feat(health-agent): add CI/CD pipeline, Uptime Kuma setup, and runtime configuration

Deploy workflows: - Integrate health-agent build (test) and image promotion (prod) into monitoring stack workflows - Add storagebox download of health-agent runtime (.env.monitoring.health-agent-runtime → health-agent/.env) and setup (.env.monitoring.health-agent-setup → health-agent/.env.setup) env files - Add "Run Uptime Kuma Setup" step: runs setup_uptime_kuma.py inside the built image only when uk_tokens.yml is missing, writes tokens to HEALTH_AGENT_CONFIG_GENERATED_DIR (/mnt/storagebox/monitoring/uk_generated) - Add health-agent/** and health-agent/deploy/prod.env path triggers to test and prod workflows respectively - Add HARBOR_CI_TOKEN login and HARBOR_PULL_TOKEN login before stack deploy in both workflows - Source health-agent/.env before docker stack deploy to expose HEALTH_AGENT_CONFIG_GENERATED_DIR Dockerfile: - Copy config/ and scripts/ into image so setup_uptime_kuma.py can run inside the container setup_uptime_kuma.py: - Load .env and .env.setup automatically via python-dotenv (no manual export needed) - Write uk_tokens.yml to config/generated/ (aligned with container volume mount) Health checks: - PATRONI_HOSTS and VAULT_HOSTS are now configurable via env vars (comma-separated host:port); no code change needed when node count changes - REDIS_SENTINEL_HOSTS now correctly parses host:port format; default updated to redis-sentinel:26379 - Fix NameError in check_patroni_cluster() caused by leftover node variable after loop refactor - Remove verify_ssl=False from Vault check; vault.iklim.co has a valid certificate Ops: - Add ops/build-and-push-health-agent.sh for manual bypass of CI pipeline - Add health-agent/deploy/prod.env template for prod image promotion manifest Project structure: - Move .env.example and .env.setup.example to health-agent/env-example/ (root .gitignore excludes health-agent/.env*) - Add root .gitignore: excludes uk_tokens.yml, __pycache__, .venv, and env files - Remove health-agent/.gitignore (superseded by root .gitignore)
2026-06-26 18:45:17 +03:00 · 2026-06-26 18:45:17 +03:00 · 58d5c24f41
commit 58d5c24f41
parent 062d3ff90d
14 changed files with 241 additions and 26 deletions
--- a/.gitea/workflows/deploy-monitoring-prod.yml
+++ b/.gitea/workflows/deploy-monitoring-prod.yml
@ -6,6 +6,7 @@ on:
      - prod-env
    paths:
      - 'docker-stack-monitoring.yml'
+      - 'health-agent/deploy/prod.env'
      - 'swag/**'
      - '.gitea/workflows/deploy-monitoring-prod.yml'

