chore: add supervisor for lpt infra (#3176)

* Adding lpt-runner script and assemble into liteprotocoltester image - to ease infra deployment
* Add supervisor that can run lpt continously in infra environment, infra.env defines defaults for run, in case image tag of lpt docker image is deploy it will build a specific image for infra deployment.
* Added message latency metrics
* DELAY_MESSAGES to MESSAGE_INTERVAL renaming
* Adjust name of START_PUBLISHING_AFTER
* Extend lpt readme with how to use make to build dockerized image and notice about infra deployment
* As fixed in discussion, we will control infra testing by built in predefined test setup
* Prevent peer switch in case using fixed service peers

Makefile

@@ -373,16 +373,16 @@ DOCKER_LPT_NIMFLAGS ?= -d:chronicles_colors:none -d:insecure
 # build a docker image for the fleet
 docker-liteprotocoltester: DOCKER_LPT_TAG ?= latest
 docker-liteprotocoltester: DOCKER_LPT_NAME ?= wakuorg/liteprotocoltester:$(DOCKER_LPT_TAG)
+# --no-cache
 docker-liteprotocoltester:
 	docker build \
-	  --no-cache \
 		--build-arg="MAKE_TARGET=liteprotocoltester" \
 		--build-arg="NIMFLAGS=$(DOCKER_LPT_NIMFLAGS)" \
 		--build-arg="NIM_COMMIT=$(DOCKER_NIM_COMMIT)" \
 		--build-arg="LOG_LEVEL=TRACE" \
 		--label="commit=$(shell git rev-parse HEAD)" \
 		--label="version=$(GIT_VERSION)" \
-		--target $(TARGET) \
+		--target $(if $(filter deploy,$(DOCKER_LPT_TAG)),deployment_lpt,standalone_lpt) \
 		--tag $(DOCKER_LPT_NAME) \
 		--file apps/liteprotocoltester/Dockerfile.liteprotocoltester.compile \
 		.

apps/liteprotocoltester/.env

@@ -1,10 +1,10 @@
-START_PUBLISHING_AFTER=45
+START_PUBLISHING_AFTER_SECS=45
 # can add some seconds delay before SENDER starts publishing
 
 NUM_MESSAGES=0
 # 0 for infinite number of messages
 
-DELAY_MESSAGES=8000
+MESSAGE_INTERVAL_MILLIS=8000
 # ms delay between messages
 
 

apps/liteprotocoltester/Dockerfile.liteprotocoltester.compile

@@ -25,9 +25,8 @@ RUN make -j$(nproc) deps QUICK_AND_DIRTY_COMPILER=1 ${NIM_COMMIT}
 RUN make -j$(nproc) ${NIM_COMMIT} $MAKE_TARGET LOG_LEVEL=${LOG_LEVEL} NIMFLAGS="${NIMFLAGS}"
 
 
-# PRODUCTION IMAGE -------------------------------------------------------------
-
-FROM alpine:3.18 AS prod
+# REFERENCE IMAGE as BASE for specialized PRODUCTION IMAGES----------------------------------------
+FROM alpine:3.18 AS base_lpt
 
 ARG MAKE_TARGET=liteprotocoltester
 
@@ -43,16 +42,33 @@ EXPOSE 30303 60000 8545
 # Referenced in the binary
 RUN apk add --no-cache libgcc pcre-dev libpq-dev \
   wget \
-  iproute2
+  iproute2 \
+  python3
 
 # Fix for 'Error loading shared library libpcre.so.3: No such file or directory'
 RUN ln -s /usr/lib/libpcre.so /usr/lib/libpcre.so.3
 
 COPY --from=nim-build /app/build/liteprotocoltester /usr/bin/
+RUN chmod +x /usr/bin/liteprotocoltester
+
+# Standalone image to be used manually and in lpt-runner -------------------------------------------
+FROM base_lpt AS standalone_lpt
+
 COPY --from=nim-build /app/apps/liteprotocoltester/run_tester_node.sh /usr/bin/
-COPY --from=nim-build /app/apps/liteprotocoltester/run_tester_node_on_fleet.sh /usr/bin/
+RUN chmod +x /usr/bin/run_tester_node.sh
 
