Hassio on Proxmox, huge disk space usage for what?

Installation Home Assistant OS
1

Hi all

I have Hassio installed via Tteck script in Proxmox 8.1.4.

The VM has 42GB disk configured like this
Schermata 2024-02-09 alle 09.36.30

From HA System monitor this is what I see

Schermata 2024-02-09 alle 09.38.45
Schermata 2024-02-09 alle 09.38.45733×354 22.3 KB

But if I go in SSH and do the command

du -h -d 1

I get

488.7M ./var
4.0K ./mnt
0 ./proc
628.0K ./run
292.7M ./usr
2.3M ./bin
4.0K ./opt
7.8M ./media
2.3M ./etc
0 ./dev
7.0M ./lib
0 ./sys
4.0K ./home
4.0K ./srv
324.0K ./sbin
14.3M ./root
76.0K ./tmp
44.0K ./data
4.0K ./addon_configs
365.9M ./homeassistant
2.3G ./backup
3.1M ./share
14.0M ./ssl
4.0K ./addons
4.0K ./lib64
5.5M ./package
52.0K ./command
3.4G .

Where are all the other GBs ?
27.9GB - 3.4GB = 24.5GB missing

Thanks

2

You are probably running df in the home assistant container, not on the OS.

3

I think that Hassio is not installed in docker, otherwise I could not install any additional components that are instead installed in docker

Or if ssh have access only to the HA part(ition), where can I access to the whole disk and try to understand who is consuming disk

4

HA OS (hassio retired more then 3 year ago) is a minimal linux os built on buildroot that has one job: running a bunch of docker containers, including a Supervisor dockor container, a Home Assistant docker container, a DNS docker container, etc. Add-ons are nothing more than docker containers with some extra glue to be able to be managed by Supervisor.

If you want to access the host, SSH to port 22222, see

1 Like
5

Ok, thanks
I’ve got DEV access on SSH using port 22222

This is what I found
Most of the used space is in /mnt/data/supervisor and in /mnt/data/docker

# # du -h -d 1 /mnt/data
# 12.7G   /mnt/data/supervisor
-sh: 12.7G: not found
# 16.0K   /mnt/data/lost+found
-sh: 16.0K: not found
# 496.0M  /mnt/data/logs
-sh: 496.0M: not found
# 17.4G   /mnt/data/docker
-sh: 17.4G: not found
# 33.2G   /mnt/data
-sh: 33.2G: not found
#

Inside supervisor with ** the consuming space

# du -h -d 1 /mnt/data/supervisor/
12.0K   /mnt/data/supervisor/dns
168.0K  /mnt/data/supervisor/apparmor
4.0K    /mnt/data/supervisor/emergency
**2.7G    /mnt/data/supervisor/mounts**
44.0K   /mnt/data/supervisor/audio
4.0K    /mnt/data/supervisor/addon_configs
360.7M  /mnt/data/supervisor/homeassistant
4.0K    /mnt/data/supervisor/.mounts_credentials
7.8M    /mnt/data/supervisor/media
**2.3G    /mnt/data/supervisor/backup** this is my needs
3.1M    /mnt/data/supervisor/share
14.0M   /mnt/data/supervisor/ssl
**5.7G    /mnt/data/supervisor/addons**
**1.7G    /mnt/data/supervisor/tmp**
12.7G   /mnt/data/supervisor/

First of all, is there a way to clear /mnt/data/supervisor/tmp ?

In the /mnt/data/supervisor/addons I found that the most space consuming is the esphome addon, and particularly in the folder
cache
idf_components

I will try to remove the esphome project that I don’t use (anymore) and see if something changes.

Regarding the /mnt/data/docker/, the space is on folder /mnt/data/docker/overlay2

Of course, I’m NOT going to do anything from this SSH session, I’m just analyzing

Any suggestions?

Thanks

6

That is where docker stores it’s containers.

1 Like