@ -47,17 +48,73 @@ jobs:
        run: |
          source ./common-functions-base.sh
          export SPRING_PROFILES_ACTIVE=PROD
-          rm -f .env .env.secrets.swag
+          rm -f .env .env.secrets.swag health-agent/.env health-agent/.env.setup
          scp -P 23 ${{ vars.STORAGEBOX_USER }}@${{ vars.STORAGEBOX_USER }}.your-storagebox.de:prod/secrets/iklim.co/.env ./.env
          scp -P 23 ${{ vars.STORAGEBOX_USER }}@${{ vars.STORAGEBOX_USER }}.your-storagebox.de:prod/secrets/iklim.co/.env.secrets.swag ./.env.secrets.swag
+          scp -P 23 ${{ vars.STORAGEBOX_USER }}@${{ vars.STORAGEBOX_USER }}.your-storagebox.de:prod/secrets/iklim.co/.env.monitoring.health-agent-runtime ./health-agent/.env
+          scp -P 23 ${{ vars.STORAGEBOX_USER }}@${{ vars.STORAGEBOX_USER }}.your-storagebox.de:prod/secrets/iklim.co/.env.monitoring.health-agent-setup ./health-agent/.env.setup
          require_env_file ./.env "Main env file"
          require_env_file ./.env.secrets.swag "SWAG secrets"
+          require_env_file ./health-agent/.env "Health-agent runtime env"
+          require_env_file ./health-agent/.env.setup "Health-agent setup env"
+
+      - name: Promote Health Agent Image
+        run: |
+          source ./common-functions-base.sh
+          export SPRING_PROFILES_ACTIVE=PROD
+          source_env_file ./health-agent/deploy/prod.env
+          if [ -z "${SOURCE_IMAGE_DIGEST:-}" ] || [ -z "${PROD_IMAGE_TAG:-}" ]; then
+            log_message "INFO" "health-agent/deploy/prod.env is empty — skipping health-agent promotion"
+            exit 0
+          fi
+          case "$SOURCE_IMAGE_DIGEST" in
+            registry.tarla.io/iklimco/health-agent@sha256:*) ;;
+            *) log_message "ERROR" "SOURCE_IMAGE_DIGEST must be registry.tarla.io/iklimco/health-agent@sha256:<digest>"; exit 1 ;;
+          esac
+          case "$PROD_IMAGE_TAG" in
+            *-rc*) log_message "ERROR" "PROD_IMAGE_TAG must not contain -rc"; exit 1 ;;
+          esac
+          PROD_IMAGE="registry.tarla.io/iklimco/health-agent:${PROD_IMAGE_TAG}"
+          echo "${{ secrets.HARBOR_CI_TOKEN }}" | \
+            docker login registry.tarla.io -u robot-ci-push-iklimco --password-stdin
+          docker pull "${SOURCE_IMAGE_DIGEST}"
+          docker tag "${SOURCE_IMAGE_DIGEST}" "${PROD_IMAGE}"
+          docker push "${PROD_IMAGE}"
+          if grep -q "^IMAGE_HEALTH_AGENT=" .env; then
+            sed -i "s|^IMAGE_HEALTH_AGENT=.*$|IMAGE_HEALTH_AGENT=health-agent:${PROD_IMAGE_TAG}|" .env
+          else
+            echo "IMAGE_HEALTH_AGENT=health-agent:${PROD_IMAGE_TAG}" >> .env
+          fi
+          echo "HEALTH_AGENT_IMAGE=${PROD_IMAGE}" >> $GITHUB_ENV
+          log_message "SUCCESS" "Promoted: ${PROD_IMAGE}"
+
+      - name: Run Uptime Kuma Setup
+        run: |
+          source ./common-functions-base.sh
+          export SPRING_PROFILES_ACTIVE=PROD
+          source_env_file ./health-agent/.env
+          mkdir -p "${HEALTH_AGENT_CONFIG_GENERATED_DIR}"
+          if [ ! -f "${HEALTH_AGENT_CONFIG_GENERATED_DIR}/uk_tokens.yml" ]; then
+            docker run --rm \
+              -v "${HEALTH_AGENT_CONFIG_GENERATED_DIR}:/app/config/generated" \
+              --env-file "$(pwd)/health-agent/.env" \
+              --env-file "$(pwd)/health-agent/.env.setup" \
+              "${HEALTH_AGENT_IMAGE}" \
+              python scripts/setup_uptime_kuma.py
+            log_message "SUCCESS" "Uptime Kuma setup complete, tokens written to ${HEALTH_AGENT_CONFIG_GENERATED_DIR}"
+          else
+            log_message "INFO" "uk_tokens.yml already exists, skipping Uptime Kuma setup"
+          fi

      - name: Deploy Monitoring Stack
        run: |
          source ./common-functions-base.sh
          export SPRING_PROFILES_ACTIVE=PROD
          source_env_file ./.env
+          source_env_file ./health-agent/.env
+          export HEALTH_AGENT_ENV_FILE="$(pwd)/health-agent/.env"
+          echo "${{ secrets.HARBOR_PULL_TOKEN }}" | \
+            docker login registry.tarla.io -u robot-swarm-pull-iklimco --password-stdin

