Setting questions

On the TrueNAS 25.x system, configuration options often don’t take effect.

For example, settings for UPS, hard drive sleep, IPv6, etc.

The first time I modify them, they work.

But if I modify the related parameters again, they stop working — unless I reboot the system.

Also, with tmux, when I manually run cb -d to start a .sh script in the background, it works fine.

But when I use the command tmux new -d -s disk_monitor bash with the -d option, it only works for a while, and then the .sh script starts constantly throwing errors.

In short, I’ve encountered many problems using TrueNAS.

Sometimes my hard drives have very small read/write activity.
It might be caused by opening the UI, logging in, or clicking on datasets or disk reports — I’m not sure if this is actually a problem.
The system logs don’t show these operations, and I don’t know if there’s an effective way to monitor them.

When using WebDAV for synchronization tasks, it always consumes a large amount of network resources and takes a long time — for example, syncing just a few dozen megabytes of files can involve several gigabytes of data transfer.

It’s prone to errors, and when a sync error occurs, it can cause a lock. This results in subsequent tasks showing as “waiting” indefinitely, even though there are no actual tasks and no real lock. I have to manually restart the related components to fix it.

However, when I manually upload files, there are far fewer problems.

Please post all details of your hardware.

UGREEN DXP4800 Plus
• CPU: Intel Pentium Gold 8505 (12th Gen x86 architecture, 5 cores, 6 threads)
• Memory: 16GB DDR5 4800 (Samsung) ×1 + 32GB DDR5X 4800 ×1
• System Storage: 128GB SSD ×1
• Hard Drives: 4 × Seagate MG08ACA16TE HDDs
• M.2 Slot: 1 × Acer GM7000HS 2TB

I bought the finished machine

I want to describe in detail: while using TrueNAS, I encountered a problem with IPv6. I disabled and re-enabled the network interface, rebooted, and tried many different operations multiple times. IPv6 would show as normal, but websites would return 522. After spending several hours, the final solution was to delete IPv6 and manually copy the generated IPv6. After that, the network worked normally.

Hard drive sleep also didn’t work — changes wouldn’t save at all. In general, it just didn’t work. I didn’t know what to do, but after a reboot, it worked normally.

My UPS could not be used. I set up the NUT service using Docker, entering an IP and port to connect to my UPS. Because it allows additional configuration, I tried entering it and saving the results, but it kept throwing errors. Even closing it in the UI didn’t work. Reverting to previous settings also didn’t help. A reboot was required. In short, my settings frequently fail.

Once I replaced a hard drive — even though the name and path were ultimately the same, SMB didn’t work. I had to delete and re-enter the same path; otherwise, it wouldn’t work.

I have been using TrueNAS for one month.

That are a lot of issues for a single post and you should probably open a single thread for each one if you expect people to help.

And then in each of these threads please provide the technical details like e.g.

So you have a network with IPv6 and after enabling a network interface on TrueNAS your client systems could not connect to web sites, anymore?

That does not make much sense. But then TrueNAS itself does not make connections to web sites, so that 522 error can hardly be experienced by TrueNAS.

  • What does your network look like - especially IPv6 connectivity?
  • What brand and model of router are you using?
  • How is IPv6 configured for clients?
  • What exactly did you configure on TrueNAS?
  • Where exactly did you observe the 522 failure?
  • How exactly did you configure hard drive sleep?
  • What is your pool layout? How many hard drives? What models?
  • How are they connected?
  • Are there more pools than one? Where is your system dataset located?
  • Where is your app dataset located?
  • Why did you install NUT in Docker instead of configuring the UPS in the system via the NUT service?
  • How exactly did you deploy and configure that Docker NUT installation?
  • How exactly did you configure it?
  • What brand and model of UPS do you have?
  • How is it connected to TrueNAS?
    [/quote]
  • How exactly did you replace the hard drive?
  • Was it part of a pool? Did you perform a ZFS replace operation followed by a resilver?
  • How exactly did SMB “not work”? What did you try to do and what error did you get?
  • Where exactly did you delete and re-enter the same path? What does that even mean?

To troubleshoot technical issues anyone trying to support you needs all the technical details. Nobody has got a working crystal ball.

Suggestions for the threads you should open, IMHO:

  • Configuring IPv6 on a TrueNAS interface breaks … <insert what exactly breaks>
  • Hard drive sleep does not work as expected when I do … <whatever you did>
  • TrueNAS NUT service cannot talk to <vendor> <model> UPS via <interface>
  • SMB access from <client system> breaks after disk replacement - how to troubleshoot?

For IPv6, it was already assigned and displayed in the UI network settings, but it could not be accessed from outside. There was no firewall, and rebooting didn’t help. However, when I deleted the assigned IPv6 and manually entered the same IPv6, then rebooted, it worked normally.

