feat: add otel operator and sample configurations

pulumi/aws/destroy.sh

@@ -91,7 +91,7 @@ if command -v aws > /dev/null; then
   validate_aws_credentials
 fi
 
-k8s_projects=(sirius grafana prometheus certmgr logagent logstore kic-helm-chart)
+k8s_projects=(sirius observability grafana prometheus certmgr logagent logstore kic-helm-chart)
 
 # Test to see if EKS has been destroy AND there are still Kubernetes resources
 # that are being managed by Pulumi. If so, we have to destroy the stack for

pulumi/aws/observability/Pulumi.yaml

@@ -0,0 +1,7 @@
+name: observability
+runtime:
+  name: python
+  options:
+    virtualenv: ../venv
+config: ../config
+description: Deploys OTEL

pulumi/aws/observability/__main__.py

@@ -0,0 +1,79 @@
+import os
+
+import pulumi
+import pulumi_kubernetes as k8s
+from pulumi_kubernetes.yaml import ConfigGroup
+
+from kic_util import pulumi_config
+
+# Removes the status field from the Nginx Ingress Helm Chart, so that i#t is
+# compatible with the Pulumi Chart implementation.
+def remove_status_field(obj):
+    if obj['kind'] == 'CustomResourceDefinition' and 'status' in obj:
+        del obj['status']
+
+def pulumi_eks_project_name():
+    script_dir = os.path.dirname(os.path.abspath(__file__))
+    eks_project_path = os.path.join(script_dir, '..', 'eks')
+    return pulumi_config.get_pulumi_project_name(eks_project_path)
+
+
+def pulumi_ingress_project_name():
+    script_dir = os.path.dirname(os.path.abspath(__file__))
+    ingress_project_path = os.path.join(script_dir, '..', 'kic-helm-chart')
+    return pulumi_config.get_pulumi_project_name(ingress_project_path)
+
+
+def otel_operator_location():
+    script_dir = os.path.dirname(os.path.abspath(__file__))
+    otel_operator_path = os.path.join(script_dir, 'otel-operator', '*.yaml')
+    return otel_operator_path
+
+def otel_deployment_location():
+    script_dir = os.path.dirname(os.path.abspath(__file__))
+    otel_deployment_path = os.path.join(script_dir, 'otel-objects', '*.yaml')
+    return otel_deployment_path
+
+def add_namespace(obj):
+    obj['metadata']['namespace'] = 'observability'
+
+
+stack_name = pulumi.get_stack()
+project_name = pulumi.get_project()
+eks_project_name = pulumi_eks_project_name()
+pulumi_user = pulumi_config.get_pulumi_user()
+
+eks_stack_ref_id = f"{pulumi_user}/{eks_project_name}/{stack_name}"
+eks_stack_ref = pulumi.StackReference(eks_stack_ref_id)
+kubeconfig = eks_stack_ref.get_output('kubeconfig').apply(lambda c: str(c))
+eks_stack_ref.get_output('cluster_name').apply(
+    lambda s: pulumi.log.info(f'Cluster name: {s}'))
+
+k8s_provider = k8s.Provider(resource_name=f'ingress-setup-sample', kubeconfig=kubeconfig)
+
+# Create the namespace
+ns = k8s.core.v1.Namespace(resource_name='observability',
+                           metadata={'name': 'observability'},
+                           opts=pulumi.ResourceOptions(provider=k8s_provider))
+
+# Config Manifests: OTEL operator
+otel_operator = otel_operator_location()
+
+otel_op = ConfigGroup(
+    'otel-op',
+    files=[otel_operator],
+    transformations=[remove_status_field], # Need to review w/ operator
+    opts=pulumi.ResourceOptions(depends_on=[ns])
+)
+
+# Config Manifests: OTEL components
+otel_deployment = otel_deployment_location()
+
+otel_dep = ConfigGroup(
+    'otel-dep',
+    files=[otel_deployment],
+    transformations=[add_namespace, remove_status_field], # Need to review w/ operator
+    opts=pulumi.ResourceOptions(depends_on=[ns,otel_op])
+)
+
+

pulumi/aws/observability/otel-objects/README.md

@@ -0,0 +1,44 @@
+## Sample Configurations
+This directory contains a number of sample configurations that can be used with the 
+[OTEL kubernetes operator](https://github.com/open-telemetry/opentelemetry-operator) that is installed as part of the
+MARA project. 
+
+Each configuration currently uses the `simplest` deployment, which uses an in-memory store for data being processed. 
+This is obviously not suited to a production deployment, but it is intended to illustrate the steps required to work 
+with the OTEL deployment.
+
+## Commonality
+
+### Listening Ports
+Each of the sample files is configured to listen on the 
+[OTLP protocol](https://opentelemetry.io/docs/reference/specification/protocol/otlp/). The listen ports configured are:
+* grpc on port 9978
+* http on port 9979
+
+### Logging 
+All of the examples log to the container's stdout. However, the basic configuration is configured to only show the 
+condensed version of the traces being received. In order to see the full traces, you need to set the logging level to
+`DEBUG`. The basic-debug object is configured to do this automatically.
+
+## Configurations
+### `otel-collector.yaml.basic`
+This is the default collector that only listens and logs summary spans to the container's stdout. 
+
+### `otel-collector.yaml.basic`
+This is a variant of the default collector that will output full spans to the container's stdout. 
+
+### `otel-collector.yaml.full`
+This is a more complex variant that contains multiple receivers, processors, and exporters. Please see the file for 
+details.
+
+### `otel-collector.yaml.lightstep`
+This configuration file deploys lightstep as an ingester. Please note you will need to have a 
+[lightstep](https://lightstep.com/) account to use this option, and you will need to add your lightstep access token 
+to the file in the field noted.
+
+## Usage
+By default, the `otel-collector.yaml.basic` configuration is copied into the live `otel-collector.yaml`. The logic for
+this project runs all files ending in `.yaml` as part of the configuration so you simply need to either rename your 
+chosen file to `otel-collector.yaml` or add ensuring only the files you want to use have the `.yaml` extension. 
+
+

