Skip to content

Latest commit

 

History

History
62 lines (53 loc) · 3.42 KB

haos_guide.md

File metadata and controls

62 lines (53 loc) · 3.42 KB
Raw

Fix the problem that the dependencies cannot be installed

Important

This custom component installation is considered advanced
should only be used if one is an expert in managing a Linux operating system.

Important

Minimum system requirements your device must meet to install dependencies:
RAM: 4GB, If your device has less than 4GB of memory, A memory error may have occurred. Storage: 64 GB or larger storage device.

SSH Add-on (Recommended)

  1. Go to the Add-on store
    Open this add-on in your Home Assistant instance.
  2. Install one of the SSH add-ons (you need to enable advanced mode in your user profile to see here)
  3. Configure the SSH add-on you chose by following the documentation for it
  4. Sure disable protection mode and Start the SSH add-on
  5. Connect to the SSH add-on
  6. Copy and paste into the Terminal to enter Home Assistant CLI
login
docker exec -it homeassistant bash

image 7. Copy and paste into the Terminal to install the dependencies

apk add e2fsprogs musl-dev gdal-dev proj-dev proj-util gcc g++ gfortran openblas-dev linux-headers
fallocate -l 4G /usr/tmp-disk
mkfs.ext4 /usr/tmp-disk
mount -o loop -t ext4 /usr/tmp-disk /tmp
pip install --no-cache-dir pandas==2.1.4 geopandas==0.14.4 matplotlib==3.9.0 scipy==1.12.0 obspy==1.4.0

image 8. If everything is successfully, Continue configuration the integration.

Note

Excessive memory use when install Matplotlib 1, A workaround is to run pip with --no-cache-dir to avoid the cache.


Docker Terminal

  1. Open a terminal and login
  2. Go inside the container with docker exec -it homeassistant bash (or similar)
  3. Copy and paste into the Terminal to install the dependencies
apk add e2fsprogs musl-dev gdal-dev proj-dev proj-util gcc g++ gfortran openblas-dev linux-headers
fallocate -l 4G /usr/tmp-disk
mkfs.ext4 /usr/tmp-disk
mount -o loop -t ext4 /usr/tmp-disk /tmp
pip install --no-cache-dir pandas==2.1.4 geopandas==0.14.4 matplotlib==3.9.0 scipy==1.12.0 obspy==1.4.0
  1. If everything is successfully, Continue configuration the integration.

Note

Excessive memory use when install Matplotlib 1, A workaround is to run pip with --no-cache-dir to avoid the cache.


Footnotes

  1. This error is coming up because, it seems, pip's caching mechanism is trying to read the entire file into memory before caching it… which poses a problem in a limited-memory environment, as matplotlib is ~50mb. ↩ ↩2