geodesic

what

• Add portal helmfile

why

To easily deploy:

• Kubernetes Dashboard
• Prometheus UI for Prometheus settings and metrics
• Grafana UI for reports
• Alert Manager UI for alerts

test

kubectl get pods -n monitoring

NAME                                                  READY     STATUS    RESTARTS   AGE
alertmanager-kube-prometheus-0                        2/2       Running   0          3d
datadog-datadog-5tz8b                                 1/1       Running   0          9h
datadog-datadog-c27gj                                 1/1       Running   0          9h
datadog-datadog-cqs6x                                 1/1       Running   0          9h
datadog-datadog-n9cmz                                 1/1       Running   0          9h
datadog-datadog-qr4dz                                 1/1       Running   0          9h
datadog-datadog-skmzg                                 1/1       Running   0          9h
datadog-datadog-xvsml                                 1/1       Running   0          9h
datadog-kube-state-metrics-6799fc6bb-wdvnl            1/1       Running   0          9h
kube-prometheus-exporter-kube-state-9c4667764-t5fw7   2/2       Running   0          3d
kube-prometheus-exporter-node-5mnl8                   1/1       Running   0          3d
kube-prometheus-exporter-node-6j92c                   1/1       Running   0          3d
kube-prometheus-exporter-node-8mhst                   1/1       Running   0          3d
kube-prometheus-exporter-node-cwqv4                   1/1       Running   0          3d
kube-prometheus-exporter-node-dzmzd                   1/1       Running   0          3d
kube-prometheus-exporter-node-gwtzf                   1/1       Running   0          3d
kube-prometheus-exporter-node-j6b9z                   1/1       Running   0          3d
kube-prometheus-exporter-node-jwbww                   1/1       Running   0          3d
kube-prometheus-exporter-node-qjzkl                   1/1       Running   0          3d
kube-prometheus-exporter-node-rgn4k                   1/1       Running   0          3d
kube-prometheus-grafana-767b89d7db-trqfg              2/2       Running   0          3d
portal-oauth2-proxy-59c679bdd6-tkzvk                  1/1       Running   0          8h
portal-portal-55969cc665-499n5                        1/1       Running   0          8h
portal-portal-55969cc665-7cpz8                        1/1       Running   0          8h
portal-proxy-f8fd4b876-l5zkk                          1/1       Running   0          8h
portal-proxy-f8fd4b876-mlrmn                          1/1       Running   0          8h
prometheus-kube-prometheus-prometheus-0               2/2       Running   0          3d

what

• Add datadog helmfile

why

• To provision Datadog chart - hosted infrastructure monitoring platform

✅ chamber exec kops -- helmfile --file helmfile.yaml --selector namespace=datadog,chart=datadog sync
✅ kubectl get pods -n datadog
NAME                                         READY     STATUS    RESTARTS   AGE
datadog-datadog-26pgw                        1/1       Running   0          50m
datadog-datadog-5h4zc                        1/1       Running   0          50m
datadog-datadog-bp2nv                        1/1       Running   0          50m
datadog-datadog-crglk                        1/1       Running   0          50m
datadog-datadog-gwhtt                        1/1       Running   0          50m
datadog-datadog-k99rw                        1/1       Running   0          50m
datadog-datadog-p8wjh                        1/1       Running   0          50m
datadog-kube-state-metrics-6799fc6bb-n582r   1/1       Running   0          50m

test

kube-state-metrics reported

kubernetes_state.container.cpu_limit
kubernetes_state.container.cpu_requested
kubernetes_state.container.memory_limit
kubernetes_state.container.memory_requested
kubernetes_state.container.ready
kubernetes_state.container.restarts
kubernetes_state.container.running
kubernetes_state.container.status_report.count.terminated
kubernetes_state.container.status_report.count.waiting
kubernetes_state.container.terminated
kubernetes_state.container.waiting
kubernetes_state.daemonset.desired
kubernetes_state.daemonset.misscheduled
kubernetes_state.daemonset.ready
kubernetes_state.daemonset.scheduled
kubernetes_state.deployment.paused
kubernetes_state.deployment.replicas
kubernetes_state.deployment.replicas_available
kubernetes_state.deployment.replicas_desired
kubernetes_state.deployment.replicas_unavailable
kubernetes_state.deployment.replicas_updated
kubernetes_state.deployment.rollingupdate.max_unavailable
kubernetes_state.job.failed
kubernetes_state.job.succeeded
kubernetes_state.limitrange.cpu.default_request
kubernetes_state.node.cpu_allocatable
kubernetes_state.node.cpu_capacity
kubernetes_state.node.memory_allocatable
kubernetes_state.node.memory_capacity
kubernetes_state.node.pods_allocatable
kubernetes_state.node.pods_capacity
kubernetes_state.node.status
kubernetes_state.nodes.by_condition
kubernetes_state.persistentvolumeclaim.status
kubernetes_state.pod.ready
kubernetes_state.pod.scheduled
kubernetes_state.replicaset.fully_labeled_replicas
kubernetes_state.replicaset.replicas
kubernetes_state.replicaset.replicas_desired
kubernetes_state.replicaset.replicas_ready
kubernetes_state.statefulset.replicas
kubernetes_state.statefulset.replicas_current
kubernetes_state.statefulset.replicas_desired
kubernetes_state.statefulset.replicas_ready
kubernetes_state.statefulset.replicas_updated

what

• Add prometheus-operator and kube-prometheus

why

• k8s cluster metrics collection, monitoring and alerting

what

• Use BANNER instead of KOPS_CLUSTER_NAME for prompt

why

• When provisioning an architecture without kops, KOPS_CLUSTER_NAME does not exist, and geodesic shows the default prompt not related to the current solution
• geodesic is designed to work with many different architectures, not only kops