Can't login to frontend (happens repeatedly after 2 weeks or so), Flash drive out of space?

[Sander de Ruiter]

Unraid 6.1.49, running stable for 2 weeks or so, and then I can no longer login to the frontend. User/pass are accepted but screen stays on login screen.

 

Running diagnostics over SSH gives the following:

Starting diagnostics collection... tail: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
tr: write error: No space left on device
done.
ZIP file '/boot/logs/tower-diagnostics-20231225-1327.zip' created.

 

I'm assuming it's the flash drive that's out of space? If so, that hasn't happened in the last 4 years and it has plenty of space left (from memory). Any hints?

[itimpi]

2 minutes ago, Sander de Ruiter said:

Unraid 6.1.49, running stable for 2 weeks or so, and then I can no longer login to the frontend. User/pass are accepted but screen stays on login screen.

 

Running diagnostics over SSH gives the following:

Starting diagnostics collection... tail: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
echo: write error: No space left on device
tr: write error: No space left on device
done.
ZIP file '/boot/logs/tower-diagnostics-20231225-1327.zip' created.

 

I'm assuming it's the flash drive that's out of space? If so, that hasn't happened in the last 4 years and it has plenty of space left (from memory). Any hints?

How much RAM do you have?   I think you need at least 4GB to run recent releases of Unraid reliably.

[Sander de Ruiter]

I have 32GB of ram.

[Sander de Ruiter]

How can I check the disk usage?

This is the output I can generate:

root@NAS:/# df
Filesystem      1K-blocks       Used  Available Use% Mounted on
rootfs           15347348   15347348          0 100% /
tmpfs               32768        636      32132   2% /run
/dev/sda1        15614032     913160   14700872   6% /boot
overlay          15347348   15347348          0 100% /lib
overlay          15347348   15347348          0 100% /usr
devtmpfs             8192          0       8192   0% /dev
tmpfs            15361108          0   15361108   0% /dev/shm
tmpfs              131072      16788     114284  13% /var/log
tmpfs             3072220          0    3072220   0% /run/user/0
/dev/md1p1     7811939620 7671723780  140215840  99% /mnt/disk1
/dev/nvme0n1p1 1953513560  190945720 1760428840  10% /mnt/cache
shfs           7811939620 7671723780  140215840  99% /mnt/user0
shfs           7811939620 7671723780  140215840  99% /mnt/user
/dev/loop2       41943040    8896468   32512092  22% /var/lib/docker

 

[itimpi]

The rootfs is 100% used which suggests you have something writing to RAM instead of persistent storage. 
 

running

du -shc /*

Might give a clue

[ctsdad]

Did you set a static IP address. I just asked because your router could have changed the IP address and that would stop access. If so you could go to your routers web access address and look for your server. Also map it as static while you were there.

[trurl]

3 hours ago, itimpi said:

rootfs is 100% used

rootfs is where the OS files are (in RAM as mentioned). If you fill that up, the OS can't work with its own files, so all sorts of things can quit working.

 

You need to figure out what is writing to RAM.

 

Often a misconfigured host path for a container will be the cause.

[Sander de Ruiter]

14 hours ago, trurl said:

rootfs is where the OS files are (in RAM as mentioned). If you fill that up, the OS can't work with its own files, so all sorts of things can quit working.

 

You need to figure out what is writing to RAM.

 

Often a misconfigured host path for a container will be the cause.

That was it (and not for the first time in my Unraid History...). I installed Homepage from the apps section, which had a default path to /mnt/user/appdata/homepage, which I think caused this?

[Kilrah]

It shouldn't, unless you changed defaults and don't have a share named appdata.

[Sander de Ruiter]

4 minutes ago, Kilrah said:

It shouldn't, unless you changed defaults and don't have a share named appdata.

Hmm, I *do* have a share named appdata...

What would be a definitive path to look out for in my container mappings, that certainly will write to RAM?

[Kilrah]

Anything that isn't an actual mount...

Could be e.g. you decided to name your cache pool something else than the default "cache", but left a template default mapping to /mnt/cache/...

[Sander de Ruiter]

20 hours ago, Sander de Ruiter said:

That was it (and not for the first time in my Unraid History...). I installed Homepage from the apps section, which had a default path to /mnt/user/appdata/homepage, which I think caused this?

That wasn't it...but I did find a container that wrote to /var/lib/folder...Moved that to /mnt/cache/appdata/folder solved this. Thanks!