Diagnosing and preventing Network Problems with Health Check

how-to

In today’s distributed and virtual environments, users will often not have full administrative control over their whole network. Health Check introduces Ping to check nodes are still healthy, and to force idle connections to be kept alive in environments with eager shutdowns of unused resources. Diagnostics requests a report from a node, giving instant health check information.

Diagnosing problems in distributed environments is far from easy, so Couchbase provides a Health Check API with Ping() for active monitoring and Diagnostics() for a look at what the client believes is the current state of the cluster. More extensive discussion of the uses of Health Check can be found in the Health Check Concept Guide.

Ping

At its simplest, ping provides information about the current state of the connections in the Couchbase Cluster, by actively polling: Ping is available at the Cluster and Bucket object levels. Below we use it at the Bucket level:

	// We'll ping the KV nodes in our cluster.
	pings, err := bucket.Ping(&gocb.PingOptions{
		ReportID:     "my-report",                                  (1)
		ServiceTypes: []gocb.ServiceType{gocb.ServiceTypeKeyValue}, (2)
	})
	if err != nil {
		panic(err)
	}

	for service, pingReports := range pings.Services {
		if service != gocb.ServiceTypeKeyValue {
			panic("we got a service type that we didn't ask for!")
		}

		for _, pingReport := range pingReports {
			if pingReport.State != gocb.PingStateOk {
				fmt.Printf(
					"Node %s at remote %s is not OK, error: %s, latency: %s\n",
					pingReport.ID, pingReport.Remote, pingReport.Error, pingReport.Latency.String(),
				)
			} else {
				fmt.Printf(
					"Node %s at remote %s is OK, latency: %s\n",
					pingReport.ID, pingReport.Remote, pingReport.Latency.String(),
				)
			}
		}
	}

	b, err := json.Marshal(pings) (3)
	if err != nil {
		panic(err)
	}

	fmt.Printf("Ping report JSON: %s", string(b))

1	ReportID is optional and assigns a name to this report, if empty then a uuid will be assigned.
2	ServiceTypes are which services to ping against.
3	The report can be marshalled down into JSON in a human friendly format.

Diagnostics

Diagnostics returns a list of the nodes that the SDK currently has (or had) a connection to, and the current status of the connection. However this call does not actively poll the nodes, reporting instead the state the last time it tried to access each node. If you want the current status, then use Ping.

As the Go SDK does not hold long lived HTTP connections the Diagnostics operation will only return information about connections to the KeyValue service.

	diagnostics, err := cluster.Diagnostics(&gocb.DiagnosticsOptions{
		ReportID: "my-report", (1)
	})
	if err != nil {
		panic(err)
	}

	if diagnostics.State != gocb.ClusterStateOnline {
		log.Printf("Overall cluster state is not online\n")
	} else {
		log.Printf("Overall cluster state is online\n")
	}

	for serviceType, diagReports := range diagnostics.Services {
		for _, diagReport := range diagReports {
			if diagReport.State != gocb.EndpointStateConnected {
				fmt.Printf(
					"Node %s at remote %s is not connected on service %s, activity last seen at: %s\n",
					diagReport.ID, diagReport.Remote, serviceType, diagReport.LastActivity.String(),
				)
			} else {
				fmt.Printf(
					"Node %s at remote %s is connected on service %s, activity last seen at: %s\n",
					diagReport.ID, diagReport.Remote, serviceType, diagReport.LastActivity.String(),
				)
			}
		}
	}

	db, err := json.Marshal(diagnostics) (2)
	if err != nil {
		panic(err)
	}

	fmt.Printf("Diagnostics report JSON: %s", string(db))

1	ReportID is optional and assigns a name to this report, if empty then a uuid will be assigned.
2	The report can be marshalled down into JSON in a human friendly format.