 ENTRYPOINT ["/usr/bin/run_tester_node.sh", "/usr/bin/liteprotocoltester"]
 
-# # By default just show help if called without arguments
-CMD ["--help"]
+# Image for infra deployment -------------------------------------------
+FROM base_lpt AS deployment_lpt
+
+# let supervisor python script flush logs immediately
+ENV PYTHONUNBUFFERED="1"
+
+COPY --from=nim-build /app/apps/liteprotocoltester/run_tester_node_at_infra.sh /usr/bin/
+COPY --from=nim-build /app/apps/liteprotocoltester/infra.env /usr/bin/
+COPY --from=nim-build /app/apps/liteprotocoltester/lpt_supervisor.py /usr/bin/
+RUN chmod +x /usr/bin/run_tester_node_at_infra.sh
+RUN chmod +x /usr/bin/lpt_supervisor.py
+
+ENTRYPOINT ["/usr/bin/lpt_supervisor.py"]

apps/liteprotocoltester/README.md

@@ -126,11 +126,11 @@ Run a SENDER role liteprotocoltester and a RECEIVER role one on different termin
 |   Variable     | Description | Default |
 | ---: | :--- | :--- |
 | NUM_MESSAGES   | Number of message to publish, 0 means infinite | 120 |
-| DELAY_MESSAGES | Frequency of messages in milliseconds | 1000 |
+| MESSAGE_INTERVAL_MILLIS | Frequency of messages in milliseconds | 1000 |
 | PUBSUB | Used pubsub_topic for testing | /waku/2/rs/66/0 |
 | CONTENT_TOPIC  | content_topic for testing | /tester/1/light-pubsub-example/proto |
 | CLUSTER_ID  | cluster_id of the network | 16 |
-| START_PUBLISHING_AFTER | Delay in seconds before starting to publish to let service node connected | 5 |
+| START_PUBLISHING_AFTER_SECS | Delay in seconds before starting to publish to let service node connected | 5 |
 | MIN_MESSAGE_SIZE | Minimum message size in bytes | 1KiB |
 | MAX_MESSAGE_SIZE | Maximum message size in bytes | 120KiB |
 
@@ -143,7 +143,7 @@ Run a SENDER role liteprotocoltester and a RECEIVER role one on different termin
 | --service-node| Address of the service node to use for lightpush and/or filter service | - |
 | --bootstrap-node| Address of the fleet's bootstrap node to use to determine service peer randomly choosen from the network. `--service-node` switch has precedence over this | - |
 | --num-messages | Number of message to publish | 120 |
-| --delay-messages | Frequency of messages in milliseconds | 1000 |
+| --message-interval | Frequency of messages in milliseconds | 1000 |
 | --min-message-size | Minimum message size in bytes | 1KiB |
 | --max-message-size | Maximum message size in bytes | 120KiB |
 | --start-publishing-after | Delay in seconds before starting to publish to let service node connected in seconds | 5 |
@@ -169,9 +169,37 @@ There are multiple benefits of using bootstrap nodes. By using them liteprotocol
 Also by using bootstrap node and peer exchange discovery, litprotocoltester will be able to simulate service peer switch in case of failures. There are built in tresholds count for service peer failures (3) after service peer will be switched during the test. Also there will be max 10 trials of switching peer before test declared failed and quit.
 These service peer failures are reported, thus extending network reliability measures.
 
