Introduction

The ONAP Operations Manager provides a set of capabilities that facilitate Carrier Grade deployments of ONAP.  ONAP deployments need to be capable of offering service while under adverse conditions typically with overall availability measured at five-nines or 99.999% uptime or about 5 minutes of downtime per year.  This requirement might be strict for an orchestration system, but keep in mind that ONAP’s closed loop control system could be providing monitoring a control for one or more critical VNFs that need to meet stringent up-time requirements as found in the TL 9000 Quality Management System Measurements Handbook.  

The Road to High Availability

The progression of the ONAP project towards a fully Carrier Grade has started and will continue over the Beijing or possibly even subsequent releases.  The steps along this progression are roughly as follows:  

For each of these steps the following sections describe the requirements in more detail and the technologies used to achieve it.

Highly Available Kubernetes Deployments

There is a high degree of variability possible in the deployment of Kubernetes.  In some cases it may be installed and managed by hand, done with 3rd party tools like Rancher or even provided by a cloud provider like Microsoft Azure Container Service - Kubernetes has a description of the options here. Kubernetes provides guidance on creating deployments that may be suitable for carrier grade deployments of ONAP on their Building High-Availability Clusters wiki page.

Reliable and Repeatable Deployment

During the Amsterdam release OOM provided a set of capabilities to deploy some or all the ONAP components rapidly and efficiently as a cloud native application with the Kubernetes container orchestration system (note that DCAE is an exception here as DCAE provides its own orchestration system). Each of the components has a deployment specification that describes not only the containers and the container requirements but the relationships or dependencies between the containers.  These dependencies dictate the order in-which the containers are started for the first time such that such dependencies are always met without arbitrary sleep times between container startups.  For example, the SDC back-end container requires the Elastic-Search, Cassandra and Kibana containers within SDC to be ready and is also dependent on DMaaP (or the message-router) to be ready before becoming fully operational. Here is the deployment specification that describes these dependencies:

apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  labels:
    app: sdc-be
  name: sdc-be
  namespace: "{{ .Values.nsPrefix }}-sdc"
spec:
  selector:
    matchLabels:
      app: sdc-be
  template:
    metadata:
      labels:
        app: sdc-be
      name: sdc-be
      annotations:
        pod.beta.kubernetes.io/init-containers: '[
          {
              "args": [
                  "--container-name",
                  "sdc-es",
                  "--container-name",
                  "sdc-cs",
                  "--container-name",
                  "sdc-kb"
              ],
              "command": [
                  "/root/ready.py"
              ],
              "env": [
                  {
                      "name": "NAMESPACE",
                      "valueFrom": {
                          "fieldRef": {
                              "apiVersion": "v1",
                              "fieldPath": "metadata.namespace"
                          }
                      }
                  }
              ],
              "image": "{{ .Values.image.readiness }}",
              "imagePullPolicy": "{{ .Values.pullPolicy }}",
              "name": "sdc-be-readiness"
          },
          {
              "args": [
                  "--container-name",
                  "dmaap"
              ],
              "command": [
                  "/root/ready.py"
              ],
              "env": [
                  {
                      "name": "NAMESPACE",
                      "value": "{{ .Values.nsPrefix }}-message-router"
                  }
              ],
              "image": "{{ .Values.image.readiness }}",
              "imagePullPolicy": "{{ .Values.pullPolicy }}",
              "name": "sdc-dmaap-readiness"
          }
          ]'
    spec:
      containers:
      - env:
        - name: ENVNAME
          value: AUTO
        - name: HOST_IP
          valueFrom:
            fieldRef:
              fieldPath: status.podIP
        image: {{ .Values.image.sdcBackend }}
        imagePullPolicy: {{ .Values.pullPolicy }}
        name: sdc-be
        volumeMounts:
        - mountPath: /usr/share/elasticsearch/data/
          name: sdc-sdc-es-es
        - mountPath: /root/chef-solo/environments/
          name: sdc-environments
        - mountPath: /etc/localtime
          name: sdc-localtime
          readOnly: true
        - mountPath: /var/lib/jetty/logs
          name: sdc-logs
        - mountPath: /var/log/onap
          name: sdc-logs-2
        - mountPath: /tmp/logback.xml
          name: sdc-logback
        lifecycle:
          postStart:
            exec:
              command: ["/bin/sh", "-c", "export LOG=wait_logback.log; touch $LOG; export SRC=/tmp/logback.xml; export DST=/var/lib/jetty/config/catalog-be/; while [ ! -e $DST ]; do echo 'Waiting for $DST...' >> $LOG; sleep 5; done; sleep 2; /bin/cp -f $SRC $DST; echo 'Done' >> $LOG"]
        ports:
        - containerPort: 8443
        - containerPort: 8080
        readinessProbe:
          tcpSocket:
            port: 8443
          initialDelaySeconds: 5
          periodSeconds: 10
      - image: {{ .Values.image.filebeat }}
        imagePullPolicy: {{ .Values.pullPolicy }}
        name: filebeat-onap
        volumeMounts:
        - mountPath: /usr/share/filebeat/filebeat.yml
          name: filebeat-conf
        - mountPath: /var/log/onap
          name: sdc-logs-2
        - mountPath: /usr/share/filebeat/data
          name: sdc-data-filebeat
      volumes:
        - name: filebeat-conf
          hostPath:
            path: /dockerdata-nfs/{{ .Values.nsPrefix }}/log/filebeat/logback/filebeat.yml
        - name: sdc-logs-2
          emptyDir: {}
        - name: sdc-data-filebeat
          emptyDir: {}
        - name: sdc-logback
          hostPath:
            path: /dockerdata-nfs/{{ .Values.nsPrefix }}/log/sdc/be/logback.xml
        - name: sdc-sdc-es-es
          hostPath:
            path: /dockerdata-nfs/{{ .Values.nsPrefix }}/sdc/sdc-es/ES
        - name: sdc-environments
          hostPath:
            path: /dockerdata-nfs/{{ .Values.nsPrefix }}/sdc/environments
        - name: sdc-localtime
          hostPath:
            path:  /etc/localtime
        - name:  sdc-logs
          hostPath:
            path:  /dockerdata-nfs/{{ .Values.nsPrefix }}/sdc/logs
      imagePullSecrets:
      - name: "{{ .Values.nsPrefix }}-docker-registry-key"

