Add otel!

charts/operator-wandb/Chart.lock

@@ -19,6 +19,9 @@ dependencies:
   version: 0.1.0
 - name: redis
   repository: https://charts.bitnami.com/bitnami
-  version: 18.1.0
-digest: sha256:0e062062405e017968fb5ad0e5064936cb55e2b441ddb1c2048f34eaf6de11a8
-generated: "2023-09-27T12:33:43.680199603-04:00"
+  version: 18.1.5
+- name: otel
+  repository: file://charts/otel
+  version: 0.1.0
+digest: sha256:d6f7dbed1f8fcbbd34d18b0911891fb27eeef0021092b69ea35e7ca5dcede038
+generated: "2023-10-16T19:07:10.090393-04:00"

charts/operator-wandb/Chart.yaml

@@ -2,7 +2,7 @@ apiVersion: v2
 name: operator-wandb
 description: A Helm chart for deploying W&B to Kubernetes
 type: application
-version: 0.10.24
+version: 0.10.25
 appVersion: 1.0.0
 icon: https://wandb.ai/logo.svg
 
@@ -40,3 +40,7 @@ dependencies:
     version: "18.*.*"
     condition: redis.install
     repository: https://charts.bitnami.com/bitnami
+  - name: otel
+    version: "*.*.*"
+    repository: file://charts/otel
+    condition: otel.install

charts/operator-wandb/charts/app/templates/deployment.yaml

@@ -76,6 +76,9 @@ spec:
             - name: prometheus
               containerPort: 8181
               protocol: TCP
+            - name: gorilla-statsd
+              containerPort: 8125
+              protocol: TCP
           env:
             - name: HOST
               value: "{{ .Values.global.host }}"
@@ -153,6 +156,11 @@ spec:
 
             - name: GORILLA_SESSION_LENGTH
               value: "{{ .Values.global.auth.sessionLengthHours }}h"
+
+            - name: GORILLA_STATSD_PORT
+              value: "8125"
+            - name: GORILLA_STATSD_HOST
+              value: "0.0.0.0"
 
             - name: BUCKET
               value: "{{ include "app.bucket" . }}"

charts/operator-wandb/charts/app/templates/service.yaml

@@ -1,4 +1,3 @@
-{{- if .Values.enabled }}
 apiVersion: v1
 kind: Service
 metadata:
@@ -23,6 +22,8 @@ spec:
     - port: 8181
       protocol: TCP
       name: prometheus
+    - port: 8125
+      protocol: UDP
+      name: gorilla-statsd
   selector:
-    {{- include "app.labels" . | nindent 4 }}
-{{- end }}
+    {{- include "app.labels" . | nindent 4 }}

charts/operator-wandb/charts/console/values.yaml

@@ -38,8 +38,8 @@ resources:
   # specify resources, uncomment the following lines, adjust them as necessary,
   # and remove the curly braces after 'resources:'.
   requests:
-    cpu: 500m
-    memory: 1Gi
+    cpu: 200m
+    memory: 200Mi
   limits:
-    cpu: 4000m
-    memory: 8Gi
+    cpu: 1
+    memory: 500Mi

charts/operator-wandb/charts/otel/.helmignore

@@ -0,0 +1,23 @@
+# Patterns to ignore when building packages.
+# This supports shell glob matching, relative path matching, and
+# negation (prefixed with !). Only one pattern per line.
+.DS_Store
+# Common VCS dirs
+.git/
+.gitignore
+.bzr/
+.bzrignore
+.hg/
+.hgignore
+.svn/
+# Common backup files
+*.swp
+*.bak
+*.tmp
+*.orig
+*~
+# Various IDEs
+.project
+.idea/
+*.tmproj
+.vscode/

charts/operator-wandb/charts/otel/Chart.yaml

@@ -0,0 +1,15 @@
+apiVersion: v2
+name: otel
+type: application
+description: A Helm chart for Kubernetes
+
+version: 0.1.0
+appVersion: "0.33.0"
+
+home: https://wandb.ai
+icon: https://wandb.ai/logo.svg
+
+maintainers:
+  - name: wandb
+    email: [email protected]
+    url: https://wandb.com

charts/operator-wandb/charts/otel/README.md

@@ -0,0 +1,7 @@
+We had to create a seperate chart, because the offical one does not support
+
+1. We need to send an otlphttp to the console server. The name of this service
+   is dynamic. TEL helm chart does not support dynamic pipeline values
+2. We could do the above as a config map, and pass it into the agent... however,
+   otel helm does not support using custom config maps names because they need
+   to be based on the release name.