-### Docker image notice
+### Building docker image
 
-#### Building for docker compose runs on simulator or standalone
+Easiest way to build the docker image is to use the provided Makefile target.
+
+```bash
+cd <your-repository>
+make docker-liteprotocoltester
+```
+This will build liteprotocoltester from the ground up and create a docker image with the binary copied to it under image name and tag `wakuorg/liteprotocoltester:latest`.
+
+#### Building public image
+
+If you want to push the image to a public registry, you can use the jenkins job to do so.
+The job is available at https://ci.status.im/job/waku/job/liteprotocoltester/job/build-liteprotocoltester-image
+
+#### Building and deployment for infra testing
+
+For specific and continuous testing purposes we have a deployment of `liteprotocoltester` test suite to our infra appliances.
+This has its own configuration, constraints and requirements. To ease this job, image shall be built and pushed with `deploy` tag.
+This can be done by the jenkins job mentioned above.
+
+or manually by:
+```bash
+cd <your-repository>
+make DOCKER_LPT_TAG=deploy docker-liteprotocoltester
+```
+
+The image created with this method will be different from under any other tag. It prepared to run a preconfigured test suite continuously.
+It will also miss prometheus metrics scraping endpoint and grafana, thus it is not recommended to use it for general testing.
+
+#### Manually building for docker compose runs on simulator or standalone
 Please note that currently to ease testing and development tester application docker image is based on ubuntu and uses the externally pre-built binary of 'liteprotocoltester'.
 This speeds up image creation. Another dokcer build file is provided for proper build of boundle image.
 
@@ -239,9 +267,9 @@ Cluster id and Pubsub-topic must be accurately set according to the network conf
 The example shows that either multiaddress or ENR form accepted.
 
 ```bash
-export START_PUBLISHING_AFTER=60
+export START_PUBLISHING_AFTER_SECS=60
 export NUM_MESSAGES=200
-export DELAY_MESSAGES=1000
+export MESSAGE_INTERVAL_MILLIS=1000
 export MIN_MESSAGE_SIZE=15Kb
 export MAX_MESSAGE_SIZE=145Kb
 export PUBSUB=/waku/2/rs/16/32
@@ -274,9 +302,9 @@ cd lpt-runner
 
 export NUM_PUBLISHER_NODES=3
 export NUM_RECEIVER_NODES=1
-export START_PUBLISHING_AFTER=120
+export START_PUBLISHING_AFTER_SECS=120
 export NUM_MESSAGES=300
-export DELAY_MESSAGES=7000
+export MESSAGE_INTERVAL_MILLIS=7000
 export MIN_MESSAGE_SIZE=15Kb
 export MAX_MESSAGE_SIZE=145Kb
 export PUBSUB=/waku/2/rs/1/4

apps/liteprotocoltester/docker-compose-on-simularor.yml

@@ -15,13 +15,13 @@ x-rln-environment: &rln_env
 
 x-test-running-conditions: &test_running_conditions
   NUM_MESSAGES: ${NUM_MESSAGES:-120}
-  DELAY_MESSAGES: "${DELAY_MESSAGES:-1000}"
+  MESSAGE_INTERVAL_MILLIS: "${MESSAGE_INTERVAL_MILLIS:-1000}"
   PUBSUB: ${PUBSUB:-/waku/2/rs/66/0}
   CONTENT_TOPIC: ${CONTENT_TOPIC:-/tester/2/light-pubsub-test/wakusim}
   CLUSTER_ID: ${CLUSTER_ID:-66}
   MIN_MESSAGE_SIZE: ${MIN_MESSAGE_SIZE:-1Kb}
   MAX_MESSAGE_SIZE: ${MAX_MESSAGE_SIZE:-150Kb}
-  START_PUBLISHING_AFTER: ${START_PUBLISHING_AFTER:-5}  # seconds
+  START_PUBLISHING_AFTER_SECS: ${START_PUBLISHING_AFTER_SECS:-5}  # seconds
 
 
 # Services definitions

apps/liteprotocoltester/docker-compose.yml

@@ -15,13 +15,13 @@ x-rln-environment: &rln_env
 
 x-test-running-conditions: &test_running_conditions
   NUM_MESSAGES: ${NUM_MESSAGES:-120}
-  DELAY_MESSAGES: "${DELAY_MESSAGES:-1000}"
+  MESSAGE_INTERVAL_MILLIS: "${MESSAGE_INTERVAL_MILLIS:-1000}"
   PUBSUB: ${PUBSUB:-/waku/2/rs/66/0}
   CONTENT_TOPIC: ${CONTENT_TOPIC:-/tester/2/light-pubsub-test/wakusim}
   CLUSTER_ID: ${CLUSTER_ID:-66}
   MIN_MESSAGE_SIZE: ${MIN_MESSAGE_SIZE:-1Kb}
   MAX_MESSAGE_SIZE: ${MAX_MESSAGE_SIZE:-150Kb}
-  START_PUBLISHING_AFTER: ${START_PUBLISHING_AFTER:-5}  # seconds
+  START_PUBLISHING_AFTER_SECS: ${START_PUBLISHING_AFTER_SECS:-5}  # seconds
   STANDALONE: ${STANDALONE:-1}
   RECEIVER_METRICS_PORT: 8003
   PUBLISHER_METRICS_PORT: 8003

apps/liteprotocoltester/filter_subscriber.nim

@@ -44,7 +44,10 @@ proc unsubscribe(
     notice "unsubscribe request successful"
 
 proc maintainSubscription(
-    wakuNode: WakuNode, filterPubsubTopic: PubsubTopic, filterContentTopic: ContentTopic
+    wakuNode: WakuNode,
+    filterPubsubTopic: PubsubTopic,
+    filterContentTopic: ContentTopic,
+    preventPeerSwitch: bool,
 ) {.async.} =
   const maxFailedSubscribes = 3
   const maxFailedServiceNodeSwitches = 10
@@ -83,7 +86,7 @@ proc maintainSubscription(
         if noFailedSubscribes < maxFailedSubscribes:
           await sleepAsync(2.seconds) # Wait a bit before retrying
           continue
-        else:
+        elif not preventPeerSwitch:
           let peerOpt = selectRandomServicePeer(
             wakuNode.peerManager, some(actualFilterPeer), WakuFilterSubscribeCodec
           )
@@ -164,4 +167,6 @@ proc setupAndSubscribe*(
   discard setTimer(Moment.fromNow(interval), printStats)
 
   # Start maintaining subscription
-  asyncSpawn maintainSubscription(wakuNode, conf.pubsubTopics[0], conf.contentTopics[0])
+  asyncSpawn maintainSubscription(
+    wakuNode, conf.pubsubTopics[0], conf.contentTopics[0], conf.fixedServicePeer
+  )

apps/liteprotocoltester/infra.env

@@ -0,0 +1,11 @@
+TEST_INTERVAL_MINUTES=180
+START_PUBLISHING_AFTER_SECS=120
+NUM_MESSAGES=300
+MESSAGE_INTERVAL_MILLIS=1000
+MIN_MESSAGE_SIZE=15Kb
+MAX_MESSAGE_SIZE=145Kb
+PUBSUB=/waku/2/rs/16/32
+CONTENT_TOPIC=/tester/2/light-pubsub-test/fleet
+CLUSTER_ID=16
+LIGHTPUSH_BOOTSTRAP=enr:-QEKuED9AJm2HGgrRpVaJY2nj68ao_QiPeUT43sK-aRM7sMJ6R4G11OSDOwnvVacgN1sTw-K7soC5dzHDFZgZkHU0u-XAYJpZIJ2NIJpcISnYxMvim11bHRpYWRkcnO4WgAqNiVib290LTAxLmRvLWFtczMuc3RhdHVzLnByb2Quc3RhdHVzLmltBnZfACw2JWJvb3QtMDEuZG8tYW1zMy5zdGF0dXMucHJvZC5zdGF0dXMuaW0GAbveA4Jyc40AEAUAAQAgAEAAgAEAiXNlY3AyNTZrMaEC3rRtFQSgc24uWewzXaxTY8hDAHB8sgnxr9k8Rjb5GeSDdGNwgnZfg3VkcIIjKIV3YWt1Mg0
+FILTER_BOOTSTRAP=enr:-QEcuED7ww5vo2rKc1pyBp7fubBUH-8STHEZHo7InjVjLblEVyDGkjdTI9VdqmYQOn95vuQH-Htku17WSTzEufx-Wg4mAYJpZIJ2NIJpcIQihw1Xim11bHRpYWRkcnO4bAAzNi5ib290LTAxLmdjLXVzLWNlbnRyYWwxLWEuc3RhdHVzLnByb2Quc3RhdHVzLmltBnZfADU2LmJvb3QtMDEuZ2MtdXMtY2VudHJhbDEtYS5zdGF0dXMucHJvZC5zdGF0dXMuaW0GAbveA4Jyc40AEAUAAQAgAEAAgAEAiXNlY3AyNTZrMaECxjqgDQ0WyRSOilYU32DA5k_XNlDis3m1VdXkK9xM6kODdGNwgnZfg3VkcIIjKIV3YWt1Mg0

apps/liteprotocoltester/lightpush_publisher.nim

@@ -113,7 +113,8 @@ proc publishMessages(
     lightpushContentTopic: ContentTopic,
     numMessages: uint32,
     messageSizeRange: SizeRange,
-    delayMessages: Duration,
+    messageInterval: Duration,
+    preventPeerSwitch: bool,
 ) {.async.} =
   var actualServicePeer = servicePeer
   let startedAt = getNowInNanosecondTime()
@@ -177,7 +178,7 @@ proc publishMessages(
       else:
         noFailedPush += 1
         lpt_service_peer_failure_count.inc(labelValues = ["publisher"])
-        if noFailedPush > maxFailedPush:
+        if not preventPeerSwitch and noFailedPush > maxFailedPush:
           info "Max push failure limit reached, Try switching peer."
           let peerOpt = selectRandomServicePeer(
             wakuNode.peerManager, some(actualServicePeer), WakuLightPushCodec
@@ -198,7 +199,7 @@ proc publishMessages(
             noFailedServiceNodeSwitches += 1
             break
 
-    await sleepAsync(delayMessages)
+    await sleepAsync(messageInterval)
 
 proc setupAndPublish*(
     wakuNode: WakuNode, conf: LiteProtocolTesterConf, servicePeer: RemotePeerInfo
@@ -252,5 +253,6 @@ proc setupAndPublish*(
     conf.contentTopics[0],
     conf.numMessages,
     (min: parsedMinMsgSize, max: parsedMaxMsgSize),
-    conf.delayMessages.milliseconds,
+    conf.messageInterval.milliseconds,
+    conf.fixedServicePeer,
   )

apps/liteprotocoltester/lpt_metrics.nim

@@ -22,6 +22,9 @@ declarePublicCounter lpt_receiver_duplicate_messages_count,
 declarePublicGauge lpt_receiver_distinct_duplicate_messages_count,
   "number of distinct duplicate messages per peer", ["peer"]
 
