4lt/docker-compose-cookbooks
1
0
Fork 0
mirror of https://github.com/nickyeoman/docker-compose-cookbooks.git synced 2026-09-03 18:36:22 +00:00
Code Issues Packages Projects Releases Wiki Activity

ollama ready, handbrake now notes

Browse Source
This commit is contained in:
nick
2026-04-17 15:23:04 +00:00
parent b13751976f
commit 83b1aa966b
12 changed files with 166 additions and 93 deletions
Download Patch File Download Diff File Expand all files Collapse all files
handbrake/README.md → handbrake_notes/README.md
+4
View File
@@ -1,4 +1,8 @@
# Handbrake
Handbrake requires changing the env file and modification of a configuration file.
I've decided these factors are not ideal for a dockhand deployment.
However all the information you need are in these notes, they just need to be modifed for proper use.
- DockerHub: https://hub.docker.com/r/jlesage/handbrake
- Env docs: https://github.com/jlesage/docker-handbrake
handbrake/compose-cpu.yaml → handbrake_notes/compose-cpu.yaml
View File
handbrake/compose-gpu.yaml → handbrake_notes/compose-gpu.yaml
View File
handbrake/sample.env → handbrake_notes/sample.env
+4 -2
View File
@@ -1,6 +1,8 @@
VOL_PATH=/data
HANDBRAKE_IMAGE=jlesage/handbrake:latest #CPU
#HANDBRAKE_IMAGE=zocker160/handbrake-nvenc:18x #GPU
#CPU
HANDBRAKE_IMAGE=jlesage/handbrake:latest
#GPU
#HANDBRAKE_IMAGE=zocker160/handbrake-nvenc:18x
#NVIDIA_VISIBLE_DEVICES=all #GPU
#NVIDIA_DRIVER_CAPABILITIES=all #GPU
HANDBRAKE_RESTART=unless-stopped
ollama/README.md
+106
View File
@@ -0,0 +1,106 @@
# Ollama with Webui
![Ollama](https://i.4lt.ca/git/ollama.png)
There is a cpu version and a gpu version.
For ollama I'm just running latest, no version lock.
Same with standard ports, no option to change.
Dockerhub: https://hub.docker.com/r/ollama/ollama
Github (instructions): https://github.com/open-webui/open-webui
Reverse Proxy Port webui: 8080 (The docker-compose sets the port to 3000)
## Preperation
You need to have nvidia-drivers, cuda and nvidia-container-toolkit installed before proceeding.
docker run for refrence:
```
docker run -d --runtime=nvidia --gpus=all -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama
docker exec -it ollama ollama pull llama3
docker exec -it ollama ollama run llama3
```
## Environment
WEBUI_SECRET_KEY is just a random secret string used to secure sessions, cookies, and auth in your WebUI. ```openssl rand -hex 32```
## Use
- [Ollama Models](https://ollama.com/search)
When using the webui for the first time after an upgrade, you will need to wait a long time. This is due to the model pulling process.
### Docker Container Cheatsheet
You can alias:
```bash
alias ollamac="docker compose exec -it ollama ollama"
#example then would be
ollamac list
```
Standard commands
```bash
# Add/Pull a model
docker compose exec -it ollama ollama pull llama3.1:8b-instruct-q4_0
# list models
docker compose exec -it ollama ollama list
# Remove a model
docker compose exec -it ollama ollama rm <model-name>
# Run CLI
docker compose exec -it ollama ollama run deebseek-r1:8b "What is Star Trek?"
```
### API
```bash
curl http://ollama.lan:11434/api/tags
```
# Notes
Straight docker run:
```bash
docker run -d -p 3000:8080 --gpus all --add-host=host.docker.internal:host-gateway -v webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:cuda
```
## Recommended Pulls
For a NVIDIA GeForce RTX 5060 Ti, your GPU may vary.
