majiayu000 d0bad07bee docs: add Hubble network observability documentation and dashboards

Add documentation explaining how to enable Hubble for network
observability in Grafana. Include four pre-built Hubble dashboards
(overview, dns-namespace, l7-http-metrics, network-overview) and
register them in the monitoring hub's dashboard list.

Closes #749

Signed-off-by: majiayu000 <1835304752@qq.com>

2026-01-02 04:17:56 +08:00

2.8 KiB

Raw Permalink Blame History

Enabling Hubble for Network Observability

Hubble is a network and security observability platform built on top of Cilium. It provides deep visibility into the communication and behavior of services in your Kubernetes cluster.

Prerequisites

Cozystack platform running with Cilium as the CNI
Monitoring hub enabled for Grafana access

Configuration

Hubble is disabled by default in Cozystack. To enable it, update the Cilium configuration.

Enable Hubble

Edit the Cilium values in your platform configuration to enable Hubble:

cilium:
  hubble:
    enabled: true
    relay:
      enabled: true
    ui:
      enabled: true
    metrics:
      enabled:
        - dns
        - drop
        - tcp
        - flow
        - port-distribution
        - icmp
        - httpV2:exemplars=true;labelsContext=source_ip,source_namespace,source_workload,destination_ip,destination_namespace,destination_workload,traffic_direction

Components

When Hubble is enabled, the following components become available:

Hubble Relay: Aggregates flow data from all Cilium agents
Hubble UI: Web-based interface for exploring network flows
Hubble Metrics: Prometheus metrics for network observability

Grafana Dashboards

Once Hubble is enabled and the monitoring hub is deployed, the following dashboards become available in Grafana under the hubble folder:

Dashboard	Description
Overview	General Hubble metrics including processing statistics
DNS Namespace	DNS query and response metrics by namespace
L7 HTTP Metrics	HTTP layer 7 metrics by workload
Network Overview	Network flow overview by namespace

Accessing Dashboards

Navigate to Grafana via the monitoring hub
Browse to the hubble folder in the dashboard browser
Select a dashboard to view network observability data

Metrics Available

Hubble exposes various metrics that can be queried in Grafana:

hubble_flows_processed_total: Total number of flows processed
hubble_dns_queries_total: DNS queries by type
hubble_dns_responses_total: DNS responses by status
hubble_drop_total: Dropped packets by reason
hubble_tcp_flags_total: TCP connections by flag
hubble_http_requests_total: HTTP requests by method and status

Troubleshooting

Verify Hubble Status

Check if Hubble is running:

kubectl get pods -n cozy-cilium -l k8s-app=hubble-relay
kubectl get pods -n cozy-cilium -l k8s-app=hubble-ui

Check Metrics Endpoint

Verify Hubble metrics are being scraped:

kubectl port-forward -n cozy-cilium svc/hubble-metrics 9965:9965
curl http://localhost:9965/metrics

Verify ServiceMonitor

Ensure the ServiceMonitor is created for Prometheus scraping:

kubectl get servicemonitor -n cozy-cilium

2.8 KiB Raw Permalink Blame History