Since about one year I am using Nextbox. I purchased this device to get an easy to maintain cloud solution. In fact, I am only using files, calendar and contacts, nothing special.
From the beginning the box stops working in average every 3 - 5 days. Mostly during the night. I then do a reboot (via SSH) and the box works as designed. Several support cases led to no solution. Since I am not the OS hero, I am unable to dig very deep.
Is there user, who experience a similar issue?
Hi, 
What do you mean by “the box stops working”?
For someone to help you, we would need to understand the problem better. Therefore: Could you describe your problem / the error in more detail?
From your post, I take away, that the RaspberryPi is working fine - You can connect to the box via ssh after all.
So: Does the Nextcloud instance go into maintenance mode, or is it simply no longer accessible at all? Do you have a dynamic DNS connection from outside the network, or do you only have local access to the box?
I receive a message, such as like “No access”, No web access", not maintenance mode. However, the latter happened also, but very rarely. Nextbox is still available by e.g. PING and of course SSH. This behaviour happened so many times, that a script in my Home Assistant reboots the Nextbox, manually triggered.
I am using Nextbox not very intensively. No individual jobs, only the standard delivery of Nitrokey.
Thanks for the clarification.
So, if I understand correctly, the NextBox (hardware) itself remains operational (SSH and ping work), but the Nextcloud web interface becomes inaccessible every few days. Rebooting the device fixes the problem temporarily.
Then, I would need to know, whether you normally access Nextcloud - not the Raspberry Pi itself – using a local IP address / the local only URL “http://nextbox.local”, or through a public hostname/domain (e.g. using DynDNS).
I’m asking because the Raspberry Pi and the OS are obviously still running – otherwise you couldn’t access the box via SSH. So the problem might be either with the NextBox setup or your connection to the webserver.
Especially your description of Nextcloud becoming inaccessible every three to five days during the night leads me to suspect that, if you use a DynDNS service, there might be a problem with updating the DNS records when your ISP changes your public IP address.
Otherwise, to narrow down your problem further, could you tell me exactly what error message you get? Also, the next time this happens, could you connect via SSH and run:
sudo docker ps
This should list all running Docker containers on your NextBox. This could help us determine whether the issue is related to one of the containers.
I am using remote access via [name].dedyn.io
Hi,
I already had some error like access denied. They got triggered by the web server not being able to read the file system. This was because the file system, that is on the ssd plucked in with USB3, was remounted. This remount was because of some USB3 error. These errors occur when your power device was not supplying enough current or if you have some device plugged in at the USB3 port.
If the trouble occurs, in ssh type ‘mount’, to see what devices are mounted. This should be something like /dev/sda/ and /dev/mmcblk0p1. Many other redirection mounts are printed, too, don’t let them confuse you.
Then have a look if the device file is really there.
Also you can have a look at the dmesg log. If the kernel looses the connection to the device, there will be errors about this.
type ‘dmesg’ into ssh console. If to much output use ‘dmesg | less’
Have you found an device error?
Regards,
Thamos
Edit: i am not sure, if the ssd is mounted as root. Maybe the mmcblk-device was root and sda is mounted somewere else. So i relativated that.
Thanks, Robert and thanks, Thamos.
I don’t want to rule out the usb problem, but as Robert access the box via dedyn.io, a DynDNS issue is also a strong candidate (many ISPs assign a new public IP at night, and a reboot of the NextBox also triggers a fresh DynDNS update – which would explain why rebooting “fixes” it).
That said, Thamos’ point is worth checking too. The NextBox’s internal drive is connected to the Raspberry Pi via USB, and the Pi 4 is known for occasional USB/power hiccups that can cause the drive to be remounted read-only. SSH would keep working (the OS runs from the SD card), but Nextcloud would lose access to its data – and again, a reboot would fix it.
Therefore a quick question in advance: when Nextcloud is unreachable and you connect via SSH – do you use the local IP or yourname.dedyn.io? If SSH via the dedyn.io name works while Nextcloud doesn’t, DNS is resolving correctly and we can rule out DynDNS. Alternativly you could compare nslookup yourname.dedyn.io with your current public IP (shown in your router or on a site like ifconfig.me). If they differ → DynDNS didn’t update.
If the dynDNS looks fine,
sudo docker ps -a – are all containers “Up”, or has one exited/restarted?mount | grep sda – does it show ro instead of rw? And dmesg | grep -iE "usb|uas|sda|ext4|i/o error|voltage" for USB resets or I/O errors.Please post the output here – with that we should be able to narrow it down quickly.
Many thanks for your analysis, Saimoen. I uploaded 2 outputs, your required one and an additional one. I am glad to hear your opinion.
Btw. I am accessing the nextbox usually via dedyn.io, but also via the IP address. Both methods are working correctly, in any situation.
Diagnostic Nextbox.zip (4.3 KB)
@RobertHasler The output of the logs you provided suggests the filesystem on the main hard disk (mounted to /srv) somehow went into a unusable/broken state. This could have multiple reasons.
The logs also suggest one possible reason. There is another hard disk connected via USB
/dev/sda1 ext4 1.8T 812G 928G 47% /media/extra-1 . As the NextBox USB ports can not supply enough power to external drives please make sure that the drive is adequately equipped with an external power supply. Otherwise you might risk further filesystem corruption to either of the hard drives (internal and external).
The other hard disc is for backup reason only and has their own power supply
If the additional hard disk is just for backups would it be an option for you to remove it temporarily, just to rule it out completely and see if the error still occurs after some days? Reasoning here is the power supply might not supply enough/stable power, so the external drive could still interfere with USB power somewhat.
I’ve done it already, no change
I found the support tickets. Seems like there may have been other (unrelated) issues with the Nextbox, which made troubleshooting the last time a bit challenging. The new information may help, you will be contacted in the new ticket for details.