Post Deployment Validations
-
Check Components in Kubernetes:
- Ensure all Yeedu components are successfully deployed in the Kubernetes cluster.
- Verify the status and health of each component, including but not limited to:
- Deployment pods
- Services
- ConfigMaps
- Secrets
- Use Kubernetes commands such as
kubectl get pods,kubectl get services, etc., to inspect the status of the deployed components.
-
Access UI from Browser:
- Open a web browser.
- Navigate to the URLs associated with the Yeedu UI components, such as:
- Yeedu REST API
- RabbitMQ UI
- Grafana
- Ensure that the UIs are accessible and responsive, indicating successful deployment and availability of Yeedu Runtime services.
1. Components in Kubernetes
Validate the created components such as configmaps, services, Persistent Volume Claims, Persistent Volumes, Deployments, Pods, Ingress, Statefulsets. The outcome should depict something as shown below.
1.1. ConfigMaps
The following should be the output when you list configmaps:
$ kubectl get configmaps -n $YEEDU_SYSTEM_KUBERNETES_NAMESPACE
NAME DATA AGE
ingress-nginx-controller 5 23m
kube-root-ca.crt 1 31d
tcp-services 9 23m
yeedu-aws-retention-json 1 4m10s
yeedu-azure-retention-json 1 4m9s
yeedu-cloud-credential-json 1 4m12s
yeedu-core-ingress-controller 1 16d
yeedu-docker-registry-credential-json 1 4m11s
yeedu-gcp-retention-json 1 4m8s
yeedu-grafana-config-files 1 3m56s
yeedu-grafana-config-files-all 1 4m1s
yeedu-grafana-config-files-grafana-home-dashboard 1 4m
yeedu-grafana-config-files-influxdb 1 4m4s
yeedu-grafana-config-files-ini 1 3m59s
yeedu-grafana-config-files-ldap 1 4m3s
yeedu-ldap-files 1 4m7s
yeedu-rabbitmq-files 2 4m6s
yeedu-rabbitmq3-plugins-conf 1 3m1s
yeedu-rabbitmq3-server-conf 2 3m1s
yeedu-redis-files 2 4m5s
yeedu-telegraf-config-files 1 3m55s
1.2. Services
You should be seeing the below output when you list services
$ kubectl get svc -n $YEEDU_SYSTEM_KUBERNETES_NAMESPACE
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
ingress-nginx-controller LoadBalancer 10.0.226.237 10.4.6.156 80:30593/TCP,443:30796/TCP,3000:32196/TCP,10000:31674/TCP,5432:31148/TCP,389:31589/TCP,6379:31539/TCP,5672:30978/TCP,15672:32195/TCP,8080:31028/TCP,8086:31665/TCP 25m
ingress-nginx-controller-admission ClusterIP 10.0.19.225 <none> 443/TCP 25m
yeedu-grafana-clusterip ClusterIP 10.0.120.102 <none> 3000/TCP 2m57s
yeedu-history-server-clusterip ClusterIP 10.0.183.238 <none> 10000/TCP 2m56s
yeedu-influxdb-clusterip ClusterIP 10.0.190.71 <none> 8086/TCP 2m55s
yeedu-ldap-clusterip ClusterIP 10.0.118.52 <none> 389/TCP 5m12s
yeedu-metadata-db-clusterip ClusterIP 10.0.201.94 <none> 5432/TCP 5m13s
yeedu-rabbitmq3 ClusterIP 10.0.144.204 <none> 5672/TCP,15672/TCP,15692/TCP 4m36s
yeedu-rabbitmq3-nodes ClusterIP None <none> 4369/TCP,25672/TCP 4m36s
yeedu-rabbitmq3-rabbitmq-messaging-topology-operator-webhook ClusterIP 10.0.89.93 <none> 443/TCP 4m54s
yeedu-redis-clusterip ClusterIP 10.0.231.179 <none> 6379/TCP 5m13s
yeedu-restapi-clusterip ClusterIP 10.0.46.19 <none> 8080/TCP 2m55s
1.3. Persistent Volume Claims
You'll see the below output when you list pvc
$ kubectl get pvc -n $YEEDU_SYSTEM_KUBERNETES_NAMESPACE