charts/operator-wandb/charts/otel/templates/_config.tpl

@@ -0,0 +1,93 @@
+{{- define "otel.config" -}}
+{{- $data := deepCopy .Values.config }}
+{{- $config := .Values.config }}
+{{- $config = mustMergeOverwrite (include "otel.hostMetricsReceiver" . | fromYaml) $config }}
+{{- $config = mustMergeOverwrite (include "otel.logsCollectionReceiver" . | fromYaml) $config }}
+{{- $config = mustMergeOverwrite (include "otel.kubeletMetricsReceiver" . | fromYaml) $config }}
+{{- $config = mustMergeOverwrite (include "otel.kubernetesEventReceiver" . | fromYaml) $config }}
+{{- $config = mustMergeOverwrite (include "otel.kubernetesClusterReceiver" . | fromYaml) $config }}
+{{- $config = mustMergeOverwrite (include "otel.sqlQueryReceiver" . | fromYaml) $config }}
+{{- $config = mustMergeOverwrite (include "otel.statsdAppReceiver" . | fromYaml) $config }}
+{{- $config = mustMergeOverwrite (include "otel.extensions" . | fromYaml) $config }}
+{{- $config = mustMergeOverwrite (include "otel.processors" . | fromYaml) $config }}
+{{- $config = mustMergeOverwrite (include "otel.service" . | fromYaml) $config }}
+{{- $config = mustMergeOverwrite (include "otel.exporter" . | fromYaml) $config }}
+{{- tpl (toYaml $config) . }}
+{{- end }}
+
+{{- define "otel.exporter" -}}
+exporters:
+  debug: {}
+  debug/detailed:
+    verbosity: detailed
+  prometheus:
+    endpoint: 0.0.0.0:9109
+{{- end }}
+
+{{- define "otel.extensions" -}}
+extensions:
+  health_check: {}
+  memory_ballast:
+    size_in_percentage: 40
+{{- end }}
+
+{{- define "otel.processors" -}}
+processors:
+  batch: {}
+  memory_limiter:
+    check_interval: 5s
+    limit_percentage: 80
+    spike_limit_percentage: 25
+  k8sattributes:
+    filter:
+      node_from_env_var: K8S_NODE_NAME
+    passthrough: false
+    pod_association:
+    - sources:
+      - from: resource_attribute
+        name: k8s.pod.ip
+    - sources:
+      - from: resource_attribute
+        name: k8s.pod.uid
+    - sources:
+      - from: connection
+    extract:
+      metadata:
+        - "k8s.namespace.name"
+        - "k8s.deployment.name"
+        - "k8s.statefulset.name"
+        - "k8s.daemonset.name"
+        - "k8s.cronjob.name"
+        - "k8s.job.name"
+        - "k8s.node.name"
+        - "k8s.pod.name"
+        - "k8s.pod.uid"
+        - "k8s.pod.start_time"
+      annotations:
+        - tag_name: $$1
+          key_regex: (.*)
+          from: pod
+      labels:
+        - tag_name: $$1
+          key_regex: (.*)
+          from: pod
+{{- end }}
+
+{{- define "otel.service" -}}
+service:
+  extensions:
+  - health_check
+  - memory_ballast
+  pipelines:
+    metrics:
+      exporters: [debug, prometheus]
+      processors: [memory_limiter, batch, k8sattributes]
+      receivers: [hostmetrics, k8s_cluster, kubeletstats, sqlquery]
+    logs:
+      exporters: [debug]
+      processors: [memory_limiter, batch]
+      receivers: [filelog]
+  telemetry:
+    metrics:
+      address: ${env:POD_IP}:8888
+{{- end }}

