ONAP - Based [I] SMO Deployment

![](https://i.imgur.com/JORnn3y.png =150x)@NTUST # ONAP - Based [I] SMO Deployment ###### tags: `osc` `smo` :::info **Outline:** [TOC] **Reference:** - [OSC SMO 部屬教學 by NYCU](https://hackmd.io/@KflrygviTAa4Bw7zSKk2CA/H1kOzqEv6) - [ONAP - Based \[H\] SMO Deployment](https://hackmd.io/Cf5E97XqTXSr00xC7bzfeA?view) ::: ## Environment ### Minimum Requirements :::success - **Hardware** ``` 6 or more CPU cores 28+ GB of RAM 50 GB of Disk ``` - **Software** ``` Ubuntu 22.04 server Kubernetes v1.27.5 python3 3.10.12 # Version should upper or equal to 3.9 ``` ::: ## Prerequisite ### 1. Install Kubernetes with kubespray ```bash= python3 --version # Version should upper or equal to 3.9 apt install -y python3-pip git clone https://github.com/kubernetes-sigs/kubespray -b release-2.23 cd kubespray pip install -r requirements.txt sed -i 's/\(kube_version: \)[^"]*/\1v1.27.5/' inventory/local/group_vars/k8s_cluster/k8s-cluster.yml ansible-playbook -i inventory/local/hosts.ini --become --become-user=root cluster.yml mkdir -p ~/.kube/config sudo cp /etc/kubernetes/admin.conf ~/.kube/config ``` ## Deployment ### 1. Download the IT/dep repository from gerrit ```bash= # Latest version cd ~ git clone https://gerrit.o-ran-sc.org/r/it/dep.git -b master --recursive ``` :::warning :bulb: **Note:** You need to ==add the recurse sub modules flag== as some parts are git submodules pointing to existing related charts (ONAP) ::: ### 2. Setup Helm Charts - Execute the following commands being logged as root. ```bash= cd dep ##Setup ChartMuseum ./smo-install/scripts/layer-0/0-setup-charts-museum.sh ##Setup HELM3 ./smo-install/scripts/layer-0/0-setup-helm3.sh ## Build ONAP/ORAN charts ./smo-install/scripts/layer-1/1-build-all-charts.sh ``` ### 3. Deploy components - Execute the following commands ==being logged as root==: ```bash= ./smo-install/scripts/layer-2/2-install-oran.sh ``` ### 4. Wait for pod finished ```bash= kubectl get pods -n onap && kubectl get pods -n nonrtric ``` ### 5. Result ![image](https://hackmd.io/_uploads/r1ZMyWuPp.png) ## Deployment of dmaap-InfluxDB-adapter, InfluxDB and Grafana  ### 0. Pre-Request - Finished the installation of ONAP SMO. ### 1. Deploy - Helm repo add and install ```bash= helm repo add winlab https://harbor.winfra.cs.nycu.edu.tw/chartrepo/winlab-oran helm install --namespace=o1ves o1ves winlab/o1ves-visualization --create-namespace ``` :::warning :red_circle: **NOTE:** Using a release name other than o1ves may cause unexpected errors. ::: ### 2. Edit the values.yaml file ```bash= helm show values winlab/o1ves-visualization > values.yaml ``` :::info - Modify the **rules**. - Add buckets to automatically create influxdb buckets. ::: #### `values.yaml` :::spoiler Example `values.yaml` ```yaml= fullnameOverride: o1ves grafana: fullnameOverride: o1ves-grafana enabled: true defaultDashboardsTimezone: Asia/Taipei adminPassword: smo service: type: NodePort nodePort: 30000 persistence: enabled: true storageClassName: "local-storage-grafana" size: 10Gi influxdb2: fullnameOverride: o1ves-influxdb2 enabled: true persistence: enabled: true storageClass: "local-storage-influxdb2" accessMode: ReadWriteOnce size: 50Gi mountPath: /var/lib/influxdb2 subPath: "" image: repository: influxdb tag: 2.3.0-alpine pullPolicy: IfNotPresent dmaap-influxdb-adapter: enabled: true image: repository: