feat: allow setting the cluster DNS domain

We use cluster.local by default to speed up DNS resolution, but this doesn't work in clusters with a custom domain. The Prometheus remote write domain doesn't use the suffix because there isn't any way to modify it from values.yaml other than manually.
SumoLogic · Oct 27, 2023 · 0b3dec3 · 0b3dec3
1 parent 6e16e14
commit 0b3dec3
Show file tree

Hide file tree

Showing 8 changed files with 16 additions and 6 deletions.
diff --git a/.changelog/3364.added.txt b/.changelog/3364.added.txt
@@ -0,0 +1 @@
+feat: allow setting the cluster DNS domain
diff --git a/deploy/helm/sumologic/README.md b/deploy/helm/sumologic/README.md
@@ -142,6 +142,7 @@ The following table lists the configurable parameters of the Sumo Logic chart an
 | `sumologic.collectionMonitoring`                                                      | If you set it to false, it would set `EXCLUDE_NAMESPACE=<release-namespace>` and not add the Prometheus remotestorage metrics.                                                                                                                                                                             | `true`                                                                                                                                                                               |
 | `sumologic.collectorName`                                                             | The name of the Sumo Logic collector that will be created in the SetUp job. Defaults to `clusterName` if not specified.                                                                                                                                                                                    | `Nil`                                                                                                                                                                                |
 | `sumologic.clusterName`                                                               | An identifier for the Kubernetes cluster. Whitespaces in the cluster name will be replaced with dashes.                                                                                                                                                                                                    | `kubernetes`                                                                                                                                                                         |
+| `sumologic.clusterDNSDomain`                                                          | The DNS domain for the cluster. Change this if you're using a custom domain.                                                                                                                                                                                                                               | `cluster.local`                                                                                                                                                                      |
 | `sumologic.cluster`                                                                   | Configuration of Kubernetes for [Terraform client](https://www.terraform.io/docs/providers/kubernetes/index.html#argument-reference).                                                                                                                                                                      | See [values.yaml]                                                                                                                                                                    |
 | `sumologic.collector.sources`                                                         | Configuration of HTTP sources. [See docs/Terraform.md for more information](/docs/terraform.md). All double quotes should be escaped here regarding Terraform syntax.                                                                                                                                      | See [values.yaml]                                                                                                                                                                    |
 | `sumologic.collector.fields`                                                          | Configuration of Sumo Logic fields. [See Sumo Logic Terraform Plugin documentation for more information](https://registry.terraform.io/providers/SumoLogic/sumologic/latest/docs/resources/collector#fields). All double quotes should be escaped here regarding Terraform syntax.                         | See [values.yaml]                                                                                                                                                                    |

diff --git a/deploy/helm/sumologic/conf/logs/collector/otelcloudwatch/config.yaml b/deploy/helm/sumologic/conf/logs/collector/otelcloudwatch/config.yaml
@@ -1,6 +1,6 @@
 exporters:
   otlphttp:
-    endpoint: http://${LOGS_METADATA_SVC}.${NAMESPACE}.svc.cluster.local.:4318
+    endpoint: http://${LOGS_METADATA_SVC}.${NAMESPACE}.svc.{{ .Values.sumologic.clusterDNSDomain }}.:4318
     sending_queue:
       queue_size: 10
     # this improves load balancing at the cost of more network traffic

diff --git a/deploy/helm/sumologic/conf/logs/collector/otelcol/config.yaml b/deploy/helm/sumologic/conf/logs/collector/otelcol/config.yaml
@@ -1,6 +1,6 @@
 exporters:
   otlphttp:
-    endpoint: http://${LOGS_METADATA_SVC}.${NAMESPACE}.svc.cluster.local.:4318
+    endpoint: http://${LOGS_METADATA_SVC}.${NAMESPACE}.svc.{{ .Values.sumologic.clusterDNSDomain }}.:4318
     sending_queue:
       queue_size: 10
     # this improves load balancing at the cost of more network traffic