charts/operator-wandb/charts/otel/templates/_helpers.tpl

@@ -0,0 +1,102 @@
+{{/* vim: set filetype=mustache: */}}
+
+{{/*
+Expand the name of the chart.
+*/}}
+{{- define "otel.name" -}}
+{{- default .Chart.Name .Values.nameOverride | trunc 63 | trimSuffix "-" }}
+{{- end }}
+
+{{/*
+Create a default fully qualified app name.
+We truncate at 63 chars because some Kubernetes name fields are limited to this (by the DNS naming spec).
+If release name contains chart name it will be used as a full name.
+*/}}
+{{- define "otel.fullname" -}}
+{{- if .Values.fullnameOverride }}
+{{- .Values.fullnameOverride | trunc 63 | trimSuffix "-" }}
+{{- else }}
+{{- $name := default .Chart.Name .Values.nameOverride }}
+{{- if contains $name .Release.Name }}
+{{- .Release.Name | trunc 63 | trimSuffix "-" }}
+{{- else }}
+{{- printf "%s-%s" .Release.Name $name | trunc 63 | trimSuffix "-" }}
+{{- end }}
+{{- end }}
+{{- end }}
+
+{{/*
+Create chart name and version as used by the chart label.
+*/}}
+{{- define "otel.chart" -}}
+{{- printf "%s-%s" .Chart.Name .Chart.Version | replace "+" "_" | trunc 63 | trimSuffix "-" }}
+{{- end }}
+
+{{/*
+Common labels
+*/}}
+{{- define "otel.labels" -}}
+helm.sh/chart: {{ include "otel.chart" . }}
+{{ include "otel.selectorLabels" . }}
+{{- if .Chart.AppVersion }}
+app.kubernetes.io/version: {{ .Chart.AppVersion | quote }}
+{{- end }}
+wandb.com/app-name: {{ include "otel.chart" . }}
+app.kubernetes.io/managed-by: {{ .Release.Service }}
+{{- end }}
+
+{{/*
+Selector labels
+*/}}
+{{- define "otel.selectorLabels" -}}
+app.kubernetes.io/name: {{ include "otel.name" . }}
+app.kubernetes.io/instance: {{ .Release.Name }}
+{{- end }}
+
+{{/*
+Create the name of the service account to use
+*/}}
+{{- define "otel.serviceAccountName" -}}
+{{- if .Values.serviceAccount.create }}
+{{- default (include "otel.fullname" .) .Values.serviceAccount.name }}
+{{- else }}
+{{- default "default" .Values.serviceAccount.name }}
+{{- end }}
+{{- end }}
+
+{{/*
+Returns the extraEnv keys and values to inject into containers.
+
+Global values will override any chart-specific values.
+*/}}
+{{- define "otel.extraEnv" -}}
+{{- $allExtraEnv := merge (default (dict) .local.extraEnv) .global.extraEnv -}}
+{{- range $key, $value := $allExtraEnv }}
+- name: {{ $key }}
+  value: {{ $value | quote }}
+{{- end -}}
+{{- end -}}
+
+{{/*
+Returns a list of _common_ labels to be shared across all
+app deployments and other shared objects.
+*/}}
+{{- define "otel.commonLabels" -}}
+{{- $commonLabels := default (dict) .Values.common.labels -}}
+{{- if $commonLabels }}
+{{-   range $key, $value := $commonLabels }}
+{{ $key }}: {{ $value | quote }}
+{{-   end }}
+{{- end -}}
+{{- end -}}
+
+{{/*
+Returns a list of _pod_ labels to be shared across all
+app deployments.
+*/}}
+{{- define "otel.podLabels" -}}
+{{- range $key, $value := .Values.pod.labels }}
+{{ $key }}: {{ $value | quote }}
+{{- end }}
+{{- end -}}
+