response. Consider me the typical dumb users. Since I have no frame of reference with that any of this means, I'm still unsure even after your reply if upgrading from BETA-0.6 to BETA-0.8 performed this or not.

There's no "What to do after upgrade" documentation, so I'm at a wall there as well. At least for TrueNas SCALE. I know it's new and will come with time.

A little more googling to get to the bottom of this, I found some older FreeNas documentation which explains "Upgrading a ZFS pool". I'm posting this mostly for Google and others in the future.

My suggestion would be to add a link to appropriate documentation in the alert to hand hold users and avoid idiots like me from wasting community resources.

TrueNAS SCALE documentation is improving and its where we need to be for BETA. The team is mostly focussed on quality testing to get to RC and then RELEASE. Documentation will also improve as we get there.

Does anyone prefer good documentation and buggy software?

 

[Ziggy]

I’m on the alpha/beta nightlies for about two months. Now that the 21.08-BETA.1 has been released, is there anyway to stay on a stable beta and not have constant updates or is this just part of the beta process I can’t avoid? There’s no option I can see in the updates section other that to untick ‘check for updates daily etc’ box.