Log relevant kubernetes events with each deployed API (#1906)

Author: Miguel Varela Ramos

CONTRIBUTING.md

@@ -194,6 +194,7 @@ image_prometheus_config_reloader: <account_id>.dkr.ecr.<region>.amazonaws.com/co
 image_prometheus_operator: <account_id>.dkr.ecr.<region>.amazonaws.com/cortexlabs/prometheus-operator:master
 image_prometheus_statsd_exporter: <account_id>.dkr.ecr.<region>.amazonaws.com/cortexlabs/prometheus-statsd-exporter:master
 image_grafana: <account_id>.dkr.ecr.<region>.amazonaws.com/cortexlabs/grafana:master
+image_event_exporter: <account_id>.dkr.ecr.<region>.amazonaws.com/cortexlabs/event-exporter:master
 ```
 
 Create `dev/config/cluster-gcp.yaml`. Paste the following config, and update `project`, `zone`, and all registry URLs (replace `<project_id>` with your project ID, and update `gcr.io` if you are using a different host):
@@ -222,6 +223,7 @@ image_prometheus_config_reloader: gcr.io/<project_id>/cortexlabs/prometheus-conf
 image_prometheus_operator: gcr.io/<project_id>/cortexlabs/prometheus-operator:master
 image_prometheus_statsd_exporter: gcr.io/<project_id>/cortexlabs/prometheus-statsd-exporter:master
 image_grafana: gcr.io/<project_id>/cortexlabs/grafana:master
+image_event_exporter: gcr.io/<project_id>/cortexlabs/event-exporter:master
 ```
 
 ### Building

build/images.sh

@@ -59,6 +59,7 @@ non_dev_images_cluster=(
   "prometheus-operator"
   "prometheus-statsd-exporter"
   "grafana"
+  "event-exporter"
 )
 non_dev_images_aws=(
   # includes non_dev_images_cluster

charts/templates/event-exporter.yaml

@@ -0,0 +1,74 @@
+apiVersion: v1
+kind: ServiceAccount
+metadata:
+  namespace: {{ .Release.Namespace }}
+  name: event-exporter
+
+---
+
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRoleBinding
+metadata:
+  name: event-exporter
+roleRef:
+  apiGroup: rbac.authorization.k8s.io
+  kind: ClusterRole
+  name: view
+subjects:
+  - kind: ServiceAccount
+    namespace: {{ .Release.Namespace }}
+    name: event-exporter
+
+---
+
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: event-exporter-config
+  namespace: {{ .Release.Namespace }}
+data:
+  config.yaml: |
+    logLevel: error
+    logFormat: json
+    route:
+      routes:
+        - match:
+            - receiver: "stdout"
+              labels:
+                cortex.dev/api: true
+    receivers:
+      - name: "stdout"
+        file:
+          path: "/dev/stdout"
+
+---
+
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: event-exporter
+  namespace: {{ .Release.Namespace }}
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      app: event-exporter
+  template:
+    metadata:
+      labels:
+        app: event-exporter
+    spec:
+      serviceAccountName: event-exporter
+      containers:
+        - name: event-exporter
+          image: {{ .Values.cortex.image_event_exporter }}
+          imagePullPolicy: IfNotPresent
+          args:
+            - -conf=/data/config.yaml
+          volumeMounts:
+            - mountPath: /data
+              name: event-exporter-config
+      volumes:
+        - name: event-exporter-config
+          configMap:
+            name: event-exporter-config

charts/templates/fluentbit.yaml

@@ -52,6 +52,8 @@ data:
 
     @INCLUDE input-kubernetes.conf
     @INCLUDE filter-kubernetes.conf
+    @INCLUDE filter-k8s-events.conf
+    @INCLUDE filter-stackdriver-format.conf
     @INCLUDE output.conf
 
   input-kubernetes.conf: |
@@ -104,6 +106,47 @@ data:
         Match               k8s_container.*
         Remove_wildcard     k8s.
 
+  filter-k8s-events.conf: |
+    [FILTER]
+        Name                nest
+        Match               k8s_container.*.event-exporter-*
+        Operation           lift
+        Nested_under        involvedObject
+        Add_prefix          involvedObject.
+
+    [FILTER]
+        Name                modify
+        Match               k8s_container.*.event-exporter-*
+        Condition           Key_exists labels
+        Rename              labels k8s.labels
+
+    [FILTER]
+        Name                modify
+        Match               k8s_container.*.event-exporter-*
+        Condition           Key_exists involvedObject.labels
+        Hard_copy           involvedObject.labels labels
+
+    [FILTER]
+        Name                nest
+        Match               k8s_container.*.event-exporter-*
+        Operation           nest
+        Wildcard            involvedObject.*
+        Nest_under          involvedObject
+        Remove_prefix       involvedObject.
+
+  filter-stackdriver-format.conf: |
+    [FILTER]
+        Name                modify
+        Match               k8s_container.*
+        Condition           Key_exists log
+        Rename              log message
+
+    [FILTER]
+        Name                modify
+        Match               k8s_container.*
+        Condition           Key_exists levelname
+        Rename              levelname level
+
   output.conf: |
     {{- if eq .Values.global.provider "aws" }}
     [OUTPUT]
@@ -122,6 +165,8 @@ data:
         resource              k8s_container
         k8s_cluster_name      {{ .Values.cortex.cluster_name }}
         k8s_cluster_location  {{ .Values.cortex.zone }}
+        severity_key          level
+        labels_key            labels
     {{- end }}
 
