Update prom rw exporter

.flake8

@@ -16,6 +16,7 @@ exclude =
   target
   __pycache__
   exporter/opentelemetry-exporter-jaeger/src/opentelemetry/exporter/jaeger/gen/
+  exporter/opentelemetry-exporter-prometheus-remote-write/src/opentelemetry/exporter/prometheus_remote_write/gen/
   exporter/opentelemetry-exporter-jaeger/build/*
   docs/examples/opentelemetry-example-app/src/opentelemetry_example_app/grpc/gen/
   docs/examples/opentelemetry-example-app/build/*

CHANGELOG.md

@@ -39,7 +39,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   ([#1253](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/1253))
 - Add metric instrumentation in starlette
   ([#1327](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/1327))
-
+- Add metric exporter for Prometheus Remote Write
+  ([#1359](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/1359))
 
 ### Fixed
 

exporter/opentelemetry-exporter-prometheus-remote-write/README.rst

@@ -0,0 +1,31 @@
+OpenTelemetry Prometheus Remote Write Exporter
+==============================================
+
+|pypi|
+
+.. |pypi| image:: https://badge.fury.io/py/opentelemetry-exporter-prometheus-remote-write.svg
+   :target: https://pypi.org/project/opentelemetry-exporter-prometheus-remote-write/
+
+This package contains an exporter to send `OTLP`_ metrics from the
+`OpenTelemetry Python SDK`_ directly to a `Prometheus Remote Write integrated backend`_
+(such as Cortex or Thanos) without having to run an instance of the
+Prometheus server.
+
+
+Installation
+------------
+
+::
+
+    pip install opentelemetry-exporter-prometheus-remote-write
+
+
+.. _OpenTelemetry: https://github.com/open-telemetry/opentelemetry-python/
+.. _Prometheus Remote Write integrated backend: https://prometheus.io/docs/operating/integrations/
+
+
+References
+----------
+
+* `OpenTelemetry Project <https://opentelemetry.io/>`_
+* `Prometheus Remote Write Integration <https://prometheus.io/docs/operating/integrations/>`_

exporter/opentelemetry-exporter-prometheus-remote-write/examples/Dockerfile

@@ -0,0 +1,11 @@
+FROM python:3.8
+
+RUN apt-get update -y && apt-get install libsnappy-dev -y
+
+WORKDIR /code
+COPY . .
+
+RUN pip install -e .
+RUN pip install -r ./examples/requirements.txt
+
+CMD ["python", "./examples/sampleapp.py"]

exporter/opentelemetry-exporter-prometheus-remote-write/examples/README.md

@@ -0,0 +1,42 @@
+# Prometheus Remote Write Exporter Example
+This example uses [Docker Compose](https://docs.docker.com/compose/) to set up:
+
+1. A Python program that creates 5 instruments with 5 unique
+aggregators and a randomized load generator
+2. An instance of [Cortex](https://cortexmetrics.io/) to receive the metrics
+data
+3. An instance of [Grafana](https://grafana.com/) to visualizse the exported
+data
+
+## Requirements
+* Have Docker Compose [installed](https://docs.docker.com/compose/install/)
+
+*Users do not need to install Python as the app will be run in the Docker Container*
+
+## Instructions
+1. Run `docker-compose up -d` in the the `examples/` directory
+
+The `-d` flag causes all services to run in detached mode and frees up your
+terminal session. This also causes no logs to show up. Users can attach themselves to the service's logs manually using `docker logs ${CONTAINER_ID} --follow`
+
+2. Log into the Grafana instance at [http://localhost:3000](http://localhost:3000)
+   * login credentials are `username: admin` and `password: admin`
+   * There may be an additional screen on setting a new password. This can be skipped and is optional
+
+3. Navigate to the `Data Sources` page
+   * Look for a gear icon on the left sidebar and select `Data Sources`
+
+4. Add a new Prometheus Data Source
+   * Use `http://cortex:9009/api/prom` as the URL
+   * (OPTIONAl) set the scrape interval to `2s` to make updates appear quickly
+   * click `Save & Test`
+
+5. Go to `Metrics Explore` to query metrics
+   * Look for a compass icon on the left sidebar
+   * click `Metrics` for a dropdown list of all the available metrics
+   * (OPTIONAL) Adjust time range by clicking the `Last 6 hours` button on the upper right side of the graph
+   * (OPTIONAL) Set up auto-refresh by selecting an option under the dropdown next to the refresh button on the upper right side of the graph
+   * Click the refresh button and data should show up on the graph
+
+6. Shutdown the services when finished
+   * Run `docker-compose down` in the examples directory