          # Remove leftover dozzle_users Docker secret from previous setup
          docker secret rm dozzle_users 2>/dev/null || true
--- a/.gitea/workflows/deploy-monitoring-test.yml
+++ b/.gitea/workflows/deploy-monitoring-test.yml
@ -6,6 +6,7 @@ on:
      - test
    paths:
      - 'docker-stack-monitoring.yml'
+      - 'health-agent/**'
      - 'swag/**'
      - '.gitea/workflows/deploy-monitoring-test.yml'

@ -43,18 +44,67 @@ jobs:
        run: |
          source ./common-functions-base.sh
          export SPRING_PROFILES_ACTIVE=TEST
-          rm -f .env .env.secrets.swag
+          rm -f .env .env.secrets.swag health-agent/.env health-agent/.env.setup
          scp -P 23 ${{ vars.STORAGEBOX_USER }}@${{ vars.STORAGEBOX_USER }}.your-storagebox.de:test/secrets/iklim.co/.env ./.env
          scp -P 23 ${{ vars.STORAGEBOX_USER }}@${{ vars.STORAGEBOX_USER }}.your-storagebox.de:test/secrets/iklim.co/.env.secrets.swag ./.env.secrets.swag
+          scp -P 23 ${{ vars.STORAGEBOX_USER }}@${{ vars.STORAGEBOX_USER }}.your-storagebox.de:test/secrets/iklim.co/.env.monitoring.health-agent-runtime ./health-agent/.env
+          scp -P 23 ${{ vars.STORAGEBOX_USER }}@${{ vars.STORAGEBOX_USER }}.your-storagebox.de:test/secrets/iklim.co/.env.monitoring.health-agent-setup ./health-agent/.env.setup
          require_env_file ./.env "Main env file"
          require_env_file ./.env.secrets.swag "SWAG secrets"
+          require_env_file ./health-agent/.env "Health-agent runtime env"
+          require_env_file ./health-agent/.env.setup "Health-agent setup env"
+
+      - name: Build and Push Health Agent
+        run: |
+          source ./common-functions-base.sh
+          export SPRING_PROFILES_ACTIVE=TEST
+          VERSION=$(sed -n 's/^version = "\(.*\)"/\1/p' health-agent/pyproject.toml)
+          IMAGE_TAG="health-agent:${VERSION}-rc"
+          IMAGE_FULL="registry.tarla.io/iklimco/${IMAGE_TAG}"
+          echo "${{ secrets.HARBOR_CI_TOKEN }}" | \
+            docker login registry.tarla.io -u robot-ci-push-iklimco --password-stdin
+          docker build -t "${IMAGE_FULL}" health-agent/
+          docker push "${IMAGE_FULL}"
+          docker pull -q "${IMAGE_FULL}"
+          DIGEST=$(docker image inspect "${IMAGE_FULL}" --format '{{index .RepoDigests 0}}')
+          if grep -q "^IMAGE_HEALTH_AGENT=" .env; then
+            sed -i "s|^IMAGE_HEALTH_AGENT=.*$|IMAGE_HEALTH_AGENT=${IMAGE_TAG}|" .env
+          else
+            echo "IMAGE_HEALTH_AGENT=${IMAGE_TAG}" >> .env
+          fi
+          echo "HEALTH_AGENT_IMAGE=${IMAGE_FULL}" >> $GITHUB_ENV
+          log_message "SUCCESS" "Pushed: ${IMAGE_FULL}"
+          log_message "INFO" "Promotion manifest — write to health-agent/deploy/prod.env on prod-env branch:"
+          echo "  SOURCE_IMAGE_DIGEST=${DIGEST}"
+          echo "  PROD_IMAGE_TAG=${VERSION}"
+
+      - name: Run Uptime Kuma Setup
+        run: |
+          source ./common-functions-base.sh
+          export SPRING_PROFILES_ACTIVE=TEST
+          source_env_file ./health-agent/.env
+          mkdir -p "${HEALTH_AGENT_CONFIG_GENERATED_DIR}"
+          if [ ! -f "${HEALTH_AGENT_CONFIG_GENERATED_DIR}/uk_tokens.yml" ]; then
+            docker run --rm \
+              -v "${HEALTH_AGENT_CONFIG_GENERATED_DIR}:/app/config/generated" \
+              --env-file "$(pwd)/health-agent/.env" \
+              --env-file "$(pwd)/health-agent/.env.setup" \
+              "${HEALTH_AGENT_IMAGE}" \
+              python scripts/setup_uptime_kuma.py
+            log_message "SUCCESS" "Uptime Kuma setup complete, tokens written to ${HEALTH_AGENT_CONFIG_GENERATED_DIR}"
+          else
+            log_message "INFO" "uk_tokens.yml already exists, skipping Uptime Kuma setup"
+          fi