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
persistence-yeedu-rabbitmq3-server-0 Bound pvc-13b9c5c5-a07b-4c92-b40b-6439d6a55793 20Gi RWO yeedu-storage-class 5m13s
persistence-yeedu-rabbitmq3-server-1 Bound pvc-6a46ff4c-3afe-480b-a878-3b1654e3bfcd 20Gi RWO yeedu-storage-class 5m13s
persistence-yeedu-rabbitmq3-server-2 Bound pvc-d15ab56e-67dc-440b-82f4-74fe114dbd49 20Gi RWO yeedu-storage-class 5m13s
yeedu-influxdb-volume Bound pvc-0742419c-7a7c-4853-b5af-93909c72136d 200Gi RWO yeedu-storage-class 3m34s
yeedu-ldap-pvc Bound pvc-14a45bb0-c7bd-4924-bdc0-06b1db854a29 10Gi RWO yeedu-storage-class 5m51s
yeedu-ldap-slapd-pvc Bound pvc-ee59d86c-97fb-4b99-8b3b-aeb309773faa 10Gi RWO yeedu-storage-class 5m51s
yeedu-metadata-db-pvc Bound pvc-3feee8a7-67c6-4f0a-be18-ec814b4bd101 20Gi RWO yeedu-storage-class 5m51s
yeedu-reactors-logs-volume Bound pvc-b718c992-9ef7-4c0d-8b45-1968ef8e6918 100Gi RWO yeedu-storage-class 3m35s
yeedu-redis-volume Bound pvc-9a1352cd-a588-455d-894d-0807c5d95c88 50Gi RWO yeedu-storage-class 5m51s
1.4. Persistent Volumes
You should be seeing the below output when you list pv
$ kubectl get pv -n $YEEDU_SYSTEM_KUBERNETES_NAMESPACE
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pvc-0742419c-7a7c-4853-b5af-93909c72136d 200Gi RWO Delete Bound yeedu/yeedu-influxdb-volume yeedu-storage-class 3m57s
pvc-13b9c5c5-a07b-4c92-b40b-6439d6a55793 20Gi RWO Delete Bound yeedu/persistence-yeedu-rabbitmq3-server-0 yeedu-storage-class 5m37s
pvc-14a45bb0-c7bd-4924-bdc0-06b1db854a29 10Gi RWO Delete Bound yeedu/yeedu-ldap-pvc yeedu-storage-class 6m14s
pvc-3feee8a7-67c6-4f0a-be18-ec814b4bd101 20Gi RWO Delete Bound yeedu/yeedu-metadata-db-pvc yeedu-storage-class 6m15s
pvc-6a46ff4c-3afe-480b-a878-3b1654e3bfcd 20Gi RWO Delete Bound yeedu/persistence-yeedu-rabbitmq3-server-1 yeedu-storage-class 5m37s
pvc-9a1352cd-a588-455d-894d-0807c5d95c88 50Gi RWO Delete Bound yeedu/yeedu-redis-volume yeedu-storage-class 6m15s
pvc-b718c992-9ef7-4c0d-8b45-1968ef8e6918 100Gi RWO Delete Bound yeedu/yeedu-reactors-logs-volume yeedu-storage-class 3m58s
pvc-d15ab56e-67dc-440b-82f4-74fe114dbd49 20Gi RWO Delete Bound yeedu/persistence-yeedu-rabbitmq3-server-2 yeedu-storage-class 5m37s
pvc-ee59d86c-97fb-4b99-8b3b-aeb309773faa 10Gi RWO Delete Bound yeedu/yeedu-ldap-slapd-pvc yeedu-storage-class 6m14s
1.5. Deployments
kubectl get deployment -n $YEEDU_SYSTEM_KUBERNETES_NAMESPACE
Every workload name is prefixed with the Helm release name it belongs to. We install Yeedu as two Helm releases, yeedu-tps (third-party services) and yeedu-ycs (Yeedu core services), plus the ingress controller. The Deployments our current charts create are:
| Deployment | Release | Created when |
|---|---|---|
<tps-release>-postgres | yeedu-tps | YEEDU_DEPLOY_LOCAL_POSTGRES=true |
<tps-release>-ldap | yeedu-tps | YEEDU_DEPLOY_LOCAL_LDAP=true |
| RabbitMQ cluster operator | yeedu-tps | Always (Bitnami rabbitmq-cluster-operator subchart) |
| RabbitMQ messaging topology operator | yeedu-tps | Always (Bitnami rabbitmq-cluster-operator subchart) |
<ycs-release>-restapi | yeedu-ycs | Always |
<ycs-release>-ui | yeedu-ycs | Always |
<ycs-release>-vault | yeedu-ycs | Always |