diff --git a/deploy/helm/sumologic/conf/metrics/collector/otelcol/config.yaml b/deploy/helm/sumologic/conf/metrics/collector/otelcol/config.yaml
@@ -2,7 +2,7 @@
 {{- $scrapeConfigsPresent := or $collectorConfig.kubelet.enabled $collectorConfig.cAdvisor.enabled $collectorConfig.annotatedPods.enabled -}}
 exporters:
   otlphttp:
-    endpoint: http://${METADATA_METRICS_SVC}.${NAMESPACE}.svc.cluster.local.:4318
+    endpoint: http://${METADATA_METRICS_SVC}.${NAMESPACE}.svc.{{ .Values.sumologic.clusterDNSDomain }}.:4318
     sending_queue:
       queue_size: 10000
       num_consumers: 10

diff --git a/deploy/helm/sumologic/templates/_helpers/_metrics.tpl b/deploy/helm/sumologic/templates/_helpers/_metrics.tpl
@@ -36,10 +36,10 @@ Generate list of remoteWrite endpoints for telegraf configuration
 {{- $endpoints := list -}}
 {{- $kps := get .Values "kube-prometheus-stack" -}}
 {{- range $remoteWrite := $kps.prometheus.prometheusSpec.remoteWrite }}
-{{- $endpoints = append $endpoints ($remoteWrite.url | trimPrefix "http://$(METADATA_METRICS_SVC).$(NAMESPACE).svc.cluster.local.:9888" | quote) -}}
+{{- $endpoints = append $endpoints ($remoteWrite.url | trimPrefix "http://$(METADATA_METRICS_SVC).$(NAMESPACE):9888" | quote) -}}
 {{- end }}
 {{- range $remoteWrite := $kps.prometheus.prometheusSpec.additionalRemoteWrite }}
-{{- $endpoints = append $endpoints ($remoteWrite.url | trimPrefix "http://$(METADATA_METRICS_SVC).$(NAMESPACE).svc.cluster.local.:9888" | quote) -}}
+{{- $endpoints = append $endpoints ($remoteWrite.url | trimPrefix "http://$(METADATA_METRICS_SVC).$(NAMESPACE):9888" | quote) -}}
 {{- end -}}
 {{- range $endpoint := .Values.metadata.metrics.config.additionalEndpoints }}
 {{- $endpoints = append $endpoints ($endpoint | quote) -}}

diff --git a/deploy/helm/sumologic/values.yaml b/deploy/helm/sumologic/values.yaml
@@ -42,6 +42,11 @@ sumologic:
   ## Cluster name: Note spaces are not allowed and will be replaced with dashes.
   clusterName: "kubernetes"
 
+  ## Cluster DNS Domain
+  ## We use the DNS domain in internal urls to speed up DNS resolution, see https://github.com/kubernetes/kubernetes/issues/56903
+  ## Change this if you have set a non-default DNS domain in your cluster
+  clusterDNSDomain: "cluster.local"
+
   ## Configuration of Kubernetes for Terraform client
   ## https://www.terraform.io/docs/providers/kubernetes/index.html#argument-reference
   ## All double quotes should be escaped here regarding Terraform syntax
@@ -1226,7 +1231,7 @@ kube-prometheus-stack:
               action: replace
       remoteWrite:
         ## infrastructure metrics
-        - url: http://$(METADATA_METRICS_SVC).$(NAMESPACE).svc.cluster.local.:9888/prometheus.metrics
+        - url: http://$(METADATA_METRICS_SVC).$(NAMESPACE):9888/prometheus.metrics
           remoteTimeout: 5s
 
     serviceMonitor:

diff --git a/tests/integration/yamls/cluster.yaml b/tests/integration/yamls/cluster.yaml
@@ -6,6 +6,7 @@ nodes:
       # these changes allow us to scrape control plane metrics
       # see kind docs on customizing kubeadm: https://kind.sigs.k8s.io/docs/user/configuration/#kubeadm-config-patches
       # and kubeadm docs on configuring control plane components: https://kubernetes.io/docs/setup/production-environment/tools/kubeadm/control-plane-flags/
+      # The DNS domain is changed to verify we work correctly in that environment
       - |
         kind: ClusterConfiguration
         scheduler:
@@ -19,3 +20,5 @@ nodes:
             dataDir: /tmp/etcd
             extraArgs:
               listen-metrics-urls: http://0.0.0.0:2381
+        networking:
+          dnsDomain: "my.cluster"