exporter/opentelemetry-exporter-prometheus-remote-write/examples/cortex-config.yml

@@ -0,0 +1,100 @@
+# This Cortex Config is copied from the Cortex Project documentation
+# Source: https://github.com/cortexproject/cortex/blob/master/docs/configuration/single-process-config.yaml
+
+# Configuration for running Cortex in single-process mode.
+# This configuration should not be used in production.
+# It is only for getting started and development.
+
+# Disable the requirement that every request to Cortex has a
+# X-Scope-OrgID header. `fake` will be substituted in instead.
+auth_enabled: false
+
+server:
+  http_listen_port: 9009
+
+  # Configure the server to allow messages up to 100MB.
+  grpc_server_max_recv_msg_size: 104857600
+  grpc_server_max_send_msg_size: 104857600
+  grpc_server_max_concurrent_streams: 1000
+
+distributor:
+  shard_by_all_labels: true
+  pool:
+    health_check_ingesters: true
+
+ingester_client:
+  grpc_client_config:
+    # Configure the client to allow messages up to 100MB.
+    max_recv_msg_size: 104857600
+    max_send_msg_size: 104857600
+    use_gzip_compression: true
+
+ingester:
+  # We want our ingesters to flush chunks at the same time to optimise
+  # deduplication opportunities.
+  spread_flushes: true
+  chunk_age_jitter: 0
+
+  walconfig:
+    wal_enabled: true
+    recover_from_wal: true
+    wal_dir: /tmp/cortex/wal
+
+  lifecycler:
+    # The address to advertise for this ingester.  Will be autodiscovered by
+    # looking up address on eth0 or en0; can be specified if this fails.
+    # address: 127.0.0.1
+
+    # We want to start immediately and flush on shutdown.
+    join_after: 0
+    min_ready_duration: 0s
+    final_sleep: 0s
+    num_tokens: 512
+    tokens_file_path: /tmp/cortex/wal/tokens
+
+    # Use an in memory ring store, so we don't need to launch a Consul.
+    ring:
+      kvstore:
+        store: inmemory
+      replication_factor: 1
+
+# Use local storage - BoltDB for the index, and the filesystem
+# for the chunks.
+schema:
+  configs:
+    - from: 2019-07-29
+      store: boltdb
+      object_store: filesystem
+      schema: v10
+      index:
+        prefix: index_
+        period: 1w
+
+storage:
+  boltdb:
+    directory: /tmp/cortex/index
+
+  filesystem:
+    directory: /tmp/cortex/chunks
+
+  delete_store:
+    store: boltdb
+
+purger:
+  object_store_type: filesystem
+
+frontend_worker:
+  # Configure the frontend worker in the querier to match worker count
+  # to max_concurrent on the queriers.
+  match_max_concurrent: true
+
+# Configure the ruler to scan the /tmp/cortex/rules directory for prometheus
+# rules: https://prometheus.io/docs/prometheus/latest/configuration/recording_rules/#recording-rules
+ruler:
+  enable_api: true
+  enable_sharding: false
+  storage:
+    type: local
+    local:
+      directory: /tmp/cortex/rules
+

exporter/opentelemetry-exporter-prometheus-remote-write/examples/docker-compose.yml

