sonda: adapt setup for deployment (#3151)

Referenced issue: https://github.com/status-im/infra-hq/issues/135

Signed-off-by: markoburcul <marko@status.im>

apps/sonda/.env.example

@@ -18,14 +18,16 @@ EXTRA_ARGS=
 RLN_RELAY_CONTRACT_ADDRESS=
 
 # -------------------- SONDA CONFIG ------------------
+METRICS_PORT=8004
+NODE_REST_ADDRESS="http://nwaku:8645"
 CLUSTER_ID=16
 SHARD=32
 # Comma separated list of store nodes to poll
-STORE_NODES="/dns4/store-01.do-ams3.shards.test.status.im/tcp/30303/p2p/16Uiu2HAmAUdrQ3uwzuE4Gy4D56hX6uLKEeerJAnhKEHZ3DxF1EfT,\
+STORE_NODES="/dns4/store-01.do-ams3.shards.test.status.im/tcp/30303/p2p/16Uiu2HAmAUdrQ3uwzuE4Gy4D56hX6uLKEeerJAnhKEHZ3DxF1EfT,
-/dns4/store-02.do-ams3.shards.test.status.im/tcp/30303/p2p/16Uiu2HAm9aDJPkhGxc2SFcEACTFdZ91Q5TJjp76qZEhq9iF59x7R,\
+/dns4/store-02.do-ams3.shards.test.status.im/tcp/30303/p2p/16Uiu2HAm9aDJPkhGxc2SFcEACTFdZ91Q5TJjp76qZEhq9iF59x7R,
-/dns4/store-01.gc-us-central1-a.shards.test.status.im/tcp/30303/p2p/16Uiu2HAmMELCo218hncCtTvC2Dwbej3rbyHQcR8erXNnKGei7WPZ,\
+/dns4/store-01.gc-us-central1-a.shards.test.status.im/tcp/30303/p2p/16Uiu2HAmMELCo218hncCtTvC2Dwbej3rbyHQcR8erXNnKGei7WPZ,
-/dns4/store-02.gc-us-central1-a.shards.test.status.im/tcp/30303/p2p/16Uiu2HAmJnVR7ZzFaYvciPVafUXuYGLHPzSUigqAmeNw9nJUVGeM,\
+/dns4/store-02.gc-us-central1-a.shards.test.status.im/tcp/30303/p2p/16Uiu2HAmJnVR7ZzFaYvciPVafUXuYGLHPzSUigqAmeNw9nJUVGeM,
-/dns4/store-01.ac-cn-hongkong-c.shards.test.status.im/tcp/30303/p2p/16Uiu2HAm2M7xs7cLPc3jamawkEqbr7cUJX11uvY7LxQ6WFUdUKUT,\
+/dns4/store-01.ac-cn-hongkong-c.shards.test.status.im/tcp/30303/p2p/16Uiu2HAm2M7xs7cLPc3jamawkEqbr7cUJX11uvY7LxQ6WFUdUKUT,
 /dns4/store-02.ac-cn-hongkong-c.shards.test.status.im/tcp/30303/p2p/16Uiu2HAm9CQhsuwPR54q27kNj9iaQVfyRzTGKrhFmr94oD8ujU6P"
 # Wait time in seconds between two consecutive queries
 QUERY_DELAY=60

apps/sonda/Dockerfile.sonda

@@ -1,3 +1,23 @@
 FROM python:3.9.18-alpine3.18
 
-RUN pip install requests argparse prometheus_client
+ENV METRICS_PORT=8004
+ENV NODE_REST_ADDRESS="http://nwaku:8645"
+ENV QUERY_DELAY=60
+ENV STORE_NODES=""
+ENV CLUSTER_ID=1
+ENV SHARD=1
+ENV HEALTH_THRESHOLD=5
+
+WORKDIR /opt
+
+COPY sonda.py /opt/sonda.py
+
+RUN pip install requests argparse prometheus_client
+
+CMD python -u /opt/sonda.py \
+    --metrics-port=$METRICS_PORT \
+    --node-rest-address="${NODE_REST_ADDRESS}" \
+    --delay-seconds=$QUERY_DELAY \
+    --pubsub-topic="/waku/2/rs/${CLUSTER_ID}/${SHARD}" \
+    --store-nodes="${STORE_NODES}" \
+    --health-threshold=$HEALTH_THRESHOLD

apps/sonda/docker-compose.yml

@@ -1,5 +1,4 @@
 
-version: "3.7"
 x-logging: &logging
   logging:
     driver: json-file
@@ -15,6 +14,8 @@ x-rln-environment: &rln_env
   RLN_RELAY_CRED_PASSWORD: ${RLN_RELAY_CRED_PASSWORD:-} # Optional: Add your RLN_RELAY_CRED_PASSWORD after the "-"
 
