8e34b1779b
Follow-up to #1382. Under the EOS server the end of a GENETIC run with parallel evaluation (workers >= 2, which is the automatic setting on machines with 3 or more cores) could hang forever in Pool.terminate()/join(), so /v1/optimize never answered. Cause: uvicorn installs a Python SIGTERM/SIGINT handler in the server process that only asks the server to shut down. The forked evaluation workers inherit it and survive the SIGTERM of Pool.terminate(). terminate() keeps the task queue's read lock (Pool._help_stuff_finish), so a surviving worker that has not yet taken its stop sentinel waits for that lock forever, and join() with it. This always happened after a batch timeout (busy workers survive) and by chance after a normal run when a worker was not scheduled in time (several EOS instances or other load; also seen with a single instance on a 2-vCPU VM). A worker that got SIGTERM before its initializer ran hit the same path. - Workers ignore SIGTERM and SIGINT. Sent to the whole process group (Ctrl-C, systemd stopping the service) they are meant for the server, which finishes or aborts the run and stops the workers itself. Restoring the default action instead was measured to be worse: a group SIGTERM killed the workers mid-batch, the batch ran into its 120 s timeout and the server did not exit within 120 s (12 s before). - _EvaluationPool stops its workers with SIGKILL at the point where Pool.terminate() would send SIGTERM (it already holds the read lock). This does not depend on handlers, initialization state or whether a worker is busy. _stop_evaluation_pool() is unchanged. Measured (8 EOS servers x 2 workers on 8 vCPU, real configuration, seed 42): 11 hangs in 160 runs before, 0 in 440 after; pool only under load: 107/300 before, 0/600 after. Plans and costs are identical, the stop takes no longer. Ctrl-C, SIGTERM to the server and SIGTERM to the group behave as before (run completes with HTTP 200, exit after ~12 s, no orphaned workers). Tests (tests/test_genetic_parallel_evaluation.py, under uvicorn's real signal handling): inherited handler, batch timeout, late worker start, late idle worker (SIGSTOP, Linux) and SIGTERM to the group during a batch. They fail without the fix. Claude-Session: https://claude.ai/code/session_01EYWR1HVM8a8src23xuyL3T Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Build optimized energy management plans for your home automation
AkkudoktorEOS is a comprehensive solution for simulating and optimizing energy systems based on renewable sources. Optimize your photovoltaic systems, battery storage, load management, and electric vehicles while considering real-time electricity pricing.
Why use AkkudoktorEOS?
AkkudoktorEOS can be used to build energy management plans that are optimized for your specific setup of PV system, battery, electric vehicle, household load and electricity pricing. It can be integrated into home automation systems such as NodeRED, Home Assistant, EVCC.
🏘️ Community
We are an open-source community-driven project and we love to hear from you. Here are some ways to get involved:
-
GitHub Issue Tracker: discuss ideas and features, and report bugs.
-
Akkudoktor Forum: get direct suppport from the community.
What do people build with AkkudoktorEOS
The community uses AkkudoktorEOS to minimize grid energy consumption and to maximize the revenue from grid energy feed in with their home automation system.
- Andreas Schmitz, the Akkudoktor, uses EOS integrated in his NodeRED home automation system for OpenSource Energieoptimierung.
- Jörg, meintechblog, uses EOS for day-ahead optimization for time-variable energy prices. See: So installiere ich EOS von Andreas Schmitz
Why not use AkkudoktorEOS?
AkkudoktorEOS does not control your home automation assets. It must be integrated into a home automation system. If you do not use a home automation system or you feel uncomfortable with the configuration effort needed for the integration you should better use other solutions.
Quick Start
Upgrading an existing installation?
The next release includes a substantial energy-planning update: 15-minute GENETIC plans, EV departure targets, flexible household loads, optional battery export, and better measurement and PV-forecast tools. These features are currently unreleased.
Existing Home Assistant automations, Node-RED flows and custom API clients may need changes.
Devices now use stable IDs, the new optimizer reads hardware settings from EOS configuration,
and incomplete forecasts or outdated battery measurements can stop a run. The legacy
POST /optimize remains available with GENETIC0; the new POST /v1/optimize has a different
request format. Back up your configuration and data before updating.
Read the upgrade guide and changelog before switching an existing installation.
Start with Docker
Run EOS with Docker (access dashboard at http://localhost:8504):
docker run -d \
--name akkudoktoreos \
-p 8503:8503 \
-p 8504:8504 \
-e OPENBLAS_NUM_THREADS=1 \
-e OMP_NUM_THREADS=1 \
-e MKL_NUM_THREADS=1 \
-e EOS_SERVER__HOST=0.0.0.0 \
-e EOS_SERVER__EOSDASH_HOST=0.0.0.0 \
-e EOS_SERVER__EOSDASH_PORT=8504 \
--ulimit nproc=65535:65535 \
--ulimit nofile=65535:65535 \
--security-opt seccomp=unconfined \
akkudoktor/eos:latest
System Requirements
- Python: 3.11 or higher
- Architecture: amd64, aarch64 (armv8)
- OS: Linux, Windows, macOS
Note
: Other architectures (armv6, armv7) require manual compilation of dependencies with Rust and GCC.
Installation
Home Assistant add-on
To install the Akkudoktor-EOS add-on in Home Assistant:
-
Add the repository URL:
In Home Assistant, go to:
Settings → Add-ons → Add-on Store → ⋮ (top-right menu) → Repositoriesand enter the URL of this Git repository:
https://github.com/Akkudoktor-EOS/EOS -
Install the add-on:
After adding the repository, the add-on will appear in the Add-on Store. Click Install.
-
Start the add-on:
Once installed, click Start in the add-on panel.
-
Access the dashboard:
Click Open Web UI in the add-on panel.
-
Configure EOS (optional): In the dashboard, go to:
Config
Docker (Recommended)
docker pull akkudoktor/eos:latest
docker compose up -d
Access the API at http://localhost:8503 (docs at http://localhost:8503/docs)
From Source
git clone https://github.com/Akkudoktor-EOS/EOS.git
cd EOS
Linux:
python -m venv .venv
.venv/bin/pip install -r requirements.txt
.venv/bin/pip install -e .
.venv/bin/python -m akkudoktoreos.server.eos
Windows:
python -m venv .venv
.venv\Scripts\pip install -r requirements.txt
.venv\Scripts\pip install -e .
.venv\Scripts\python -m akkudoktoreos.server.eos
Configuration
EOS uses EOS.config.json for configuration. If the file doesn't exist, a default configuration is
created automatically.
Custom Configuration Directory
export EOS_DIR=/path/to/your/config
Configuration Methods
- EOSdash (Recommended) - Web interface at
http://localhost:8504 - Manual - Edit
EOS.config.jsondirectly - API - Use the Server API
See the documentation for all configuration options.
Port Configuration
Default ports: 8503 (API), 8504 (Dashboard)
If running on shared systems (e.g., Synology NAS), these ports may conflict with system services. Reconfigure port mappings as needed:
docker run -p 8505:8503 -p 8506:8504 ...
API Documentation
Interactive API docs available at:
- Swagger UI:
http://localhost:8503/docs - OpenAPI Spec: View Online
Resources
Contributing
We welcome contributions! See CONTRIBUTING for guidelines.
License
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.