TOS 6.0.460 (X86) Beta is Released for Update

[abs1ck]

What is your device model? What TOS version were you using before upgrading to version 461? What type of array did you set up? Have you performed any other operations previously?

Hi, I have an F4-424 Pro, I update as soon as the update comes out. Raid array I have a TRAID, never had such a thing. For the first time this started happening on assembly 461, even on 460 there was no such thing, it worked quietly, but here it just turns off stably once every 1-2 days. I have exported a zip archive for contacting via mail, maybe you can find something.

You can send the System report to: [email protected]

Yes of course. I have already sent it, the report was created immediately after turning on, after randomly turning off the device

[DanielB]

Applicable Models:
F2-221, F2-223, F2-422, F2-423, F4-424 F4-221, F4-223, F4-421, F4-422, F4-423, F4-424, F4-424 Pro
F5-221, F5-421, F5-422 T6-423 F8-421, F8-422 T9-423, T9-450 T12-423, T12-450

I have a F2-424 and am still running v6.0.229 as my model has been left off the list the last couple of TOS 6 update posts. Was that by accident or is there another reason?

[TMroy]

That is incorrect. I have mounted exfat, NTfs and ext4 partitions. It does seem a bit clunky, but you do have to select the actual partitions (not DISK) before selecting the 'mount disk' function. You may also have to reboot device for the mount to be recognised.

Sorry, no Partition, no mount.

The feature of 'Mount Disk' has certain conditions for use. The disk you want to mount must have partitions, and there cannot be any volumes on the partitions. The file system needs to be directly created on the partitions.

[OttoG]

How do we determine the manufacturing date?

It's within the device's serial number. The first two numbers after the second hyphen is the year, the next two are the week. Yours would need to be 2025 or higher to not be effected by the limitation

Thanks for the info. My primary NAS is 2105 so it should be okay. I'll hold off until the release is out of beta before updating in case that changes. The other is the older of my two NASs and probably won't be compatible. It's in a remote location and I haven't been able to check it yet.

[DCCentR]

Today I have some strange issue with one of my HDD. I have two ST22000NM001E-3HM103 (on latest FW) HDD's in RAID1. One in slot 1 and second one in slot 2:

- HDD in slot 1 sudenly dropped out: system starts showing that RAID1 is degraded, HDD1 stops properly shows in "Disk > Overview". There was only that - Good x1, Epmty x2, In risk x0, Defective x0 (disk in slot 1 didn't count in any of this types O_o) (pic - https://i.imgur.com/jGj8yzu.png);
- But it still shows in "Disk > Disk", without "Normal" label (pic - https://i.imgur.com/SEgxfxJ.png);
- And both disk shows as "Good" in my degraded storage pool (pic - https://i.imgur.com/6Vbaa0N.png);
- Buzzer is beeping all that time and led of slot 1 shows as static green;
- S.M.A.R.T. Info shows empty on that disk. Starting the S.M.A.R.T. test stays in "Testing..." status.

After rebooting the NAS disk in slot 1 didn't show anywhere in UI and RAID1 was still in degraded state. Led of slot 1 shows as static green. I took the HDD out of slot 1 and put it back in. It start showing in UI as "Normal". S.M.A.R.T. check went normally and didn't show any abnormal red values on it. After that I rebooted NAS again and it start rebuilding process.

I am on 6.0.461 F4-424 Pro. I will generate the arhive with logs and send them to support mail. Disk's are new and working in NAS about four months.

[TMzethar]

We observed the screenshots you provided and the system report in your mailbox. Unfortunately, they were not enough to pinpoint the cause of the problem. Please continue to communicate with the technical support e-mail address to schedule a time for an engineer to investigate the issue remotely.

[crisisacting]

The other is the older of my two NASs and probably won't be compatible. It's in a remote location and I haven't been able to check it yet.

@TMzethar, is there a method through SSH to retrieve the device serial number?

Couldn't find it anywhere within the web interface & dmesg (via SSH) didn't seem to have it anywhere either; would be nice to have it visible within the super user profile on the web interface, when in a situation like OttoG's.

[bonerserk]

We observed the screenshots you provided and the system report in your mailbox. Unfortunately, they were not enough to pinpoint the cause of the problem. Please continue to communicate with the technical support e-mail address to schedule a time for an engineer to investigate the issue remotely.

I've been dealing with this exact problem for a while. I reported it and sent a log report, and I've already tried what the TM engineers suggested (changing disk slots, reinstalling TOS, recreating the RAID), but with no result.

After many tests, I believe that TOS has some incompatibility with the sleep disk status of Seagate disks. I also have Seagate disks ST8000VN004, which should be fully compatible with TOS according to the official pages. When the disks go to sleep and something tries to write or read, sometimes these disks can't wake up, and TOS thinks the RAID is corrupted.

I was able to replicate this issue many times: I have installed Immich (a photo manager) on a server, which has its photo library on my RAID. When all disks are sleeping, if I take a photo with my phone, the Immich app will upload the photo, and the Immich server tries to write it directly to the folder on the RAID. As soon as it tries to write the photo, the RAID enters degraded mode. Less frequently, this also happens when trying to read a file, like a film.

When this happens, I can hear the NAS beeping, and soon after, I can hear the disks starting their reading and wake-up noises. I had installed TOS only on an SSD NVMe disk.

Now, I have reinstalled TOS on the problematic Seagate disk instead of the SSD disk, and it’s been three days without this problem occurring. However, I think that installing TOS on the RAID disks prevents them from going to sleep. I can always hear disk noises, and files are always available with no waiting time, unlike when the disks are really asleep.

When the degraded RAID status occurs, if I shut down the NAS and then start it again, everything works correctly as if nothing happened. This method (shutting down) prevents the RAID rebuilding process, unlike what happens with a reboot.

I want to ask @crisisacting if you want to reinstall TOS on the disks to verify if the problem persists.

I also noticed that this issue is not only related to RAID but also to single disk configuration. I put one of my Seagate disks as a single disk, and the problem is the same. TOS can't wake up the disk, and I can't access files.

I am available to provide any information and support to the TM team to help find and solve this problem.

Thanks.

[Gremlin]

In TOS6 the device ID is shown on Overview page in Hex. I wonder if this translates to the Serial no

[Gremlin]

I had installed TOS only on an SSD NVMe disk.

That might be the intention, but I'm not certain it is the reality. My testing has shown that TOS will still be installed on ALL the drives, but that some may be "inactive" (and thus not updated). Have you checked the drive partitioning? Have you tried viewing a console screen when booting - you can see which arrays are being rebuilt. On my 424 there is always a distinct pause (seconds) while the os array is 'rebuilt'* and mounted.

In any case, I work from the presumption that the nas is a server which may be accessed at any time by outside clients (mine is primarily a backup repository) and really needs to be awake. On my test machine I have, currently been mounting drives as 'external' disks (TM wording) in the spare drive slots with a variety of file systems (without the TOS bits) created on another system. Those could easily sleep if needed. (I'm not sure how a raid would do in that scenario but presume it would not).

Have you considered just disabling the sleep function (in control panel) just to ensure that is the problem?

*"re-assembled" might be a better word