feat: run qdrant and ollama inside thothii

docker/core.Dockerfile

@@ -91,10 +91,10 @@ ENV PATH="/opt/venv/bin:/usr/local/bin:$PATH" \
     HOME=/home/thoth
 
 COPY scripts/verify-line-endings.sh /usr/local/bin/verify-line-endings
-COPY docker/core-entrypoint.sh docker/session-migrate.sh docker/ensure-pi-trust.mjs /app/docker/
+COPY docker/core-entrypoint.sh docker/session-migrate.sh docker/ensure-pi-trust.mjs docker/embedding-model-init.sh /app/docker/
 COPY docker/smoke/core-smoke.sh /app/docker/smoke/core-smoke.sh
 RUN /usr/local/bin/verify-line-endings /app/docker \
- && chmod +x /app/docker/core-entrypoint.sh /app/docker/session-migrate.sh /app/docker/smoke/core-smoke.sh
+ && chmod +x /app/docker/core-entrypoint.sh /app/docker/session-migrate.sh /app/docker/embedding-model-init.sh /app/docker/smoke/core-smoke.sh
 
 WORKDIR /app/backend
 USER thoth

docker/embedding-model-init.sh

@@ -0,0 +1,81 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+ollama_base_url=${OLLAMA_BASE_URL:-http://embedding:11434}
+ollama_model=${OLLAMA_MODEL:-qwen3-embedding:0.6b}
+wait_timeout_sec=${OLLAMA_WAIT_TIMEOUT_SEC:-180}
+
+case "$wait_timeout_sec" in
+  ''|*[!0-9]*)
+    echo "OLLAMA_WAIT_TIMEOUT_SEC must be an integer number of seconds" >&2
+    exit 1
+    ;;
+esac
+
+case "$ollama_base_url" in
+  http://*)
+    host_and_path=${ollama_base_url#http://}
+    ;;
+  *)
+    echo "OLLAMA_BASE_URL must use http://" >&2
+    exit 1
+    ;;
+esac
+
+host_port=${host_and_path%%/*}
+ollama_host=${host_port%%:*}
+ollama_port=${host_port##*:}
+if [[ "$host_port" == "$ollama_host" ]]; then
+  ollama_port=80
+fi
+
+deadline=$((SECONDS + wait_timeout_sec))
+export OLLAMA_HOST="$ollama_base_url"
+
+fetch_tags() {
+  local response body
+  response=$(
+    exec 3<>"/dev/tcp/$ollama_host/$ollama_port"
+    printf 'GET /api/tags HTTP/1.1\r\nHost: %s\r\nConnection: close\r\n\r\n' "$ollama_host" >&3
+    cat <&3
+  ) || return 1
+  [[ "$response" == *$' 200 '* || "$response" == HTTP/1.1$' 200'* || "$response" == HTTP/1.0$' 200'* ]] || return 1
+  body=${response#*$'\r\n\r\n'}
+  if [[ "$body" == "$response" ]]; then
+    body=${response#*$'\n\n'}
+  fi
+  printf '%s' "$body"
+}
+
+model_present() {
+  local compact_json
+  compact_json=$(printf '%s' "$1" | tr -d '[:space:]')
+  grep -Fq "\"name\":\"$ollama_model\"" <<<"$compact_json"
+}
+
+wait_for_tags() {
+  local tags_json
+  while (( SECONDS <= deadline )); do
+    if tags_json=$(fetch_tags 2>/dev/null); then
+      printf '%s' "$tags_json"
+      return 0
+    fi
+    sleep 1
+  done
+  echo "timed out waiting for Ollama tags at $ollama_base_url/api/tags" >&2
+  return 1
+}
+
+tags_json=$(wait_for_tags)
+if model_present "$tags_json"; then
+  echo "embedding model already cached: $ollama_model"
+  exit 0
+fi
+
+ollama pull "$ollama_model"
+tags_json=$(fetch_tags)
+model_present "$tags_json" || {
+  echo "embedding model missing after pull: $ollama_model" >&2
+  exit 1
+}
+echo "embedding model ready: $ollama_model"