TrueNAS SCALE 22.02-RC.1 Now available!

[mye]

It seems encryption is a common factor. In the system I've been able to inspect, the middleware was stuck in a job trying to recursively unlock the pool.

That's my guess. Encrypted pool. beta2->RC1 - on boot stuck on importing zfs pools. Rebooted to Beta1, delete pools, Boot to RC1, create pools with encryption, reboot to RC1 - stuck on importing pools. Fresh RC1 install, import pools from gui - reboot - stuck on importing.

 

[rmr]

Yes, that's it. On a fresh install, putting the "Apps" pool on an encrypted pool causes it to hang at the next reboot. Putting it on an unencrypted pool (leaving everything else the same) causes it to hang.

It seems encryption is a common factor. In the system I've been able to inspect, the middleware was stuck in a job trying to recursively unlock the pool.

 

[uRyn]

I have try no matter fresh install or upgrade, same problem. I think it's not connected to the xattr. it's the pool's problem.

1. If I shutdown the server, open the box and remove the hard disks which pool cannot import during system booting. Then turn on server, trueNAS RC1 can boot success. Login the web ui, disconnect the pool. Then shutdown. reconnect the harddisk and turn on the server. while boot success, import the pool just disconnect before and everything working as expected. But the problem is while I restart the system with the pool connected same problem happen again(stuck in import pool).
2. I can also boot to beta2 and disconnect the pool, then restart and boot to RC1, after boot success, import the pool again, everything working as expected. but same problem while restart system with pool connected.

So the problem is the RC1 cannot import pool while system booting. And it's not every pool cause this while system booting. I have several pools, only one of my pool will have this problem. I don't know what's special connect to the pool types. The pool blocked import during booting is a raid-Z with 3 harddisk, several datasets and one of the dataset encrypted with a key.

 

[rmr]

There is a PR in the notes for https://jira.ixsystems.com/browse/NAS-113052, https://github.com/truenas/middleware/pull/7796, which I manually applied. It fixes the issue for me.

Code:

--- a/usr/lib/python3/dist-packages/middlewared/plugins/pool.py
+++ b/usr/lib/python3/dist-packages/middlewared/plugins/pool.py
@@ -1852,7 +1852,9 @@ def import_on_boot(self, job):
                         'Failed to inherit mountpoints for %s', pool['name'], exc_info=True,
                     )
 
-                unlock_job = self.middleware.call_sync('pool.dataset.unlock', pool['name'], {'recursive': True})
+                unlock_job = self.middleware.call_sync(
+                    'pool.dataset.unlock', pool['name'], {'recursive': True, 'toggle_attachments': False}
+                )
                 unlock_job.wait_sync()
                 if unlock_job.error or unlock_job.result['failed']:
                     failed = ', '.join(unlock_job.result['failed']) if not unlock_job.error else ''

To apply this patch, boot back into your working 21.08-BETA.2. Save the patch into a file (for example, ~root/7796.patch) Then, on the command line,

Code:

$ sudo bash
# mkdir /mnt/2202
# mount -t zfs boot-pool/ROOT/22.02-RC.1 /mnt/2202
# cd /mnt/2202
# patch -p1 < 7796.patch
# cd /
# umount /mnt/2202
# rmdir /mnt/2202

You can now reboot into 22.02-RC.1.

 

[marmoset]

There is a PR in the notes for https://jira.ixsystems.com/browse/NAS-113052, https://github.com/truenas/middleware/pull/7796, which I manually applied. It fixes the issue for me.

Code:

--- a/usr/lib/python3/dist-packages/middlewared/plugins/pool.py
+++ b/usr/lib/python3/dist-packages/middlewared/plugins/pool.py
@@ -1852,7 +1852,9 @@ def import_on_boot(self, job):
                         'Failed to inherit mountpoints for %s', pool['name'], exc_info=True,
                     )
 
-                unlock_job = self.middleware.call_sync('pool.dataset.unlock', pool['name'], {'recursive': True})
+                unlock_job = self.middleware.call_sync(
+                    'pool.dataset.unlock', pool['name'], {'recursive': True, 'toggle_attachments': False}
+                )
                 unlock_job.wait_sync()
                 if unlock_job.error or unlock_job.result['failed']:
                     failed = ', '.join(unlock_job.result['failed']) if not unlock_job.error else ''

To apply this patch, boot back into your working 21.08-BETA.2. Save the patch into a file (for example, ~root/7796.patch) Then, on the command line,

Code:

$ sudo bash
# mkdir /mnt/2202
# mount -t zfs boot-pool/ROOT/22.02-RC.1 /mnt/2202
# cd /mnt/2202
# patch -p1 < 7796.patch
# cd /
# mount /mnt/2202
# rmdir /mnt/2202

You can now reboot into 22.02-RC.1.

I think that last mount should be umount? I'm not impacted by this so I can't confirm, but just looking at what you're doing umount makes more sense than mount.

 

[rmr]

I think that last mount should be umount? I'm not impacted by this so I can't confirm, but just looking at what you're doing umount makes more sense than mount.

yes you’re right. I edited the post.

 

[openwinz]

I think that last mount should be umount? I'm not impacted by this so I can't confirm, but just looking at what you're doing umount makes more sense than mount.

There is a PR in the notes for https://jira.ixsystems.com/browse/NAS-113052, https://github.com/truenas/middleware/pull/7796, which I manually applied. It fixes the issue for me.

