deployment works

Author: raphaeltm

samples/ollama/app/Dockerfile

@@ -1,15 +1,20 @@
-FROM ubuntu:22.04
+FROM debian:bookworm-slim
 
 # install curl
 RUN apt-get update && apt-get install -y curl && apt-get clean
 
-# create ollama user and group
-RUN groupadd -r ollama && useradd -r -g ollama ollama
+# Define build-time arguments with default values
+ARG OLLAMA_HOST=0.0.0.0:8000
+ARG OLLAMA_MODELS=/usr/share/ollama/.ollama/models
+
+# Use the build-time arguments to set environment variables
+ENV OLLAMA_HOST=${OLLAMA_HOST}
+ENV OLLAMA_MODELS=${OLLAMA_MODELS}
 
 RUN curl -fsSL https://ollama.com/install.sh | sh
 
-COPY ./run /run
+WORKDIR /app
 
-USER ollama
+COPY ./run.sh ./run.sh
 
-CMD ["/run"]
+CMD ["./run.sh"]

samples/ollama/app/run

@@ -0,0 +1,9 @@
+#!/bin/bash
+
+nohup ollama serve > ollama.log 2>&1 &
+time curl --retry 5 --retry-connrefused --retry-delay 0 -sf http://localhost:8000
+
+ollama pull llama3.2:1b
+
+# tail the log file infinitely
+tail -f ollama.log

samples/ollama/app/run.sh

@@ -1,5 +1,3 @@
-# Sample compose file. #REMOVE_ME_AFTER_EDITING
-
 services:
   app:
     # uncomment to add your own domain
@@ -8,10 +6,15 @@ services:
       context: ./app
       dockerfile: Dockerfile
     ports:
-      - target: 3000
-        published: 3000
+      - target: 8000
+        published: 8000
         mode: ingress
+    deploy:
+      resources:
+        limits:
+          cpus: '2'
+          memory: 4G
     healthcheck:
       # wget or curl required for healthchecks on services with a published port
       # this gets parsed by Defang and provided to the load balancers as well
-      test: [ "CMD", "curl", "-s", "http://localhost:3000/" ]
+      test: [ "CMD", "curl", "-s", "http://localhost:8000/" ]