```
Generating prompts for InvokeAI
ollama pull llama3.1:8b
ollama pull mistral:7b-instruct
ollama pull mythomax-l2:13b
Generating PHP code
ollama pull codellama:13b-code
ollama pull deepseek-coder:6.7b-instruct
ollama pull wizardcoder:15b
Help with writing a book
ollama pull llama3.1:70b-instruct
ollama pull mythomax-l2:13b
ollama pull openhermes2.5-mistral:7b
Help with psychology questions/learning
ollama pull samantha-mistral:7b
ollama pull openchat:7b
ollama pull nous-hermes2-mixtral:8x7b-dpo
n8n agents
ollama pull qwen2.5-coder:7b-instruct
ollama pull llama3.1:8b-instruct
ollama pull gorilla-openfunctions-v2:7b
Summarizing documents
ollama pull granite3.2:8b-instruct
ollama pull cognitivetech/obook_summary:7b
ollama pull bge-m3:8b
```
ollama/compose-cpu.yaml
+22
View File
@@ -0,0 +1,22 @@
# Ollama CPU
services:
ollama-cpu:
image: ollama/ollama
ports:
- "11434:11434"
volumes:
- "${VOL_PATH:-/data}/ollama:/root/.ollama"
webui-cpu:
image: ${WEBUI_IMAGE:-ghcr.io/open-webui/open-webui:main}
volumes:
- "${VOL_PATH:-/data}/webui/cache:/root/.cache"
- "${VOL_PATH:-/data}/webui/data:/app/backend/data"
environment:
- WEBUI_URL=${WEBUI_URL:-http://localhost:8080}
- WEBUI_SECRET_KEY=${WEBUI_SECRET_KEY:?must_be_set}
- OLLAMA_BASE_URL=http://ollama-cpu:11434
ports:
- "${WEBUI_PORT:-8080}:8080"
ollama/compose-gpu.yaml
+24
View File
@@ -0,0 +1,24 @@
# Ollama GPU
services:
ollama:
image: ollama/ollama
volumes:
- ${VOL_PATH}/ollama:/root/.ollama
runtime: nvidia
environment:
- NVIDIA_VISIBLE_DEVICES=all
- NVIDIA_DRIVER_CAPABILITIES=compute,utility
ports:
- "11434:11434"
webui:
image: ${WEBUI_IMAGE:-ghcr.io/open-webui/open-webui:main}
volumes:
- "${VOL_PATH:-/data}/webui:/root/.cache"
- "${VOL_PATH:-/data}/open-webui:/app/backend/data"
environment:
- WEBUI_URL=${WEBUI_URL:-http://localhost:8080}
- WEBUI_SECRET_KEY=${WEBUI_SECRET_KEY:?must_be_set}
- OLLAMA_BASE_URL=http://ollama:11434
ports:
- "${WEBUI_PORT:-8080}:8080"
ollama/sample.env
+5
View File
@@ -0,0 +1,5 @@
VOL_PATH=/data
WEBUI_IMAGE=ghcr.io/open-webui/open-webui:main
WEBUI_PORT=8080
WEBUI_SECRET_KEY=a2dddd88c9da5f25786a0eb6e48e66305ee999d2adac8f8af821f82f11f56441 # ChangeMe openssl rand -hex 32
WEBUI_URL=ollama.lan
ollama_notes/README.md
-25
View File
@@ -1,25 +0,0 @@
# Ollama
Dockerhub: https://hub.docker.com/r/ollama/ollama
Github (instructions): https://github.com/open-webui/open-webui
Reverse Proxy Port webui: 8080 (The docker-compose sets the port to 3000)
## Preperation
You need to have nvidia-drivers, cuda and nvidia-container-toolkit
installed before proceeding.
docker run for refrence:
```
docker run -d --runtime=nvidia --gpus=all -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama
docker exec -it ollama ollama pull llama3
docker exec -it ollama ollama run llama3
```
# Notes
this worked:
```bash
docker run -d -p 3000:8080 --gpus all --add-host=host.docker.internal:host-gateway -v webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:cuda
```
ollama_notes/docker-compose.yml
-56
View File
@@ -1,56 +0,0 @@
services:
ollama:
image: ollama/ollama
runtime: nvidia
environment:
- NVIDIA_VISIBLE_DEVICES=all
ports:
- "11434:11434"
volumes:
- ./data/ollama:/root/.ollama
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]
webui:
image: ${WEBUI_IMAGE:-ghcr.io/open-webui/open-webui:cuda}
runtime: nvidia # Use NVIDIA runtime for GPU support
deploy:
resources:
reservations:
devices:
- capabilities: [gpu]
environment:
- NVIDIA_VISIBLE_DEVICES=all
- NVIDIA_DRIVER_CAPABILITIES=compute,utility
volumes:
- "${VOL_PATH:-./data}/webui:/root/.cache"
- "${VOL_PATH:-./data}/open-webui:/app/backend/data"
extra_hosts:
- "host.docker.internal:host-gateway"
ports:
- "3000:8080" # Open WebUI access
ollama-cpu:
image: ollama/ollama
ports:
- "11434:11434"
volumes:
- "${VOL_PATH:-./data}/ollama:/root/.ollama"
webui-cpu:
image: ${WEBUI_IMAGE:-ghcr.io/open-webui/open-webui:cuda}
volumes:
- "${VOL_PATH:-./data}/webui:/root/.cache"
- "${VOL_PATH:-./data}/open-webui:/app/backend/data"
extra_hosts:
- "host.docker.internal:host-gateway"
environment:
- WEBUI_SECRET_KEY=${WEBUI_SECRET_KEY}
- USE_CUDA_DOCKER=${USE_CUDA_DOCKER}
ports:
- "3000:8080" # Open WebUI access
ollama_notes/sample.env
-9
View File
@@ -1,9 +0,0 @@
VOL_PATH=./data
WEBUI_IMAGE=ghcr.io/open-webui/open-webui:cuda
# For CPU
# WEBUI_IMAGE=ghcr.io/open-webui/open-webui:main
WEBUI_SECRET_KEY=keyher
WEBUI_URL=myai.example.com
USE_CUDA_DOCKER=true
openarchiver_dev/sample.env
+1 -1
View File
@@ -1,4 +1,4 @@
VOL_PATH=./data
VOL_PATH=/data
# --- Application Settings ---
# Set to 'production' for production environments