Another feature that may assist in achieving a repeatable deployment in the presence of faults that may have reduced the capacity of the cloud is assigning priority to the containers such that mission critical components have the ability to evict less critical components.  Kubernetes provides this capability with Pod Priority and Preemption.

Prior to having more advanced carrier grade features available, the ability to at least be able to re-deploy ONAP (or a subset of) reliably provides a level of confidence that should an outage occur the system can be brought back on-line predictably.

Backup and Restore

A critical factor in being able to recover from an ONAP outage is to ensure that critical state isn't lost after a failure.  Much like ephemeral storage on VMs; any state information stored within a container will be lost once the container is restarted - containers are managed as Cattle, not Pets. To ensure that critical state information is retained after a failure, the OOM deployment specifications for the ONAP components use the Kubernetes concept of Persistent Volumes, an external storage facility that has its own lifecycle. The use of a persistent volume is specified in the ONAP deployment specifications.  Here is an example from the sdnc db-deployment.yaml:

apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: sdnc-dbhost
  namespace: "{{ .Values.nsPrefix }}-sdnc"
spec:
  selector:
    matchLabels:
      app: sdnc-dbhost
  template:
    metadata:
      labels:
        app: sdnc-dbhost
      name: sdnc-dbhost
    spec:
      containers:
      - env:
        - name: MYSQL_ROOT_PASSWORD
          value: openECOMP1.0
        - name: MYSQL_ROOT_HOST
          value: '%'
        image: {{ .Values.image.mysqlServer }}
        imagePullPolicy: {{ .Values.pullPolicy }}
        name: sdnc-db-container
        volumeMounts:
        - mountPath: /etc/localtime
          name: localtime
          readOnly: true
        - mountPath: /var/lib/mysql
          name: sdnc-data
        ports:
        - containerPort: 3306
        readinessProbe:
          tcpSocket:
            port: 3306
          initialDelaySeconds: 5
          periodSeconds: 10
      volumes:
      - name: localtime
        hostPath:
          path: /etc/localtime
      - name: sdnc-data
        persistentVolumeClaim:
          claimName: sdnc-db
      imagePullSecrets:
      - name: "{{ .Values.nsPrefix }}-docker-registry-key"