pulumi/aws/observability/otel-objects/otel-collector.yaml

@@ -0,0 +1,28 @@
+apiVersion: opentelemetry.io/v1alpha1
+kind: OpenTelemetryCollector
+metadata:
+  name: simplest
+  namespace: observability
+spec:
+  config: |
+    receivers:
+      otlp:
+        protocols:
+          grpc:
+            endpoint: 0.0.0.0:9978
+          http:
+            endpoint: 0.0.0.0:9979
+
+    processors:
+      batch:
+
+    exporters:
+      logging:
+        logLevel:
+
+    service:
+      pipelines:
+        traces:
+          receivers: [otlp]
+          processors: [batch]
+          exporters: [logging]

pulumi/aws/observability/otel-objects/otel-collector.yaml.basic

@@ -0,0 +1,28 @@
+apiVersion: opentelemetry.io/v1alpha1
+kind: OpenTelemetryCollector
+metadata:
+  name: simplest
+  namespace: observability
+spec:
+  config: |
+    receivers:
+      otlp:
+        protocols:
+          grpc:
+            endpoint: 0.0.0.0:9978
+          http:
+            endpoint: 0.0.0.0:9979
+
+    processors:
+      batch:
+
+    exporters:
+      logging:
+        logLevel:
+
+    service:
+      pipelines:
+        traces:
+          receivers: [otlp]
+          processors: [batch]
+          exporters: [logging]

pulumi/aws/observability/otel-objects/otel-collector.yaml.basic-debug

@@ -0,0 +1,28 @@
+apiVersion: opentelemetry.io/v1alpha1
+kind: OpenTelemetryCollector
+metadata:
+  name: simplest
+  namespace: observability
+spec:
+  config: |
+    receivers:
+      otlp:
+        protocols:
+          grpc:
+            endpoint: 0.0.0.0:9978
+          http:
+            endpoint: 0.0.0.0:9979
+
+    processors:
+      batch:
+
+    exporters:
+      logging:
+        logLevel: debug
+
+    service:
+      pipelines:
+        traces:
+          receivers: [otlp]
+          processors: [batch]
+          exporters: [logging]

pulumi/aws/observability/otel-objects/otel-collector.yaml.full

@@ -0,0 +1,65 @@
+apiVersion: opentelemetry.io/v1alpha1
+kind: OpenTelemetryCollector
+metadata:
+  name: simplest
+  namespace: observability
+spec:
+  config: |
+    extensions:
+      health_check:
+      pprof:
+        endpoint: 0.0.0.0:1777
+      zpages:
+        endpoint: 0.0.0.0:55679
+
+    receivers:
+      otlp:
+        protocols:
+          grpc:
+            endpoint: 0.0.0.0:9978
+          http:
+            endpoint: 0.0.0.0:9979
+      opencensus:
+      jaeger:
+        protocols:
+          grpc:
+          thrift_binary:
+          thrift_compact:
+          thrift_http:
+      zipkin:
+
+      # Collect own metrics
+      prometheus:
+        config:
+          scrape_configs:
+            - job_name: 'otel-collector'
+              scrape_interval: 120s
+              static_configs:
+                - targets: [ '0.0.0.0:8080']
+              metrics_path: '/z/prometheus'
+
+    processors:
+      batch:
+
+    exporters:
+      prometheus:
+        endpoint: "0.0.0.0:8889"
+
+      logging:
+        logLevel: debug
+
+      jaeger:
+        endpoint: "0.0.0.0:14250"
+
+    service:
+      pipelines:
+        traces:
+          receivers: [otlp, opencensus, jaeger, zipkin]
+          processors: [batch]
+          exporters: [logging, jaeger]
+        metrics:
+          receivers: [otlp, opencensus, prometheus]
+          processors: [batch]
+          exporters: [logging]
+
+      extensions: [health_check, pprof, zpages]

pulumi/aws/observability/otel-objects/otel-collector.yaml.lightstep

@@ -0,0 +1,35 @@
+apiVersion: opentelemetry.io/v1alpha1
+kind: OpenTelemetryCollector
+metadata:
+  name: simplest
+  namespace: observability
+spec:
+  config: |
+    receivers:
+      otlp:
+        protocols:
+          grpc:
+            endpoint: 0.0.0.0:9978
+          http:
+            endpoint: 0.0.0.0:9979
+
+    exporters:
+      logging:
+      otlp:
+        endpoint: ingest.lightstep.com:443
+        headers:
+          "lightstep-access-token":"YOURTOKEN"
+
+    processors:
+      batch:
+
+    service:
+      pipelines:
+        traces:
+          receivers: [otlp]
+          processors: [batch]
+          exporters: [logging, otlp]
+        metrics:
+          receivers: [otlp]
+          processors: [batch]
+          exporters: [logging, otlp]

pulumi/aws/observability/otel-operator/__main__.py

@@ -0,0 +1,12 @@
+import pulumi
+import pulumi_kubernetes as kubernetes
+
+opentelemetry_operator_system_namespace = kubernetes.core.v1.Namespace("opentelemetry_operator_systemNamespace",
+    api_version="v1",
+    kind="Namespace",
+    metadata=kubernetes.meta.v1.ObjectMetaArgs(
+        labels={
+            "control-plane": "controller-manager",
+        },
+        name="opentelemetry-operator-system",
+    ))