      - name: Deploy Monitoring Stack
        run: |
          source ./common-functions-base.sh
          export SPRING_PROFILES_ACTIVE=TEST
          source_env_file ./.env
-
+          source_env_file ./health-agent/.env
+          export HEALTH_AGENT_ENV_FILE="$(pwd)/health-agent/.env"
+          echo "${{ secrets.HARBOR_PULL_TOKEN }}" | \
+            docker login registry.tarla.io -u robot-swarm-pull-iklimco --password-stdin
          docker stack deploy \
            --with-registry-auth \
            --resolve-image changed \
--- a/.gitignore
+++ b/.gitignore
@ -0,0 +1,6 @@
+health-agent/config/generated/uk_tokens.yml
+*__pycache__*
+health-agent/.venv/*
+health-agent/prod-env/
+health-agent/test-env/
+health-agent/.env*
--- a/common-functions-base.sh
+++ b/common-functions-base.sh
@ -67,9 +67,8 @@ lookup_env_value() {
 }

 # Matematiksel veya mantıksal işlem gerektiren env değerlerini hesaplar.
-# Örn: Milisaniye cinsinden JWT süresini saniyeye çevirir.
 refresh_calculated_env_vars() {
-  export JWT_ACCESS_TOKEN_EXPIRATION_SEC=$(( JWT_ACCESS_TOKEN_EXPIRATION / 1000 ))
+  
 }

 # Tüm çevre dosyalarını (ana env, ortak sırlar ve servis sırları) tazeleyerek yükler.
--- a/health-agent/.gitignore
+++ b/health-agent/.gitignore
@ -1,3 +0,0 @@
-config/generated/uk_tokens.yml
-.env
-.env.setup
--- a/health-agent/Dockerfile
+++ b/health-agent/Dockerfile
@ -6,6 +6,8 @@ RUN apt-get update && apt-get install -y --no-install-recommends curl && rm -rf

 COPY pyproject.toml ./
 COPY src/ ./src/
+COPY config/ ./config/
+COPY scripts/ ./scripts/
 RUN pip install --no-cache-dir .
 ENV PYTHONPATH=/app/src

--- a/health-agent/README.md
+++ b/health-agent/README.md
@ -52,7 +52,7 @@ Environment_Monitoring/health-agent/
 ├── Dockerfile
 ├── pyproject.toml
 ├── .env.example                  # health-agent runtime değişkenleri (credentials, ENV, CLUSTER_SIZE_*)
-└── .env.setup.example            # setup script değişkenleri (UK_API_KEY, Slack webhook'ları)
+└── .env.setup.example            # setup script değişkenleri (UK_URL, UK_USER, UK_PASS, Slack webhook'ları)
 ```

 ---
@ -76,6 +76,8 @@ Environment_Monitoring/health-agent/
 | `EXTERNAL_DOMAIN` | Base domain — `iklim.co` in both environments |
 | `EXTERNAL_SUBDOMAIN_SUFFIX` | Subdomain suffix — empty for prod, `-test` for test → `api-test.iklim.co` |
 | `SLACK_WEBHOOK_IKLIM_{ENV}_OPS` | Direct Slack webhook for container crash/OOM events — e.g. `SLACK_WEBHOOK_IKLIM_PROD_OPS` |
+| `PATRONI_HOSTS` | Patroni node list (comma-separated `host:port`) — e.g. `patroni-01:8008,patroni-02:8008` |
+| `VAULT_HOSTS` | Vault node subdomain list (comma-separated) — e.g. `vault-1,vault-2,vault-3` |
 | `RABBITMQ_USER` / `RABBITMQ_PASS` | RabbitMQ management credentials |
 | `MONGO_URI` | MongoDB connection URI |
 | `REDIS_PASSWORD` | Redis / Sentinel password |
@ -104,13 +106,19 @@ Push token'ları `config/generated/uk_tokens.yml`'den otomatik okunur — bu dos

 Health-agent deploy edilmeden önce kurulum script'i çalıştırılır. Script, `monitors.yml`'i okuyarak tüm monitor, tag, group ve status page'leri Uptime Kuma'da oluşturur; push token'larını `config/generated/uk_tokens.yml`'e yazar.

-Script [`uptime-kuma-api`](https://pypi.org/project/uptime-kuma-api/) kütüphanesini kullanır. API key authentication desteği implementasyon öncesi doğrulanmalıdır; desteklenmiyorsa `requests` + `Authorization: Bearer <api_key>` ile REST API doğrudan çağrılır.
+Script `uptime-kuma-api-v2` kütüphanesini kullanır; Socket.IO üzerinden username/password ile bağlanır.

