Need help with excessive disk usage

Configuration
1

Hello,

Can someone please help me figuring out what is causing this weird excessive disk usage?

➜ ~ cd /
➜ / du -hs
du: can’t open ‘./proc/573/task/573/fdinfo’: Permission denied
du: can’t open ‘./proc/573/fdinfo’: Permission denied
66.3G .
➜ /
➜ / df -h
Filesystem Size Used Available Use% Mounted on
overlay 109.3G 109.1G 0 100% /
/dev/sda8 109.3G 109.1G 0 100% /addons
/dev/sda8 109.3G 109.1G 0 100% /backup
/dev/sda8 109.3G 109.1G 0 100% /share
/dev/sda8 109.3G 109.1G 0 100% /media
/dev/sda8 109.3G 109.1G 0 100% /data
/dev/sda8 109.3G 109.1G 0 100% /config
devtmpfs 1.7G 0 1.7G 0% /dev
tmpfs 1.9G 0 1.9G 0% /dev/shm
/dev/sda8 109.3G 109.1G 0 100% /ssl
/dev/sda8 109.3G 109.1G 0 100% /etc/resolv.conf
/dev/sda8 109.3G 109.1G 0 100% /run/audio
tmpfs 758.4M 1.8M 756.6M 0% /run/docker.sock
tmpfs 758.4M 1.8M 756.6M 0% /run/dbus
/dev/sda8 109.3G 109.1G 0 100% /etc/hosts
/dev/sda8 109.3G 109.1G 0 100% /etc/hostname
/dev/sda8 109.3G 109.1G 0 100% /etc/asound.conf
tmpfs 1.9G 0 1.9G 0% /dev/shm
tmpfs 758.4M 1.8M 756.6M 0% /run/log/journal
/dev/sda8 109.3G 109.1G 0 100% /var/log/journal
/dev/sda8 109.3G 109.1G 0 100% /etc/pulse/client.conf
tmpfs 1.9G 0 1.9G 0% /proc/asound
devtmpfs 1.7G 0 1.7G 0% /proc/keys
devtmpfs 1.7G 0 1.7G 0% /proc/latency_stats
devtmpfs 1.7G 0 1.7G 0% /proc/timer_list
tmpfs 1.9G 0 1.9G 0% /sys/firmware
➜ /

Once I reboot the host the disk utilization goes back to normal around 70% (I do have some files in the smb share folder), but after couple of hours it climbs back quickly to 100%.

I added these to the configuration.yaml, but it did not help:

system_log:
max_entries: 100

recorder:
commit_interval: 30

image
image684×624 22.5 KB

Thank you for any suggestions!

2 
du --maxdepth=1|sort -n

Then rinse and repeat through the biggest dir.

3

Unfortunately --maxdepth=1 isn’t working. I’m running Home assistant OS, maybe that’s why… also du shows only 60GB disk usage out of 110GB, meanwhile HAOS and df are showing 100% disk usage.

➜ / du -hs
du: can’t open ‘./proc/573/task/573/fdinfo’: Permission denied
du: can’t open ‘./proc/573/fdinfo’: Permission denied
66.3G .

4

Probably unused images. They can easily eat a lot of space. Install portainer addon and check for unused images and or stop containers.

5

Installed Portrainer, images taking up only 5.5 GB

image
image1563×825 86.1 KB

6

Yes on HAOS du is the version supplied by busybox, which is a bit different. For that version, try

du -d 1

If it is filling rapidly and reverts when you restart, it is probably something temporary, like a runaway log file in a container. But it won’t be unused images, as they are only installed when you update docker image.

7

No need to install portainer - docker does all that.

8

Yes, exactly. I tried to drill down using du -d 1 yesterday, but du was just not even showing that I’m utilized at 100%. Anyways I found out what was causing it. It was the google drive backup addon, as I forgot to exclude the /share folder (even though I thought I did). After excluding this folder (it contains about 60GB of data), all is working fine!
Thank you for your help and time.

image
image673×645 22.5 KB

9

Interesting, as the bard said, “All’s Well That Ends Well” :slight_smile:

1 Like