At the bottom of the deployment specification is a list of the volumes, localtime and sdnc-data which uses an external persistentVolumeClaim of sdnc-db.  This claim is specified as follows:

apiVersion: v1
kind: PersistentVolume
metadata:
  name: "{{ .Values.nsPrefix }}-sdnc-db"
  namespace: "{{ .Values.nsPrefix }}-sdnc"
  labels:
    name: "{{ .Values.nsPrefix }}-sdnc-db"
spec:
  capacity:
    storage: 2Gi
  accessModes:
    - ReadWriteMany
  persistentVolumeReclaimPolicy: Retain
  hostPath:
    path: /dockerdata-nfs/{{ .Values.nsPrefix }}/sdnc/data
---
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: sdnc-db
  namespace: "{{ .Values.nsPrefix }}-sdnc"
spec:
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 2Gi
  selector:
    matchLabels:
      name: "{{ .Values.nsPrefix }}-sdnc-db"

Many different types of storage are supported by this capability such as: GCEPersistentDisk, AWSElasticBlockStore, AzureFile, AzureDisk, FC (Fibre Channel), FlexVolume, Flocker, NFS, iSCSI, RBD (Ceph Block Device), CephFS, Cinder (OpenStack block storage), Glusterfs, VsphereVolume, Quobyte Volumes, HostPath (Single node testing only), VMware Photon, Portworx Volumes, ScaleIO Volumes, and StorageOS. 

As critical state is stored outside of the ONAP containers on a storage media specific to the cloud environment, specific instructions on how to backup and restore such storage is outside of the scope of ONAP.

Health Monitoring

All highly available systems include at least one facility to monitor the health of components within the system.  Such health monitors are often used as inputs to distributed coordination systems (such as etcd, zookeeper, or consul) and monitoring systems (such as nagios or zabbix).  Within ONAP Consul is the monitoring system of choice and deployed by OOM in two parts.  A three-way, centralized Consul server cluster is deployed as a highly available monitor of all of the ONAP components.  The Consul server provides a user interface that allows a user to graphically view the current health status of all of the ONAP components for which agents have been created - a sample from the ONAP Integration labs follows.  Monitoring of ONAP components is configured in the agents within JSON files and stored in gerrit under the consul-agent-config.

Initially the Consul agents are using the same health monitoring facilities as the robot test infrastructure which are typically just validating that the end-point is reachable.  Some health checks already support more advanced checking - such as validating that a database is able to create, update and delete an entry. Consul exposes an API that allows external agents to use the results of the health check, such as the Kubernetes "liveness" probes described below.  

Component Recoverability

OOM deploys ONAP with Kubernetes defined by deployment specifications as mentioned earlier.  These same deployment specifications are also used to implement automatic recoverability of ONAP components when individual components fail. Once ONAP is deployed, a "liveness" probe starts checking the health of the components after a specified startup time.  These liveness probes can simply check that a port is available, that a built-in health check is reporting good health, or that the Consul health check is positive. Should a liveness probe indicate a failed container it will be restarted as described in the deployment specification.  Should the deployment specification indicate that there are one or more dependencies to this container or component (for example a dependency on a database) the dependency will be satisfied before the container/component is restarted. This mechanism ensures that, after a failure, all of the ONAP components restart successfully.  Note that, during the Amsterdam release, deployment specifications were created for all ONAP components but not all of these deployment specifications are restartable (idempotent).  Further work is required during the Beijing release to ensure recoverability of all the ONAP components.

Centralized Logging

An important tool in achieving minimal downtime is the ability to rapidly diagnose problems and determine the root cause.  The Logging Enhancements Project have been building a centralized log collection system based on the Elastic Stack and a Filebeat collector container that is instantiated alongside the containers for each of the ONAP components.  Here is an example from the aai-traversal deployment specification:

    spec:
      containers:
      - name: aai-traversal
[...]
      - name: filebeat-onap-aai-traversal
        image: {{ .Values.image.filebeat }}
        imagePullPolicy: {{ .Values.pullPolicy }}
        volumeMounts:
        - mountPath: /usr/share/filebeat/filebeat.yml
          name: filebeat-conf
        - mountPath: /var/log/onap
          name: aai-traversal-logs
        - mountPath: /usr/share/filebeat/data
          name: aai-traversal-filebeat

Filebeat collects logs from within the namespace of each component and ships them to the centralized logging stack that was deployed by OOM with the other ONAP components.  Users are able to point their web browsers to the Kibana component and see all of the raw logs as well as predefined dashboards that show the state of ONAP in real-time. 