 ```bash
 cd Environment_Monitoring/health-agent

-# setup değişkenlerini doldur
-cp .env.setup.example .env.setup
+# Python 3.12 venv oluştur ve aktive et
+python3.12 -m venv .venv
+source .venv/bin/activate
+pip install -e ".[dev]"
+
+# runtime ve setup değişkenlerini doldur
+cp .env.example .env        # ENV, EXTERNAL_DOMAIN vb.
+cp .env.setup.example .env.setup  # UK_URL, UK_USER, UK_PASS, Slack webhook'ları

 # önce dry-run ile ne yapılacağını gör
 python scripts/setup_uptime_kuma.py --dry-run
@ -118,10 +126,8 @@ python scripts/setup_uptime_kuma.py --dry-run
 # tüm kaynakları oluştur
 python scripts/setup_uptime_kuma.py

-# sadece belirli bileşenleri güncelle
-python scripts/setup_uptime_kuma.py --only monitors
-python scripts/setup_uptime_kuma.py --only notifications
-python scripts/setup_uptime_kuma.py --only status-page
+# sadece belirli bir monitörü işle (monitor adıyla)
+python scripts/setup_uptime_kuma.py --only SWARM-CLUSTER
 ```

 Script idempotent çalışır — CI/CD pipeline'ında her deploy'da güvenle tetiklenebilir.
@ -141,6 +147,10 @@ Planlı bakım/deploy sırasında etkilenecek group için Uptime Kuma'da Mainten
 ```bash
 cd Environment_Monitoring/health-agent

+# Python 3.12 venv oluştur ve aktive et
+python3.12 -m venv .venv
+source .venv/bin/activate
+
 # bağımlılıkları kur
 pip install -e ".[dev]"

--- a/health-agent/deploy/prod.env
+++ b/health-agent/deploy/prod.env
@ -0,0 +1,2 @@
+SOURCE_IMAGE_DIGEST=registry.tarla.io/iklimco/health-agent@sha256:fadf229d4423075d2871f9dc4a5a0afdf6dfe7c5fcd04d866b2d6d6fe8942b56
+PROD_IMAGE_TAG=0.1.0
--- a/health-agent/env-example/.env.example
+++ b/health-agent/env-example/.env.example
@ -9,6 +9,8 @@ EXTERNAL_DOMAIN=iklim.co
 EXTERNAL_SUBDOMAIN_SUFFIX=
 UK_PUSH_URL_BASE=https://uptime.tarla.io/api/push
 SLACK_WEBHOOK_IKLIM_PROD_OPS=
+PATRONI_HOSTS=patroni-01:8008,patroni-02:8008,patroni-03:8008
+VAULT_HOSTS=vault
 RABBITMQ_USER=
 RABBITMQ_PASS=
 MONGO_URI=
@ -17,3 +19,6 @@ REDIS_MASTER_NAME=
 REDIS_SENTINEL_HOSTS=
 STORAGEBOX_PATH=
 APISIX_ADMIN_KEY=
+PATRONI_HOSTS=
+VAULT_HOSTS=
+HEALTH_AGENT_CONFIG_GENERATED_DIR=
--- a/health-agent/env-example/.env.setup.example
+++ b/health-agent/env-example/.env.setup.example
--- a/health-agent/scripts/setup_uptime_kuma.py
+++ b/health-agent/scripts/setup_uptime_kuma.py
@ -2,11 +2,16 @@ import os
 import argparse
 import yaml
 import logging
+from dotenv import load_dotenv
 from uptime_kuma_api import UptimeKumaApi, MonitorType

 logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")
 logger = logging.getLogger("uk-setup")