 x-sonda-env: &sonda_env
+  METRICS_PORT: ${METRICS_PORT:-8004}
+  NODE_REST_ADDRESS: ${NODE_REST_ADDRESS:-"http://nwaku:8645"}
   CLUSTER_ID: ${CLUSTER_ID:-1}
   SHARD: ${SHARD:-0}
   STORE_NODES: ${STORE_NODES:-}
@@ -24,7 +25,8 @@ x-sonda-env: &sonda_env
 # Services definitions
 services:
   nwaku:
-    image: ${NWAKU_IMAGE:-harbor.status.im/wakuorg/nwaku:v0.30.1}
+    image: ${NWAKU_IMAGE:-harbor.status.im/wakuorg/nwaku:deploy-status-prod}
+    container_name: nwaku
     restart: on-failure
     ports:
       - 30304:30304/tcp
@@ -54,29 +56,27 @@ services:
     entrypoint: sh
     command:
       - /opt/run_node.sh
+    networks:
+      - nwaku-sonda
 
   sonda:
     build:
       context: .
       dockerfile: Dockerfile.sonda
+    container_name: sonda
     ports:
-      - 127.0.0.1:8004:8004
+      - 127.0.0.1:${METRICS_PORT}:${METRICS_PORT}
     environment:
       <<:
         - *sonda_env
-    command: >
-      python -u /opt/sonda.py
-      --delay-seconds=${QUERY_DELAY}
-      --pubsub-topic=/waku/2/rs/${CLUSTER_ID}/${SHARD}
-      --store-nodes=${STORE_NODES}
-      --health-threshold=${HEALTH_THRESHOLD}
-    volumes:
-      - ./sonda.py:/opt/sonda.py:Z
     depends_on:
       - nwaku
+    networks:
+      - nwaku-sonda
   
   prometheus:
     image: docker.io/prom/prometheus:latest
+    container_name: prometheus
     volumes:
       - ./monitoring/prometheus-config.yml:/etc/prometheus/prometheus.yml:Z
     command:
@@ -86,9 +86,12 @@ services:
     restart: on-failure:5
     depends_on:
       - nwaku
+    networks:
+      - nwaku-sonda
 
   grafana:
     image: docker.io/grafana/grafana:latest
+    container_name: grafana
     env_file:
       - ./monitoring/configuration/grafana-plugins.env
     volumes:
@@ -104,4 +107,8 @@ services:
     restart: on-failure:5
     depends_on:
       - prometheus
+    networks:
+      - nwaku-sonda
 
+networks:
+  nwaku-sonda:

apps/sonda/sonda.py

@@ -26,10 +26,12 @@ node_health = Gauge('node_health', "Binary indicator of a node's health. 1 is he
 
 # Argparser configuration
 parser = argparse.ArgumentParser(description='')
-parser.add_argument('-p', '--pubsub-topic', type=str, help='pubsub topic', default='/waku/2/rs/1/0')
+parser.add_argument('-m', '--metrics-port',      type=int, default=8004,                help='Port to expose prometheus metrics.')
-parser.add_argument('-d', '--delay-seconds', type=int, help='delay in second between messages', default=60)
+parser.add_argument('-a', '--node-rest-address', type=str, default="http://nwaku:8645", help='Address of the waku node to send messages to.')
-parser.add_argument('-n', '--store-nodes', type=str, help='comma separated list of store nodes to query', required=True)
+parser.add_argument('-p', '--pubsub-topic',      type=str, default='/waku/2/rs/1/0',    help='PubSub topic.')
-parser.add_argument('-t', '--health-threshold', type=int, help='consecutive successful store requests to consider a store node healthy', default=5)
+parser.add_argument('-d', '--delay-seconds',     type=int, default=60,                  help='Delay in seconds between messages.')
+parser.add_argument('-n', '--store-nodes',       type=str, required=True,               help='Comma separated list of store nodes to query.')
+parser.add_argument('-t', '--health-threshold',  type=int, default=5,                   help='Consecutive successful store requests to consider a store node healthy.')
 args = parser.parse_args()
 
 
@@ -178,22 +180,21 @@ def main():
       store_nodes = [s.strip() for s in args.store_nodes.split(",")]
   log_with_utc(f'Store nodes to query: {store_nodes}')
 
-  # Start Prometheus HTTP server at port 8004
+  # Start Prometheus HTTP server at port set by the CLI(default 8004)
-  start_http_server(8004)
+  start_http_server(args.metrics_port)
 
-  node_rest_address = 'http://nwaku:8645'
   while True:
     timestamp = time.time_ns()
       
     # Send Sonda message
-    res = send_sonda_msg(node_rest_address, args.pubsub_topic, SONDA_CONTENT_TOPIC, timestamp)
+    res = send_sonda_msg(args.node_rest_address, args.pubsub_topic, SONDA_CONTENT_TOPIC, timestamp)
 
     log_with_utc(f'sleeping: {args.delay_seconds} seconds')
     time.sleep(args.delay_seconds)
 
     # Only send store query if message was successfully published
     if(res):
-      send_store_queries(node_rest_address, store_nodes, args.pubsub_topic, SONDA_CONTENT_TOPIC, timestamp)
+      send_store_queries(args.node_rest_address, store_nodes, args.pubsub_topic, SONDA_CONTENT_TOPIC, timestamp)
     
     # Update node health metrics
     for store_node in store_nodes: