initial

.dockerignore

@@ -0,0 +1,3 @@
+.venv
+.git
+.gitignore

.github/workflows/docker-build-push.yml

@@ -0,0 +1,37 @@
+name: Build docker image
+on:
+  create:
+    tags:
+      - 'v*'
+  push:
+    branches:
+      - main
+
+jobs:
+  docker:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Login to GitHub Container Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ghcr.io
+          username: ${{ github.repository_owner }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Build and push
+        uses: docker/build-push-action@v6
+        with:
+          platforms: linux/amd64,linux/arm64
+          push: true
+          tags: |
+            ghcr.io/dysnix/block_lag_exporter:latest
+            ghcr.io/user/block_lag_exporter:${{ github.ref_name }}

.gitignore

@@ -0,0 +1,2 @@
+.venv
+.idea

Dockerfile

@@ -0,0 +1,17 @@
+# Use an official Python runtime as a parent image
+FROM python:3.12-slim-bookworm
+
+# Set the working directory in the container to /app
+WORKDIR /app
+
+# Add Pipfile and Pipfile.lock to the WORKDIR
+ADD Pipfile Pipfile.lock /app/
+
+RUN apt-get update && apt-get -y install gcc && \
+    pip install pipenv && pipenv install --system --deploy && \
+    apt-get -y remove gcc && apt-get -y autoremove
+
+# Add the rest of the code
+ADD ./exporter/ /app/
+
+CMD [ "python3", "main.py" ]

Pipfile

@@ -0,0 +1,17 @@
+[[source]]
+url = "https://pypi.org/simple"
+verify_ssl = true
+name = "pypi"
+
+[packages]
+prometheus-client = "*"
+web3 = "*"
+websockets = "*"
+aiohttp = "*"
+prometheus_async = "*"
+
+[dev-packages]
+
+[requires]
+python_version = "3.12"
+python_full_version = "3.12.4"

README.md

@@ -1,2 +1,15 @@
 # block_lag_exporter
-Metrics exporter to export latest block lag for evm-based chains
+Simple metrics exporter to export latest block lag for evm-based chains. It exports 2 metrics:
+* `head_lag_seconds` - histogram with values from uptime
+* `head_lag_seconds_last` - gauge with latest value of block lag
+
+and 2 endpoints on `http://0.0.0.0:${LISTENER_PORT}`:
+* `/metrics` - to scrape prometheus metrics
+* `/health` - to check liveness 
+
+Default values are optimized to use it as a k8s geth/geth-like sidecar 
+Use following environment variables to override defaults:
+* `LISTENER_PORT=8000` - port to listen 
+* `WS_URL=ws://localhost:8545` - websocket URL to connect and subscribe to new blocks
+* `HIST_BUCKETS=0.05,0.08,0.1,0.15,0.2,0.3,0.4,0.6,0.8,1.0,1.2,1.6,2.0,2.5,3.0,4.0,8.0,+Inf` - override prometheus 
+histogram buckets for histogram metric

exporter/main.py

@@ -0,0 +1,92 @@
+import asyncio
+import signal
+import json
+import sys
+import os
+import time
+
+from web3 import Web3
+from websockets import connect
+from aiohttp import web
+from prometheus_async import aio
+from prometheus_client import Histogram, Gauge
+
+
+def process_block(block):
+    timestamp = int(block['timestamp'], 16)
+    block_number = int(block['number'], 16)
+    ts = time.time()
+    lag = ts - timestamp
+    hist.observe(lag)
+    gauge.set("{:+.4f}".format(lag))
+    print("ts=%d block=%d lag=%2.4f" % (timestamp, block_number, lag))
+    return
+
+
+async def get_event():
+    try:
+        async with connect(ws_url) as ws:
+            await ws.send('{"jsonrpc": "2.0", "id": 1, "method": "eth_subscribe", "params": ["newHeads"]}')
+            subscription_response = await ws.recv()
+            print("ws subscription: %s" % subscription_response)
+            while True:
+                message = await asyncio.wait_for(ws.recv(), timeout=5)
+                response = json.loads(message)
+                block = response['params']['result']
+                process_block(block)
+    except Exception as e:
+        print(e)
+        return
+
+
+async def event_wrapper():
+    # await asyncio.sleep(0)
+    counter = 0
+    while True:
+        print("Starting wrapper tasks #%d" % counter)
+        await get_event()
+        print("await app done")
+        await asyncio.sleep(2)
+        counter += 1
+    # yield
+
+
+async def background_tasks(app):
+    app[ws_listener] = asyncio.create_task(event_wrapper())
+    yield
+    app[ws_listener].cancel()
+    await app[ws_listener]
+
+
+async def on_shutdown(app: web.Application) -> None:
+    print("Shutting down ...")
+    # there is no sense to wait, metrics server is stopped anyway
+    # await asyncio.sleep(30)
+    sys.exit(0)
+
+
+async def health(self):
+    return web.Response(text="OK")
+
+
+if __name__ == "__main__":
+    metrics_port = int(os.environ.get("LISTENER_PORT", 8000))
+    ws_url = os.environ.get("WS_URL", "ws://localhost:8545")
+    buckets = os.environ.get(
+        "HIST_BUCKETS", "0.05,0.08,0.1,0.15,0.2,0.3,0.4,0.6,0.8,1.0,1.2,1.6,2.0,2.5,3.0,4.0,8.0,+Inf")
+
+    hist = Histogram('head_lag_seconds', 'Last block lag',
+                     buckets=buckets.split(','))
+    gauge = Gauge('head_lag_seconds_last', 'Last block lag')
+
+    loop = asyncio.new_event_loop()
+    asyncio.set_event_loop(loop)
+
+    app = web.Application()
+    ws_listener = web.AppKey("ws_listener", asyncio.Task[None])
+    app.cleanup_ctx.append(background_tasks)
+    app.on_shutdown.append(on_shutdown)
+    app.router.add_get("/metrics", aio.web.server_stats)
+    app.router.add_get("/health", health)
+
+    web.run_app(app, port=metrics_port, loop=loop, handle_signals=True)