+_root = os.path.join(os.path.dirname(__file__), "..")
+load_dotenv(os.path.join(_root, ".env"))
+load_dotenv(os.path.join(_root, ".env.setup"))
+
 def format_str(text, env_name, project):
    if not isinstance(text, str):
        return text
@ -118,7 +123,7 @@ def setup_uptime_kuma(dry_run=False, only=None):
                logger.warning(f"Status page ops failed: {e}")
                
    # 4. Write tokens to uk_tokens.yml
-    token_file = os.path.join(os.path.dirname(__file__), "..", "config", "uk_tokens.yml")
+    token_file = os.path.join(os.path.dirname(__file__), "..", "config", "generated", "uk_tokens.yml")
    if not dry_run:
        with open(token_file, "w") as f:
            yaml.dump(tokens, f)
--- a/health-agent/src/health_agent/checks/http.py
+++ b/health-agent/src/health_agent/checks/http.py
@ -29,19 +29,27 @@ def http_check(url, expected_status=None, auth=None, verify_ssl=True, timeout=5,
        return False, None, ping_ms, str(e)

 def check_patroni_cluster():
-    nodes = ["patroni-01", "patroni-02", "patroni-03"]
+    hosts_env = os.getenv("PATRONI_HOSTS", "patroni-01:8008,patroni-02:8008,patroni-03:8008")
+    nodes = []
+    for h in hosts_env.split(","):
+        h = h.strip()
+        if ":" in h:
+            host, port = h.rsplit(":", 1)
+            nodes.append((host, int(port)))
+        else:
+            nodes.append((h, 8008))
    cluster_data = None
    error_msg = "All Patroni nodes unreachable"
    start_t = time.time()
-    
-    for node in nodes:
-        url = f"http://{node}:8008/cluster"
+
+    for host, port in nodes:
+        url = f"http://{host}:{port}/cluster"
        ok, resp, _, err = http_check(url, timeout=3)
        if ok and resp:
            cluster_data = resp.json()
            break
        elif err:
-            error_msg = f"{node} error: {err}"
+            error_msg = f"{host}:{port} error: {err}"
            
    ping_ms = int((time.time() - start_t) * 1000)
    
@ -117,7 +125,8 @@ def check_apisix():
        push("APISIX-GATEWAY", "down", f"admin API unreachable: {err or resp.status_code}", ping_ms)

 def check_vault():
-    nodes = ["vault-1", "vault-2", "vault-3"]
+    hosts_env = os.getenv("VAULT_HOSTS", "vault")
+    nodes = [h.strip() for h in hosts_env.split(",")]
    domain = os.getenv("EXTERNAL_DOMAIN", "iklim.co")
    unsealed_count = 0
    total = len(nodes)
@ -127,7 +136,7 @@ def check_vault():
    start_t = time.time()
    for node in nodes:
        url = f"https://{node}.{domain}:8200/v1/sys/health"
-        ok, resp, ms, err = http_check(url, verify_ssl=False, expected_status=[200, 429, 473])
+        ok, resp, ms, err = http_check(url, expected_status=[200, 429, 473])
        max_ping = max(max_ping, ms)
        
        if resp:
--- a/health-agent/src/health_agent/checks/redis_sentinel.py
+++ b/health-agent/src/health_agent/checks/redis_sentinel.py
@ -9,8 +9,15 @@ logger = logging.getLogger(__name__)
 def check_redis_sentinel():
    start_t = time.time()
    