+declarePublicGauge lpt_receiver_latencies,
+  "Message delivery latency per peer (min-avg-max)", ["peer", "latency"]
+
 declarePublicCounter lpt_receiver_lost_subscription_count,
   "number of filter service peer failed PING requests - lost subscription"
 

apps/liteprotocoltester/lpt_supervisor.py

@@ -0,0 +1,54 @@
+#!/usr/bin/env python3
+
+import os
+import time
+from subprocess import Popen
+import sys
+
+def load_env(file_path):
+  predefined_test_env = {}
+  with open(file_path) as f:
+    for line in f:
+      if line.strip() and not line.startswith('#'):
+        key, value = line.strip().split('=', 1)
+        predefined_test_env[key] = value
+  return predefined_test_env
+
+def run_tester_node(predefined_test_env):
+  role = sys.argv[1]
+  # override incoming environment variables with the ones from the file to prefer predefined testing environment.
+  for key, value in predefined_test_env.items():
+      os.environ[key] = value
+
+  script_cmd = "/usr/bin/run_tester_node_at_infra.sh /usr/bin/liteprotocoltester {role}".format(role=role)
+  return os.system(script_cmd)
+
+if __name__ == "__main__":
+  if len(sys.argv) < 2 or sys.argv[1] not in ["RECEIVER", "SENDER"]:
+    print("Error: First argument must be either 'RECEIVER' or 'SENDER'")
+    sys.exit(1)
+
+  predefined_test_env_file = '/usr/bin/infra.env'
+  predefined_test_env = load_env(predefined_test_env_file)
+
+  test_interval_minutes = int(predefined_test_env.get('TEST_INTERVAL_MINUTES', 60))  # Default to 60 minutes if not set
+  print(f"supervisor: Start testing loop. Interval is {test_interval_minutes} minutes")
+  counter = 0
+
+  while True:
+    counter += 1
+    start_time = time.time()
+    print(f"supervisor: Run #{counter} started at {time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(start_time))}")
+    print(f"supervisor: with arguments: {predefined_test_env}")
+
+    exit_code = run_tester_node(predefined_test_env)
+
+    end_time = time.time()
+    run_time = end_time - start_time
+    sleep_time = max(5 * 60, (test_interval_minutes * 60) - run_time)
+
+    print(f"supervisor: Tester node finished at {time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(end_time))}")
+    print(f"supervisor: Runtime was {run_time:.2f} seconds")
+    print(f"supervisor: Next run scheduled in {sleep_time // 60:.2f} minutes")
+
+    time.sleep(sleep_time)

apps/liteprotocoltester/run_tester_node.sh

@@ -88,7 +88,7 @@ if [ -z "${SERIVCE_NODE_ADDR}" ]; then
 fi
 
 if $SERVICE_NODE_DIRECT; then
-  FULL_NODE=--service-node="${SERIVCE_NODE_ADDR}"
+  FULL_NODE=--service-node="${SERIVCE_NODE_ADDR} --fixed-service-peer"
 else
   FULL_NODE=--bootstrap-node="${SERIVCE_NODE_ADDR}"
 fi
@@ -107,8 +107,8 @@ if [ -n "${CLUSTER_ID}" ]; then
     CLUSTER_ID=--cluster-id="${CLUSTER_ID}"
 fi
 
-if [ -n "${START_PUBLISHING_AFTER}" ]; then
-    START_PUBLISHING_AFTER=--start-publishing-after="${START_PUBLISHING_AFTER}"
+if [ -n "${START_PUBLISHING_AFTER_SECS}" ]; then
+    START_PUBLISHING_AFTER_SECS=--start-publishing-after="${START_PUBLISHING_AFTER_SECS}"
 fi
 