When replacing the drive, it was a single solid-state drive. There were some SMB configurations before. Even though the hard drive path did not change after replacement, the SMB share didn’t work. Deleting and re-entering it made it work normally.

Regarding hard drive sleep, my drives are in RAIDZ1. I set them to sleep after 30 minutes, and it worked the first time. But when I changed it to a different number of minutes, the setting didn’t take effect. Only after a reboot would it work.

For UPS, I am using the US3000 model, which is not supported by the system. I compiled one and connected it to the system through an external NUT. It had been running fine, but today I accidentally modified the settings, and the UPS system log reported an error saying something like “us3000 driver service failed to start.” I reverted the configuration back to my previously working setup, but the system log still kept reporting errors. I disabled the UPS service from the system UI, but when checking via command line, it was still running, and the log continued to report errors. I manually stopped the UPS service via command line.

What I want to say is that operations in the UI often have no effect. Even if something cannot be used, there is no prompt or warning. Right now, my system works fine as long as I don’t perform operations through the UI — otherwise, it’s likely to not take effect.

What I mean is that when you perform operations in the UI, the first time can succeed, but any further modifications will not be saved or take effect — even if the parameters are exactly the same, they will not be saved.

If you modify related things, for example replacing a hard drive and copying the files to be exactly the same, even with the same path, some changes will take effect and some will not. For example, for the dataset path of an SMB share — after a reboot or when opening it, even if permissions are fine, it will not work. You must delete it, reselect the same path, and save it for it to work.

I have solved all these issues. What I’m saying is that operations in the UI often don’t actually work as shown. It may display that a service is stopped but it’s still running. It may display “saved successfully” but it doesn’t work, and sometimes a reboot is required.

Currently, my approach to these problems is:
• Only make the first change, and avoid saving a second modification.
• If rebooting doesn’t fix it, delete the item, then repeat exactly the same operation and save again.
• If it still doesn’t work, a reboot will make it normal.

It’s very strange.

Again - what does that even mean? TrueNAS picked up an IPv6 address automatically? Fine - that’s how it should work.

What was that address? A link local or a GUA one?

What do you mean by “accessed from outside”? From the Internet? Hopefully your router prevented that.

Etc …

Without showing the address and describing what you mean by “outside” and by “access” (browser? ping? SMB?) - this is not diagnosable.

Then that drive “a single solid-state drive” - so you built a ZFS pool from a single drive? OK, although not recommended. Then you recreated that ZFS pool under the same name? And the SMB share did not work? Rebooting or restarting the SMB service would probably have helped.

Also did you create the share on the top level “pool name” dataset? You should not do that. Always create at least one dataset first and share that.

Etc. … this is really tiresome. I can assure you settings from the UI do work and most probably every problem you encounter is particular to your specific configuration. So without sharing this configuration - no chance of help.

E.g. what was that IPv6 address and what do you mean by “access”?

Did you check with

nut-scanner -U

from shell which driver your ups expects? Even if it’s not listed explicitly, if its something like the usbhid-ups driver you can simply select any ups vendor that lists that driver and it should work

The same ipv6 address is automatically assigned, and the other is fixed manual input. The static ipv6 address is the same, but when it is automatically generated, sometimes it cannot be accessed from the outside. Manual input can be saved outside to connect.

Until you define “accessed” and “outside” I’m out of this discussion.

These commands don’t work. They can’t be detected at all. I bought a brand machine and UPS of the same brand before, but I changed the system and it doesn’t support it anymore. I built a nut service with docker and filled it in the system for control.

They work here:

root@truenas[~]# nut-scanner -U

Scanning USB bus.

Are you even using TrueNAS? :man_shrugging:

I have an anti-generation service. One day, I modified the ipv6 settings and there was a problem. It can only be accessed by the local machine and the local area network. The website cannot be accessed. The ipv6 external, that is, other devices, the Internet, cannot be accessed. The ping is not connected. But I copied the generated ip to the static ipv6. The website is normal. Ping is also possible.

What is that and how does it relate to TrueNAS?

What were the settings before the modification? What were the settings after the modification? What exactly was the problem?

What is “it”? If it is TrueNAS of course it can only be accessed locally. It is not advisable to connect TrueNAS to the Internet for access. Not at all.

Which website?

What is “the IPv6 external”? From which system do you try to access it? What is your network topology, your router, your ISP uplink, your IPv6 connectivity model?

What website?

None of this makes any sense. :frowning_face:

according to that github issue you linked, the ups expects the usbhid-ups driver. I’d try connecting it to truenas, select any vendor that lists that driver and see if you can communicate with it, nothing should stop it from doing it.

root@truenas[~]# nut-scanner -U

Scanning USB bus.

root@truenas[~]# nut-scanner -U

Scanning USB bus.

root@truenas[~]#

This is the docker I made. It can work normally