-    hosts = os.getenv("REDIS_SENTINEL_HOSTS", "redis-sentinel-01,redis-sentinel-02,redis-sentinel-03")
-    sentinel_nodes = [(h.strip(), 26379) for h in hosts.split(",")]
+    hosts = os.getenv("REDIS_SENTINEL_HOSTS", "redis-sentinel:26379")
+    sentinel_nodes = []
+    for h in hosts.split(","):
+        h = h.strip()
+        if ":" in h:
+            host, port = h.rsplit(":", 1)
+            sentinel_nodes.append((host, int(port)))
+        else:
+            sentinel_nodes.append((h, 26379))
    
    master_name = os.getenv("REDIS_MASTER_NAME", "prod-master")
    password = os.getenv("REDIS_PASSWORD", None)
--- a/ops/build-and-push-health-agent.sh
+++ b/ops/build-and-push-health-agent.sh
@ -0,0 +1,66 @@
+#!/usr/bin/env bash
+# Builds the health-agent Docker image and pushes it to Harbor as a release candidate.
+# Use this to bypass the CI pipeline for manual builds/hotfixes.
+#
+# Usage (run from Environment_Monitoring/ root):
+#   HARBOR_CI_TOKEN=<token> ./ops/build-and-push-health-agent.sh
+#
+# Optional env vars:
+#   HARBOR_REGISTRY   (default: registry.tarla.io)
+#   HARBOR_PROJECT    (default: iklimco)
+#   HARBOR_CI_USER    (default: robot-ci-push-iklimco)
+#   TAG_SUFFIX        (default: -rc)
+
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+cd "$(cd "$SCRIPT_DIR/.." && pwd)"
+
+HARBOR_REGISTRY="${HARBOR_REGISTRY:-registry.tarla.io}"
+HARBOR_PROJECT="${HARBOR_PROJECT:-iklimco}"
+HARBOR_CI_USER="${HARBOR_CI_USER:-robot-ci-push-iklimco}"
+TAG_SUFFIX="${TAG_SUFFIX:--rc}"
+
+log() { echo "[$(date +%H:%M:%S)] $*"; }
+die() { echo "[$(date +%H:%M:%S)] ERROR: $*" >&2; exit 1; }
+
+sep() {
+  local title="$*"
+  local line
+  line=$(printf '─%.0s' {1..70})
+  echo
+  echo "$line"
+  printf '  %s\n' "$title"
+  echo "$line"
+}
+
+[[ -z "${HARBOR_CI_TOKEN:-}" ]] && die "HARBOR_CI_TOKEN env var is required."
+[[ -f "health-agent/pyproject.toml" ]] || die "health-agent/pyproject.toml not found — run from Environment_Monitoring/ root."
+[[ -f "health-agent/Dockerfile" ]] || die "health-agent/Dockerfile not found."
+
+VERSION=$(sed -n 's/^version = "\(.*\)"/\1/p' health-agent/pyproject.toml)
+[[ -n "$VERSION" ]] || die "Could not determine version from health-agent/pyproject.toml"
+
+IMAGE="${HARBOR_REGISTRY}/${HARBOR_PROJECT}/health-agent:${VERSION}${TAG_SUFFIX}"
+
+sep "health-agent  →  ${IMAGE}"
+
+log "Logging in to ${HARBOR_REGISTRY} as ${HARBOR_CI_USER}"
+echo "$HARBOR_CI_TOKEN" | docker login "$HARBOR_REGISTRY" -u "$HARBOR_CI_USER" --password-stdin
+log "✔ Harbor login successful"
+
+log "Building..."
+docker build -t "$IMAGE" health-agent/
+
+log "Pushing..."
+docker push "$IMAGE"
+docker pull -q "$IMAGE"
+DIGEST=$(docker image inspect "$IMAGE" --format '{{index .RepoDigests 0}}')
+
+log "✔ Pushed: ${IMAGE}"
+
+sep "Promotion Manifest — write to health-agent/deploy/prod.env on prod-env branch"
+echo
+echo "  SOURCE_IMAGE_DIGEST=${DIGEST}"
+echo "  PROD_IMAGE_TAG=${VERSION}"
+echo