 if [ -n "${MIN_MESSAGE_SIZE}" ]; then
@@ -124,8 +124,8 @@ if [ -n "${NUM_MESSAGES}" ]; then
     NUM_MESSAGES=--num-messages="${NUM_MESSAGES}"
 fi
 
-if [ -n "${DELAY_MESSAGES}" ]; then
-    DELAY_MESSAGES=--delay-messages="${DELAY_MESSAGES}"
+if [ -n "${MESSAGE_INTERVAL_MILLIS}" ]; then
+    MESSAGE_INTERVAL_MILLIS=--message-interval="${MESSAGE_INTERVAL_MILLIS}"
 fi
 
 echo "Running binary: ${BINARY_PATH}"
@@ -136,14 +136,15 @@ echo "My external IP: ${MY_EXT_IP}"
 exec "${BINARY_PATH}"\
       --log-level=INFO\
       --nat=extip:${MY_EXT_IP}\
+      --test-peers\
       ${FULL_NODE}\
-      ${DELAY_MESSAGES}\
+      ${MESSAGE_INTERVAL_MILLIS}\
       ${NUM_MESSAGES}\
       ${PUBSUB}\
       ${CONTENT_TOPIC}\
       ${CLUSTER_ID}\
       ${FUNCTION}\
-      ${START_PUBLISHING_AFTER}\
+      ${START_PUBLISHING_AFTER_SECS}\
       ${MIN_MESSAGE_SIZE}\
       ${MAX_MESSAGE_SIZE}
       # --config-file=config.toml\

apps/liteprotocoltester/run_tester_node_at_infra.sh

@@ -0,0 +1,105 @@
+#!/bin/sh
+
+#set -x
+#echo "$@"
+
+if test -f .env; then
+  echo "Using .env file"
+  . $(pwd)/.env
+fi
+
+
+echo "I am a lite-protocol-tester node"
+
+BINARY_PATH=$1
+
+if [ ! -x "${BINARY_PATH}" ]; then
+  echo "Invalid binary path '${BINARY_PATH}'. Failing"
+  exit 1
+fi
+
+if [ "${2}" = "--help" ]; then
+  echo "You might want to check nwaku/apps/liteprotocoltester/README.md"
+  exec "${BINARY_PATH}" --help
+  exit 0
+fi
+
+FUNCTION=$2
+if [ "${FUNCTION}" = "SENDER" ]; then
+  FUNCTION=--test-func=SENDER
+  SERIVCE_NODE_ADDR=${LIGHTPUSH_SERVICE_PEER:-${LIGHTPUSH_BOOTSTRAP:-}}
+  NODE_ARG=${LIGHTPUSH_SERVICE_PEER:+--service-node="${LIGHTPUSH_SERVICE_PEER}"}
+  NODE_ARG=${NODE_ARG:---bootstrap-node="${LIGHTPUSH_BOOTSTRAP}"}
+  METRICS_PORT=--metrics-port="${PUBLISHER_METRICS_PORT:-8003}"
+fi
+
+if [ "${FUNCTION}" = "RECEIVER" ]; then
+  FUNCTION=--test-func=RECEIVER
+  SERIVCE_NODE_ADDR=${FILTER_SERVICE_PEER:-${FILTER_BOOTSTRAP:-}}
+  NODE_ARG=${FILTER_SERVICE_PEER:+--service-node="${FILTER_SERVICE_PEER}"}
+  NODE_ARG=${NODE_ARG:---bootstrap-node="${FILTER_BOOTSTRAP}"}
+  METRICS_PORT=--metrics-port="${RECEIVER_METRICS_PORT:-8003}"
+fi
+
+if [ -z "${SERIVCE_NODE_ADDR}" ]; then
+  echo "Service/Bootsrap node peer_id or enr is not provided. Failing"
+  exit 1
+fi
+
+MY_EXT_IP=$(wget -qO- --no-check-certificate https://api4.ipify.org)
+
+if [ -n "${PUBSUB}" ]; then
+    PUBSUB=--pubsub-topic="${PUBSUB}"
+else
+    PUBSUB=--pubsub-topic="/waku/2/rs/66/0"
+fi
+
+if [ -n "${CONTENT_TOPIC}" ]; then
+    CONTENT_TOPIC=--content-topic="${CONTENT_TOPIC}"
+fi
+
+if [ -n "${CLUSTER_ID}" ]; then
+    CLUSTER_ID=--cluster-id="${CLUSTER_ID}"
+fi
+
+if [ -n "${START_PUBLISHING_AFTER_SECS}" ]; then
+    START_PUBLISHING_AFTER_SECS=--start-publishing-after="${START_PUBLISHING_AFTER_SECS}"
+fi
+
+if [ -n "${MIN_MESSAGE_SIZE}" ]; then
+    MIN_MESSAGE_SIZE=--min-test-msg-size="${MIN_MESSAGE_SIZE}"
+fi
+
+if [ -n "${MAX_MESSAGE_SIZE}" ]; then
+    MAX_MESSAGE_SIZE=--max-test-msg-size="${MAX_MESSAGE_SIZE}"
+fi
+
+
+if [ -n "${NUM_MESSAGES}" ]; then
+    NUM_MESSAGES=--num-messages="${NUM_MESSAGES}"
