Updated both my NAS’es today.
24.10.2 → 24.10.3 → 25.04.2
No issues so far.
Thnx IX 
Can I ask a couple of naive questions? If one wants to prepare VMs in EE for 25.04.2 what VNC ports are appropriate? I thought I read somewhere that 5901 and above were ok, and 5900 not? And do the VNC passwords have to be valid user passwords on the VM or are they just arbitrary?
Port validation has been in place for a number of releases. Only VMs created before 23.10 are impacted where a port could be set outside the range.
Display device password is a password specifically for accessing that VM, not linked to a user. Release notes forum section latest post has a link to the UI guide for this area.
Will there be a method to migrate Incus VMs back to classic VMs or will they need to be rebuilt? I scanned the change logs and documentation but didn’t find anything. Please let me know if I missed something.
2 Likes
Thanks! I checked the PR for 25.04.2.1, which clarifies what was being checked for!
May I ask if you left your apps and VMs running, or did you shut them down before updating?
1 Like
Mine were running when I updated.
Hey, I think you forgot to mention the fix to SMB server-side copy with Apple clients. If @awalkerix hadn’t mentioned in this thread, I would have thought it was delayed for the next release.
This is a monumental improvement for environments with macOS users.
2 Likes
Is this the fix that doesn’t move data over the cables when using Finder to move data between different locations on the server (only within datasets or even between them?)? If so, does it happen automatically or are any settings needed on TrueNAS and/or MacOS?
Good news !
Since I now does understand the cause of ^not showing up^ of the VM’s,
running 24.10 I did redefine all the display settings of all my VM’s, where I also added a display device to VM’s not having one.
After those changes I did save the config and upgraded to 25.04.2 …
Guess what the VM’s are there.
So there is an issue, but it you make sure you VM’s do have:
- a display device
- a display port above 5900
- a display password
- an assigned display IP
The upgrade seems to work.
What I did NOT test is, what happens if you have VM’s without a display device, like I had before I did this correction action.
During the correction I did found two errors:
- display devices attached to an IP which did not exist any more … which does not seems a problem when running 24.10 …
However … I really do not have any idea what the goal of that address is …
If I go to the VM and open spice … the GUI-address opens … and NOT the address defined as display address …
- as said potentiality a problem if a VM does not have a display device !???
I migrated another machine with a VM not having a display device
No Problem! 
Two machines here. One baremetal with lots of rust, one virtualized, host full of nvme’s. Both updated to the latest, no issues. Haven’t really stressed them much and don’t run VM’s but have a few apps, no weirdness yet.
Yeah, it also uses block cloning if available so the copies are instant and don’t occupy extra space.
AFAIK you don’t have to enable anything, it just works.
Anyone else no longer receiving app update notifications (in the UI or email) when a new image is available for custom apps or community apps after upgrading from 25.04.1 to 25.04.2? Emails notifications for new apps have always been unreliable for me in 25.04.1 (very rarely received), but at least the UI always let me know when a new image was available to update an app. Now, I only see the notification that an update is available if I reboot TrueNAS. It shows immediately after a reboot. Obviously, this is not a viable workaround.
I’ve been on 25.04.2 since it was available and have not seen a single notification for an app update unless I reboot. There’s been at least 3 different instances of updates available for apps between then and today.
Edit: Now even a restart of TrueNAS does not show available updates to apps. Not sure what changed between updates.
I upgraded from 24.10.2.3 to 25.04.2 3 days ago, and have not seen any notifications to upgrade. Apps appear to display update availability normally.
UPDATE: Now I see the notification of an available update as well. It took a few days, but now it seems to be there all the time.
When you say apps appear to display update availability normally, have seen an update available for any of your apps since upgrading to 25.04.2?
All my apps show “Up to date” in the UI currently, but two of them I know for sure have an update available in their repositories as I checked manually. Still no update shown. Even if I go to the " Manage Container Images" tab and try to pull from there nothing changes.
Check the current app version status on the webpage, if you want a second opinion:
If you want updates the moment the original repository updates you will need to move over to custom apps pointing directly at that repository (and even then there will often be a slight delay to to mirrors/CDNs not caching all data immediately).
1 Like
All of my apps are custom apps just so that I have better control of their configs, updates, etc. Their respective docker configs are configured to pull images directly from who publishes them.
I know it’s not an availability issue because spinning up a fresh VM and installing a docker instance pulled the correct newest image. Also, if I go to TrueNAS and spin down the app, then delete its related image from " Manage Container Images", I can use the “Pull Image” function, and it will pull the latest image also.
I checked my DNS server logs for my TrueNAS host overnight and I saw there was no traffic to the repos in a 12-hour period, so it appears to not have even tried to reach out and check for a new image.
Does anyone know the frequency that TrueNAS checks for image updates?
I’ve noticed the same thing…almost all custom apps on my server, but updates are not flagged in a timely manner, often coming in batches of 3 or 4 apps at once, when I know updates have been available for a few days or more.
If you know an update is available, you can force it by having your custom app set to “always pull an image, even if it’s present on the host” then restart it.
1 Like
That’s work-in-progress for Goldeye… in the meantime, you can just keep operating the existing VMs.