[Help] NAS Keeps “502” erroring shortly after a clean install

[benfozard]

Hi,

The other day, my F4-424 TOS_X642.0_6.0.673 NAS suddenly said "user does not exist" when trying to log in. I safely powered down and then on reboot it does not come back up, only giving me a "502 Bad Gateway" when I try and log in. The TNAS app also does not allow me to log in.

The NAS has been 100% stable since I got it in January 2024 and is running the latest TOS since I upgraded to TOS 6 many months ago.

My boot/system drives are two M2 SSDs, the only way I could get the device to do a factory reset/initiate the boot loader (there's no reset button on my device!) is to delete the partitions on each of these and do a clean install, removing the HDDs with the data on from the bays. This works. Once TOS is installed, I can then plug in my HDDs and re set up the system. However, after about an hour or so, the desktop becomes un-responsive and I get the same error, user does not exist, and then on restart, 502 and the device is unusable.

There's no apps installed as I never get chance to install them. Anyone any ideas what's going on here and what is a clean install of TOS (the existing HDDs do not contain the system)?

[OrionXie]

We are very sorry for the bad experience. If your device can still access the terminal via SSH, you can try the following operations:
1.
Go to the root directory of the device to check which files are occupied
```bash
cd /
du -sh *
```
2.
Go to the etc directory to check which files/directories occupy a large space
```bash
cd /etc
du -sh *
```
3.
Delete the large occupied `P` file
```bash
rm /etc/p
```
After deletion, refresh the system. Deletion may take some time as the file is relatively large.

If you cannot solve the problem on your own, please keep communicating with technical support via email or LiveChat.

[benfozard]

So, I did a clean install again, enabled ssh from the start. The system was stable until after a few hours I could not install apps. I rebooted, and got a 502.

I was able to ssh in and ran "rm /etc/p", followed by "reboot". This worked! I also captured the results of the other two commands if you need them too, but "/etc/p" was "5.5G" in size.

However, I then configured the network adaptor to use bind both ports (which is the configuration I have been running since I got it) and rebooted and got a "502". Sadly, I can no longer ssh in (maybe changing the network interface has turned off ssh?) so my device is dead again.

<UPDATE> I plugged in a keyboard and monitor direcly into the NAS and was able to delete "/etc/p" that way to get my device to boot again. It's only a matter of time before it dies again though. <UPDATE>

Bottom line is, what is causing this issue to keep happening? The NAS has been stable for 18 months.

[TMjack]

We sincerely apologize for the inconvenience caused by this issue. After confirmation, it has been identified as a bug related to the backup system configuration. Our technical team has prioritized the fix, and it is scheduled to be resolved in an upcoming version. We truly appreciate you sharing your workaround and thank you for your continued support of TerraMaster!

[benfozard]

Many thanks. I will hold off backing up for now, but have a workaround if I need to to get it to boot if it happens again. I look forward to the fix.

[OrionXie]

Thank you for your understanding. We have prioritized this issue and expect to fix it in the next version.