+fi
+
+if [ -n "${MESSAGE_INTERVAL_MILLIS}" ]; then
+    MESSAGE_INTERVAL_MILLIS=--message-interval="${MESSAGE_INTERVAL_MILLIS}"
+fi
+
+echo "Running binary: ${BINARY_PATH}"
+echo "Node function is: ${FUNCTION}"
+echo "Using service/bootstrap node as: ${NODE_ARG}"
+echo "My external IP: ${MY_EXT_IP}"
+
+exec "${BINARY_PATH}"\
+      --log-level=INFO\
+      --nat=extip:${MY_EXT_IP}\
+      --test-peers\
+      ${NODE_ARG}\
+      ${MESSAGE_INTERVAL_MILLIS}\
+      ${NUM_MESSAGES}\
+      ${PUBSUB}\
+      ${CONTENT_TOPIC}\
+      ${CLUSTER_ID}\
+      ${FUNCTION}\
+      ${START_PUBLISHING_AFTER_SECS}\
+      ${MIN_MESSAGE_SIZE}\
+      ${MAX_MESSAGE_SIZE}\
+      ${METRICS_PORT}

apps/liteprotocoltester/statistics.nim

@@ -189,8 +189,11 @@ proc dupMsgs(self: Statistics): string =
     )
   return dupMsgs
 
-proc echoStat*(self: Statistics) =
+proc echoStat*(self: Statistics, peerId: string) =
   let (minL, maxL, avgL) = self.calcLatency()
+  lpt_receiver_latencies.set(labelValues = [peerId, "min"], value = minL.nanos())
+  lpt_receiver_latencies.set(labelValues = [peerId, "avg"], value = avgL.nanos())
+  lpt_receiver_latencies.set(labelValues = [peerId, "max"], value = maxL.nanos())
 
   let printable = catch:
     """*------------------------------------------------------------------------------------------*
@@ -248,7 +251,7 @@ proc echoStats*(self: var PerPeerStatistics) =
       echo "Error while printing statistics"
     else:
       echo peerLine.get()
-      stats.echoStat()
+      stats.echoStat(peerId)
 
 proc jsonStats*(self: PerPeerStatistics): string =
   try:

apps/liteprotocoltester/tester_config.nim

@@ -89,10 +89,10 @@ type LiteProtocolTesterConf* = object
     name: "start-publishing-after"
   .}: uint32
 
-  delayMessages* {.
+  messageInterval* {.
     desc: "Delay between messages in milliseconds.",
     defaultValue: 1000,
-    name: "delay-messages"
+    name: "message-interval"
   .}: uint32
 
   pubsubTopics* {.
@@ -142,7 +142,7 @@ type LiteProtocolTesterConf* = object
 
   testPeers* {.
     desc: "Run dial test on gathered PeerExchange peers.",
-    defaultValue: true,
+    defaultValue: false,
     name: "test-peers"
   .}: bool
 
@@ -158,6 +158,13 @@ type LiteProtocolTesterConf* = object
     name: "rest-port"
   .}: uint16
 
+  fixedServicePeer* {.
+    desc:
+      "Prevent changing the service peer in case of failures, the full test will stict to the first service peer in use.",
+    defaultValue: false,
+    name: "fixed-service-peer"
+  .}: bool
+
   restAllowOrigin* {.
     desc:
       "Allow cross-origin requests from the specified origin." &

ci/Jenkinsfile.lpt

@@ -61,6 +61,7 @@ pipeline {
           "--build-arg=MAKE_TARGET='liteprotocoltester' " +
           "--build-arg=NIMFLAGS='${params.NIMFLAGS}' " +
           "--build-arg=LOG_LEVEL='${params.LOWEST_LOG_LEVEL_ALLOWED}' "  +
+          "--target ${params.IMAGE_TAG == 'deploy' ? 'deployment_lpt' : 'standalone_lpt'} " +
           "--file=apps/liteprotocoltester/Dockerfile.liteprotocoltester.compile " +
           " ."
         )