Code:

--- a/usr/lib/python3/dist-packages/middlewared/plugins/pool.py
+++ b/usr/lib/python3/dist-packages/middlewared/plugins/pool.py
@@ -1852,7 +1852,9 @@ def import_on_boot(self, job):
                         'Failed to inherit mountpoints for %s', pool['name'], exc_info=True,
                     )
 
-                unlock_job = self.middleware.call_sync('pool.dataset.unlock', pool['name'], {'recursive': True})
+                unlock_job = self.middleware.call_sync(
+                    'pool.dataset.unlock', pool['name'], {'recursive': True, 'toggle_attachments': False}
+                )
                 unlock_job.wait_sync()
                 if unlock_job.error or unlock_job.result['failed']:
                     failed = ', '.join(unlock_job.result['failed']) if not unlock_job.error else ''

To apply this patch, boot back into your working 21.08-BETA.2. Save the patch into a file (for example, ~root/7796.patch) Then, on the command line,

Code:

$ sudo bash
# mkdir /mnt/2202
# mount -t zfs boot-pool/ROOT/22.02-RC.1 /mnt/2202
# cd /mnt/2202
# patch -p1 < 7796.patch
# cd /
# umount /mnt/2202
# rmdir /mnt/2202

You can now reboot into 22.02-RC.1.

Confirmed, this patch worked for me as well. Hoping the patch will be part of RC.2. Really appreciate Truenas SCALE and the community support.

 

[Andreuha]

I just tried upgrading from Core to Scale and out of 3 pools only one is successfully brought online.

• Physical drives are available in the disk list.
• 'smartctl -H /dev/... ' returns PASSED for all devices
  • SMART tests initiated through the UI it hang at 99.00% for disks from both online and undetected pools (unrelated bug?)
• pool still not visible after export
• ruled out controller by shuffling drives around (mixing mainboard connectors & an LSI 9205-8I).
• Pools come online with no problems after replacing old boot media and booting into TrueNAS CORE

Is this a known issue?

 

[morganL]

There's an update here. https://www.truenas.com/community/threads/scale-22-02-rc-1-feature-flags-and-import.96461/

In short there are two separate issues:
1. xattr_compat feature flag - if this is "active" the pool needs to be backed-up and recreated (described in release notes)
2. there is a pool import issues with encrypted datasets and SMB -- that will be fixed in an RC1.1 planned for this week. Described in this bug report: https://jira.ixsystems.com/browse/NAS-113052

 

[georgelza]

... so came here all "READY" to do... reading 5 pages later and it's hmmm ye maybe wait.

I created a VM on VirtualBox to test the water, well Plex install as a start did not go well, it seems to be complaining about the name I want to give the Application name,

Then can't find QBitTorrent either.

... not "clear" either how to maybe add external containers, ie the Unifi Controller.

at the moment my base TrueNAS Core OS is on a thumb drive, with all my physical internal drives allocated to two pools.
Easy enough to push a TrueNAS SCALE OS onto a 2nd thumb drive, and boot onto that... well my first expectation would have been it should have been smooth sailing to import my Pools... reading the above, seems not, and then guessing all my settings/sync jobs etc, none will be there, as I also can't pull in a previous config backup (would have been nice if it was able to import it and indicate the following can be restored and the following not...)

G

 

[cap]

Two questions:
Which kernel (version) is used?
Is PowerTOP included?

"Powertop is a tool provided by Intel to enable various powersaving modes in userspace, kernel and hardware. It is possible to monitor processes and show which of them are utilizing the CPU and wake it from its Idle-States, allowing to identify applications with particular high power demands."
https://wiki.archlinux.org/title/Powertop

 

[ClassicGOD]

Which kernel (version) is used?

Code:

Linux version 5.10.70+truenas (root@tnsbuilds01.tn.ixsystems.net) (gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2) #1 SMP Mon Nov 22 21:32:24 UTC 2021

no powertop from what I can tell.

 

[truecharts]

PowerTop issue added:

https://jira.ixsystems.com/browse/NAS-113898

And PR added as well:

NAS-113898 / 22.12 / Add PowerTop to SCALE by PrivatePuffin · Pull Request #206 · truenas/scale-build

Powertop is a vital Intel-provided tool for monitoring per-process powerconsumption on power-efficiency focussed systems. It also allows access to some intel-adviced optimalisations, which is, ofco...

github.com

 

[Bruce Ho]

When TrueNAS SCALE 22.02-RC.1, I can't use apt-get update command.

 

[NugentS]

TrueNAS is an appliance. Trying to manually update it yourself is an excellent way of breaking things

 

[Bruce Ho]

TrueNAS is an appliance. Trying to manually update it yourself is an excellent way of breaking things

I want to install vim to edit text files.

 

[NugentS]

Try vi instead

 

[truecharts]

TrueNAS is an appliance. Trying to manually update it yourself is an excellent way of breaking things

apt-get update is 100% safe (apt-get update is NOT apt-get upgrade!).
and works fine, the error thrown above is not relevant to apt-get update.

apt-get update && apt-get install powertop works perfectly fine on RC1-2.

 

[NugentS]

apt-get update is 100% safe (apt-get update is NOT apt-get upgrade!).
and works fine, the error thrown above is not relevant to apt-get update.

apt-get update && apt-get install powertop works perfectly fine on RC1-2.

Listen to @truecharts on this one, not me. Could still use vi though

 

[cap]

PowerTop issue added:

https://jira.ixsystems.com/browse/NAS-113898

And PR added as well:

NAS-113898 / 22.12 / Add PowerTop to SCALE by PrivatePuffin · Pull Request #206 · truenas/scale-build

Powertop is a vital Intel-provided tool for monitoring per-process powerconsumption on power-efficiency focussed systems. It also allows access to some intel-adviced optimalisations, which is, ofco...

github.com

Please fix this. PowerTOP is one of the main reasons I want to switch from Core to Scale.