feat(sage): create the project sage-otel-collector (ARCH-324)

.gitignore

@@ -126,4 +126,4 @@ docs/_build/
 target/
 
 # Ipython Notebook
-.ipynb_checkpoints
+.ipynb_checkpoints

apps/sage/otel-collector/Dockerfile

@@ -0,0 +1,5 @@
+FROM public.ecr.aws/aws-observability/aws-otel-collector:v0.41.1 as aws-otel
+
+FROM otel/opentelemetry-collector-contrib:0.113.0
+
+COPY --from=aws-otel /healthcheck /healthcheck

apps/sage/otel-collector/README.md

@@ -0,0 +1,34 @@
+# Purpose
+
+As discussed in this Github Issue: <https://github.com/open-telemetry/opentelemetry-collector-contrib/issues/30798#issuecomment-2009233014>
+The official opentelemetry (OTEL) collector image does not contain cURL or related shell
+commands required to do container level health checks. It is reliant on external
+services such as the application load balancer in AWS to perform these checks. This is
+problematic with our deployment of the OTEL collector as we are using AWS
+service connect with AWS ECS to allow other containers within the namespace to connect
+to the collector. As such, there is no load balancer in-front of the container to handle
+its lifecycle. Within ECS, the recommended way from AWS to handle container level health
+checks is to let ECS perform commands in the container.
+Source: <https://docs.aws.amazon.com/AmazonECS/latest/APIReference/API_HealthCheck.html>
+
+Since the OTEL collector does not have a shell, nor cURL available we need to accomplish
+this another way. In the official AWS OTEL collector distro they accomplish this by
+compiling a golang script down into a binary that can be run within the container.
+Unfortunately we cannot use the AWS OTEL collector because they are not supporting the
+`oauth2clientauthextension`: <https://github.com/aws-observability/aws-otel-collector/issues/1492>.
+
+For our purposes we are creating a new image based off the `otel/opentelemetry-collector-contrib` image,
+but with the addition of the healthcheck binary from the AWS OTEL distro. This
+combination lets us use the oauth2 extension, and have container level health checks.
+
+## Creating a new image (To automate later on)
+
+As new base images are updated we will need to in-turn create a new otel collector
+image that we deploy to ECS.
+
+1. Update values in the `Dockerfile`
+2. Run `docker build -t ghcr.io/sage-bionetworks/sage-otel-collector:vX.X.X .` (Replace the version)
+3. Run `docker push ghcr.io/sage-bionetworks/sage-otel-collector:vX.X.X` (Replace the version)
+
+Once a new image is built and pushed, then you'll want to update the values in the CDK
+scripts to use the new image version.

apps/sage/otel-collector/project.json

@@ -0,0 +1,32 @@
+{
+  "name": "sage-otel-collector",
+  "$schema": "../../../node_modules/nx/schemas/project-schema.json",
+  "projectType": "application",
+  "targets": {
+    "serve-detach": {
+      "executor": "nx:run-commands",
+      "options": {
+        "command": "docker/sage/serve-detach.sh sage-otel-collector"
+      }
+    },
+    "publish-image": {
+      "executor": "@nx-tools/nx-container:build",
+      "options": {
+        "context": "apps/sage/otel-collector",
+        "metadata": {
+          "images": ["ghcr.io/sage-bionetworks/{projectName}"],
+          "tags": ["type=edge,branch=main", "type=sha"]
+        },
+        "push": true
+      },
+      "dependsOn": ["build-image"]
+    },
+    "scan-image": {
+      "executor": "nx:run-commands",
+      "options": {
+        "command": "trivy image ghcr.io/sage-bionetworks/{projectName}:local --quiet",
+        "color": true
+      }
+    }
+  }
+}

docker/sage/networks.yml

@@ -0,0 +1,4 @@
+networks:
+  sage:
+    name: sage
+    driver: bridge

docker/sage/serve-detach.sh

@@ -0,0 +1,12 @@
+#!/usr/bin/env bash
+
+args=(
+  # List of services in alphanumeric order
+  --file docker/sage/services/otel-collector.yml
+
+  --file docker/sage/networks.yml
+
+  up $1 --detach --remove-orphans
+)
+
+docker compose "${args[@]}"

docker/sage/services/otel-collector.yml

@@ -0,0 +1,18 @@
+services:
+  sage-otel-collector:
+    image: ghcr.io/sage-bionetworks/sage-otel-collector:${SAGE_VERSION:-local}
+    container_name: sage-otel-collector
+    restart: always
+    networks:
+      - sage
+    ports:
+      - '1888:1888' # pprof extension
+      - '8888:8888' # Prometheus metrics exposed by the collector
+      - '8889:8889' # Prometheus exporter metrics
+      - '13133:13133' # health_check extension
+      - '4317:4317' # OTLP gRPC receiver
+      - '55679:55679' # zpages extension
+    deploy:
+      resources:
+        limits:
+          memory: 200M

libs/sage-monorepo/nx-plugin/src/plugins/plugin.ts

@@ -34,7 +34,7 @@ function writeProjectConfigurationsToCache(
   writeJsonFile(cachePath, results);
 }
 
-const projectFilePattern = '{apps,libs}/{openchallenges,agora,sandbox}/**/project.json';
+const projectFilePattern = '{apps,libs}/{openchallenges,agora,sage,sandbox}/**/project.json';
 
 export const createNodesV2: CreateNodesV2<SageMonorepoPluginOptions> = [
   projectFilePattern,