harbor.winfra.cs.nycu.edu.tw/winlab-oran/dmaap-influxdb-adapter influxdb: host: o1ves-influxdb2 port: 80 tokenSecret: o1ves-influxdb2-auth org: influxdata logLevel: DEBUG rules: - topic: unauthenticated.SEC_FAULT_OUTPUT rules: - bucket: influxdb measurement: fault matches: - key: event.commonEventHeader.domain value: fault tags: - key: type value: test - key: sourceId field: event.commonEventHeader.sourceId fields: - key: faultFieldsVersion field: event.faultFields.faultFieldsVersion type: string buckets: - name: influxdb image: pullPolicy: IfNotPresent ``` ::: <br> In this case, we use this `values_Akmal.yaml` file. ```yaml= dmaap-influxdb-adapter: image: pullPolicy: Always logLevel: DEBUG rules: - topic: unauthenticated.VES_MEASUREMENT_OUTPUT rules: - bucket: my-bucket measurement: slice-measurement matches: - key: event.commonEventHeader.domain value: stndDefined tags: - key: vendor value: ntust-taiwan-lab - key: sourceName field: event.commonEventHeader.sourceName fields: - key: avg-throughput-downlink field: event.stndDefinedFields.data.measurements[0].value type: int - key: ue-avg-mcs-index field: event.stndDefinedFields.data.measurements[1].value type: float buckets: - name: my-bucket ``` ### Delete ```bash= helm uninstall --namespace o1ves o1ves kubectl delete ns o1ves sudo rm -r /dockerdata-nfs/o1ves-* ``` :::warning #### :red_circle: **NOTE:** - If you edit the `values.yaml` , you need to **upgrade the pod** or delete and redeploy. ```bash= ## Upgrade helm upgrade --namespace=o1ves o1ves winlab/o1ves-visualization --create-namespace -f value2.yml ## Redeploy helm uninstall --namespace o1ves o1ves kubectl delete ns o1ves sudo rm -r /dockerdata-nfs/o1ves-* helm install --namespace=o1ves o1ves winlab/o1ves-visualization --create-namespace -f values.yaml ``` - You can write a ves event then use this command to test sent the event to influxdb. ```bash= curl -X POST \ -H 'Content-Type: application/json' \ -u sample1:sample1 \ -d @file.json \ https://192.168.8.12:30417/eventListener/v7 ``` - Check whether influxdb-adapter receive the event or not. ```bash= kubectl logs -n o1ves o1ves-dmaap-influxdb-adapter-64cf8bf7b8-qf6sg ``` ::: ### InfluxDB, Grafana Web UI and Set InfluxDB Data Source #### InfluxDB Web UI - http://<server-ip/>:30001 > Here we use http://192.168.8.12:30001 - username and password : (admin/) :::warning - Use this command to find the password. ```bash= kubectl get secret -n o1ves o1ves-influxdb2-auth -o json # find the password then echo echo OGMwdDRYRG1xUXNyR29BQ3hCZFdVZkZGeGJqUlRpTlc= | base64 -d ``` - If InfluxDB port didn't open, you can refer this. ```bash= kubectl edit svc -n o1ves o1ves-influxdb2 # line 33 add: nodePort:30001 # line 41 type change to NodePort ``` ::: #### Grafana Web UI - http://<server-ip/>:30000 > Here we use http://192.168.8.12:30000 - username and password : (admin/smo) #### Set influxdb data source - Go Administration/Data sources you will see InfluxDB default. - Check the seeting: - Query Language: Flux - URL: http://o1ves-influxdb2.o1ves - Auth - Basic auth: Disable - InfluxDB Details - Organization: influxdb - Token: my-token - Save & test ### Result - Check pods are running ![image](https://hackmd.io/_uploads/ryf11-dD6.png) - InfluxDB Web UI ![](https://hackmd.io/_uploads/ryMHCeNY2.png) - Grafana Web UI ![](https://hackmd.io/_uploads/rJlJmrJPh.png)