<ycs-release>-history-server | yeedu-ycs | Always |
<ycs-release>-reactors-monitor | yeedu-ycs | Always |
<ycs-release>-grafana | yeedu-ycs | YEEDU_ENABLE_GRAFANA_MONTORING=true |
<ycs-release>-influxdb | yeedu-ycs | YEEDU_ENABLE_GRAFANA_MONTORING=true |
<ycs-release>-functions-scheduler | yeedu-ycs | Always |
<ycs-release>-functions-proxy | yeedu-ycs | Always |
<ycs-release>-functions-celery | yeedu-ycs | Always |
| Apache Airflow deployments (subchart) | yeedu-ycs | YEEDU_ENABLE_SCHEDULER=true |
ingress-nginx-controller | ingress | Always |
note
yeedu-reactors-cosi is a StatefulSet, not a Deployment, and log synchronisation runs as the yeedu-reactors-logs-sync container inside that StatefulSet's pods. There's no separate log-sync Deployment.
1.6. Pods
kubectl get pods -n $YEEDU_SYSTEM_KUBERNETES_NAMESPACE
Expect one pod per Deployment replica listed above, the StatefulSet pods listed in section 1.7, and a number of completed one-shot Jobs from the install (ingress admission jobs, the Yeedu init job, and the ingress/DNS patch jobs). All long-running pods should reach Running with all containers ready; the Jobs should show Completed.
1.7. Statefulsets
kubectl get statefulset -n $YEEDU_SYSTEM_KUBERNETES_NAMESPACE
Our current charts create:
| StatefulSet | Release | Notes |
|---|---|---|
<ycs-release>-reactors-cosi | yeedu-ycs | Runs the reactors container and the log-sync container |
| RabbitMQ cluster server | yeedu-tps | Created by the RabbitmqCluster custom resource, 3 replicas |
| Redis master and replicas | yeedu-tps | Bitnami Redis subchart, replica.replicaCount: 3 |
1.8. Secrets
kubectl get secret -n $YEEDU_SYSTEM_KUBERNETES_NAMESPACE
Expect at minimum:
yeedu-certs(kubernetes.io/tls): the ingress certificate built fromssl/yeedu.crtandssl/yeedu.keyyeedu-image-pull-secret(kubernetes.io/dockerconfigjson): the container registry credentials- The RabbitMQ default-user and Erlang-cookie secrets, and the messaging-topology-operator webhook certificate
- The ingress admission webhook secret
- One
sh.helm.release.v1.<release>.v<n>secret per Helm release. With the current layout that isyeedu-tps,yeedu-ycsand the ingress release
You can list the Helm releases directly instead:
helm list -n $YEEDU_SYSTEM_KUBERNETES_NAMESPACE
1.9. Ingress Services
You'll see the below output when you list ingress services
$ kubectl get svc -n yeedu
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
ingress-nginx-controller LoadBalancer 10.0.226.237 10.4.6.156 80:30593/TCP,443:30796/TCP,3000:32196/TCP,10000:31674/TCP,5432:31148/TCP,389:31589/TCP,6379:31539/TCP,5672:30978/TCP,15672:32195/TCP,8080:31028/TCP,8086:31665/TCP 6m35s
ingress-nginx-controller-admission ClusterIP 10.0.19.225 <none> 443/TCP 6m35s
1.10. Ingress Pods
You should be seeing the below output when you list Ingress Pods
$ kubectl get pods -n yeedu
NAME READY STATUS RESTARTS AGE
ingress-nginx-admission-create-fzvnz 0/1 Completed 0 5m25s
ingress-nginx-admission-patch-g565f 0/1 Completed 0 5m25s
ingress-nginx-controller-6fdbb469b9-nv8bh 1/1 Running 0 5m26s
2. Access from Browser
2.1. RabbitMQ
2.2. History Server
2.3. Grafana
2.4. Rest-API