Intra Component Clustering

The OOM project is not responsible to creating highly available versions of all of the ONAP components but does provide via Kubernetes many built in facilities to build clustered, highly available systems including: Services with load-balancers (including support for External Load Balancers), Ingress Resources, and Replica Sets. Some of the open-source projects that form the basis of ONAP components directly support clustered configurations like ODL with instructions on Setting Up Clustering or MariaDB Getting Started with MariaDB Galera Cluster. .  

OOM uses the Kubernetes service abstraction to provide a consistent access point for each of the ONAP components, independent of the pod or container architecture of that component.  For example, the SDN-C component may introduce OpenDaylight clustering as some point and change the number of pods in this component to three or more, but this change will be isolated from the other ONAP components by the service abstraction.  A service can include a load balancer on its ingress to distribute traffic between the pods and even react to dynamic changes in the number of pods if they are part of a replica set. A replica set is a construct that is used to describe the desired state of the cluster.  For example 'replicas: 3' indicates to Kubernetes that a cluster of 3 instances is the desired state.  Should one of the members of the cluster fail, a new member will be automatically started to replace it.

Some of the ONAP components many need a more deterministic deployment; for example to enable intra-cluster communication. For these applications the component can be deployed as a Kubernetes StatefulSet which will maintain a persistent identifier for the pods and thus a stable network id for the pods. For example: the pod names might be web-0, web-1, web-{N-1} for N 'web' pods with corresponding DNS entries such that intra service communication is simple even if the pods are physically distributed across multiple nodes. An example of how these capabilities can be used is described in the Running Consul on Kubernetes tutorial.  

Anti-Affinity Rules

OOM will use the rich set of Kubernetes node and pod affinity / anti-affinity rules to minimize the chance of a single failure resulting in a loss of ONAP service. Node affinity / anti-affinity is used to guide the Kubernetes orchestrator in the placement of pods on nodes (physical or virtual machines).  For example if a container used Intel DPDK technology the pod may state that it as affinity to an Intel processor based node.  For more information refer to Assigning Pods to Nodes. Within the context of Carrier Grade, node affinity or anti-affinity is typically not applicable however, pod affinity is as explained below.

Pod affinity / anti-affiity is the concept of creating a spacial relationship between pods when the Kubernetes orchestrator does assignment (both initially an in operation) to nodes as explained in Inter-pod affinity and anti-affinity. For example, one might choose to co-located all of the ONAP SDC containers on a single node as they are not cirtical runtime components and co-location minimizes overhead. On the other hand, one might choose to ensure that all of the containers in an ODL cluster (SDNC and APPC) are placed on separate nodes such that a node failure has minimal impact to the operation of the cluster.  An example of how pod affinity / anti-affinity is shown below:

apiVersion: v1
kind: Pod
metadata:
  name: with-pod-affinity
spec:
  affinity:
    podAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
      - labelSelector:
          matchExpressions:
          - key: security
            operator: In
            values:
            - S1
        topologyKey: failure-domain.beta.kubernetes.io/zone
    podAntiAffinity:
      preferredDuringSchedulingIgnoredDuringExecution:
      - weight: 100
        podAffinityTerm:
          labelSelector:
            matchExpressions:
            - key: security
              operator: In
              values:
              - S2
          topologyKey: kubernetes.io/hostname
  containers:
  - name: with-pod-affinity
    image: gcr.io/google_containers/pause:2.0

 This example contains both podAffinity and podAntiAffinity rules, the first rule is is a must (requiredDuringSchedulingIgnoredDuringExecution) while the second will be met pending other considerations (preferredDuringSchedulingIgnoredDuringExecution).

  Taints and Tolerations

ONAP S/W Upgrades & Rollbacks

Container Lifecycle Hooks

Geo-Redundant Deployments

Federation,

List of Epics

The following list of JIRA Epics represent the development activities required to complete the OOM related carrier grade activities (to be confirmed):