@@ -0,0 +1,33 @@
+# Copyright The OpenTelemetry Authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+version: "3.8"
+
+services:
+  cortex:
+    image: quay.io/cortexproject/cortex:v1.5.0
+    command:
+      - -config.file=./config/cortex-config.yml
+    volumes:
+      - ./cortex-config.yml:/config/cortex-config.yml:ro
+    ports:
+      - 9009:9009
+  grafana:
+    image: grafana/grafana:latest
+    ports:
+      - 3000:3000
+  sample_app:
+    build:
+      context: ../
+      dockerfile: ./examples/Dockerfile

exporter/opentelemetry-exporter-prometheus-remote-write/examples/requirements.txt

@@ -0,0 +1,7 @@
+psutil
+protobuf>=3.13.0
+requests>=2.25.0
+python-snappy>=0.5.4
+opentelemetry-api
+opentelemetry-sdk
+opentelemetry-proto

exporter/opentelemetry-exporter-prometheus-remote-write/examples/sampleapp.py

@@ -0,0 +1,100 @@
+import logging
+import random
+import sys
+import time
+from logging import INFO
+
+import psutil
+
+from opentelemetry import metrics
+from opentelemetry.exporter.prometheus_remote_write import (
+    PrometheusRemoteWriteMetricsExporter,
+)
+from opentelemetry.metrics import Observation
+from opentelemetry.sdk.metrics import MeterProvider
+from opentelemetry.sdk.metrics.export import PeriodicExportingMetricReader
+
+logging.basicConfig(stream=sys.stdout, level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+testing_labels = {"environment": "testing"}
+
+exporter = PrometheusRemoteWriteMetricsExporter(
+    endpoint="http://cortex:9009/api/prom/push",
+    headers={"X-Scope-Org-ID": "5"},
+)
+reader = PeriodicExportingMetricReader(exporter, 1000)
+provider = MeterProvider(metric_readers=[reader])
+metrics.set_meter_provider(provider)
+meter = metrics.get_meter(__name__)
+
+
+# Callback to gather cpu usage
+def get_cpu_usage_callback(observer):
+    for (number, percent) in enumerate(psutil.cpu_percent(percpu=True)):
+        labels = {"cpu_number": str(number)}
+        yield Observation(percent, labels)
+
+
+# Callback to gather RAM usage
+def get_ram_usage_callback(observer):
+    ram_percent = psutil.virtual_memory().percent
+    yield Observation(ram_percent, {})
+
+
+requests_counter = meter.create_counter(
+    name="requests",
+    description="number of requests",
+    unit="1",
+)
+
+request_min_max = meter.create_counter(
+    name="requests_min_max",
+    description="min max sum count of requests",
+    unit="1",
+)
+
+request_last_value = meter.create_counter(
+    name="requests_last_value",
+    description="last value number of requests",
+    unit="1",
+)
+
+requests_active = meter.create_up_down_counter(
+    name="requests_active",
+    description="number of active requests",
+    unit="1",
+)
+
+meter.create_observable_counter(
+    callbacks=[get_ram_usage_callback],
+    name="ram_usage",
+    description="ram usage",
+    unit="1",
+)
+
+meter.create_observable_up_down_counter(
+    callbacks=[get_cpu_usage_callback],
+    name="cpu_percent",
+    description="per-cpu usage",
+    unit="1",
+)
+
+request_latency = meter.create_histogram("request_latency")
+
+# Load generator
+num = random.randint(0, 1000)
+while True:
+    # counters
+    requests_counter.add(num % 131 + 200, testing_labels)
+    request_min_max.add(num % 181 + 200, testing_labels)
+    request_last_value.add(num % 101 + 200, testing_labels)
+
+    # updown counter
+    requests_active.add(num % 7231 + 200, testing_labels)
+
+    request_latency.record(num % 92, testing_labels)
+    logger.log(level=INFO, msg="completed metrics collection cycle")
+    time.sleep(1)
+    num += 9791

exporter/opentelemetry-exporter-prometheus-remote-write/proto/README.md

@@ -0,0 +1,3 @@
+## Instructions
+1. Install protobuf tools. Can use your package manager or download from [GitHub](https://github.com/protocolbuffers/protobuf/releases/tag/v21.7)
+2. Run `generate-proto-py.sh` from inside the `proto/` directory