   parsers.conf: |

charts/values.yaml

@@ -29,6 +29,7 @@ cortex:
   image_prometheus_operator: quay.io/cortexlabs/prometheus-operator:master
   image_prometheus_statsd_exporter: quay.io/cortexlabs/prometheus-statsd-exporter:master
   image_grafana: quay.io/cortexlabs/grafana:master
+  image_event_exporter: quay.io/cortexlabs/event-exporter:master
 
 networking:
   istio-discovery:

cli/cmd/lib_cluster_config_aws.go

@@ -428,6 +428,10 @@ func setConfigFieldsFromCached(userClusterConfig *clusterconfig.Config, cachedCl
 		return clusterconfig.ErrorConfigCannotBeChangedOnUpdate(clusterconfig.ImageGrafanaKey, cachedClusterConfig.ImageGrafana)
 	}
 
+	if s.Obj(cachedClusterConfig.ImageEventExporter) != s.Obj(userClusterConfig.ImageEventExporter) {
+		return clusterconfig.ErrorConfigCannotBeChangedOnUpdate(clusterconfig.ImageEventExporterKey, cachedClusterConfig.ImageEventExporter)
+	}
+
 	if userClusterConfig.Spot != nil && *userClusterConfig.Spot != *cachedClusterConfig.Spot {
 		return clusterconfig.ErrorConfigCannotBeChangedOnUpdate(clusterconfig.SpotKey, *cachedClusterConfig.Spot)
 	}

dev/versions.md

@@ -330,6 +330,13 @@ supported (<https://github.com/awslabs/amazon-eks-ami/issues/176>)
 1. Update the base image version in `images/grafana/Dockerfile`.
 1. Update `grafana.yaml` as necessary, if that's the case.
 
+## Event Exporter
+
+1. Find the latest release
+   on [GitHub](https://github.com/opsgenie/kubernetes-event-exporter).
+1. Update the base image version in `images/event-exporter/Dockerfile`.
+1. Update `event-exporter.yaml` as necessary, if that's the case.
+
 ## aws-iam-authenticator
 
 1. Find the latest release [here](https://docs.aws.amazon.com/eks/latest/userguide/install-aws-iam-authenticator.html)

docs/clusters/aws/install.md

@@ -109,4 +109,5 @@ image_prometheus_config_reloader: quay.io/cortexlabs/prometheus-config-reloader:
 image_prometheus_operator: quay.io/cortexlabs/prometheus-operator:master
 image_prometheus_statsd_exporter: quay.io/cortexlabs/prometheus-statsd-exporter:master
 image_grafana: quay.io/cortexlabs/grafana:master
+image_event_exporter: quay.io/cortexlabs/event-exporter:master
 ```

docs/clusters/gcp/install.md

@@ -83,4 +83,5 @@ image_prometheus_config_reloader: quay.io/cortexlabs/prometheus-config-reloader:
 image_prometheus_operator: quay.io/cortexlabs/prometheus-operator:master
 image_prometheus_statsd_exporter: quay.io/cortexlabs/prometheus-statsd-exporter:master
 image_grafana: quay.io/cortexlabs/grafana:master
+image_event_exporter: quay.io/cortexlabs/event-exporter:master
 ```

docs/clusters/gcp/logging.md

@@ -7,18 +7,18 @@ RealtimeAPI:
 ```text
 resource.type="k8s_container"
 resource.labels.cluster_name="<INSERT CLUSTER NAME>"
-jsonPayload.labels.apiKind="RealtimeAPI"
-jsonPayload.labels.apiName="<INSERT API NAME>"
+labels.apiKind="RealtimeAPI"
+labels.apiName="<INSERT API NAME>"
 ```
 
 TaskAPI:
 
 ```text
 resource.type="k8s_container"
 resource.labels.cluster_name="<INSERT CLUSTER NAME>"
-jsonPayload.labels.apiKind="TaskAPI"
-jsonPayload.labels.apiName="<INSERT API NAME>"
-jsonPayload.labels.jobID="<INSERT JOB ID>"
+labels.apiKind="TaskAPI"
+labels.apiName="<INSERT API NAME>"
+labels.jobID="<INSERT JOB ID>"
 ```
 
 Please make sure to navigate to the project containing your cluster and adjust the time range accordingly before running queries.