What does eBPF need on the node?

Loading eBPF programs into the kernel needs hostPID, a privileged security context and root, so one Helm flag reconfigures the node DaemonSet with all three. It is opt-in, amd64 and arm64 only, and the flag on its own collects nothing until the portal pushes the receiver config.

Do pods restart when we turn tracing on?

eBPF attaches to sockets, so nothing restarts and nothing gets rebuilt. Operator auto-instrumentation injects a language SDK at admission, so those pods are recreated once. Both layers run together.

Will pod labels wreck our cardinality?

Pod IDs and request IDs are the usual cause. A topk query in the docs ranks your metrics by series count, then write_relabel_configs drops the label or coarsens it to a deployment name before it leaves the cluster. Nothing dropped at the source is billed.

Kubernetes without a sidecar per service.

One DaemonSet enrols into Fleet Management and the pipelines arrive from the portal. Nodes, pods, namespaces and workloads report CPU, memory, network and disk, with no code changes and no per-language agent to keep current.

Start free Read the docs

Included on the free tier: 20,000 active series, 50 GB of logs and 50 GB of traces a month. No card.

insight / kubernetes monitoringLIVE

checkout-7f9c4640m

payments-api-2b1d410m

inventory-5d81a120m

notify-worker-340m

ingress-nginx-x9k880m

Runs on

Amazon EKS
Google GKE
Azure AKS
Red Hat OpenShift
Rancher
k3s

Trademarks

What you get

Included on every plan, including the free tier.

cluster / prod-eu-1

data22.4 GiB

checkout14.2 GiB

payments9.8 GiB

ingress3.1 GiB

observability2.6 GiB

Nodes18

Pods214

Namespaces6

Routes
ROUTE	RATE	ERRORS	P95
POST /api/checkout	42/s	0.3%	412 ms
GET /api/cart	188/s	0.0%	38 ms
POST /api/pay	39/s	2.1%	1.21 s
GET /api/stock	96/s	0.0%	22 ms

Nothing restarted, nothing rebuilt

Server and client spans arrive alongside the metrics, read from the socket.

How it gets there

Three steps, then it is reporting.

One DaemonSet per cluster. The chart enrols it and Fleet Management sends the rest.

Install the agent chart and enrol the cluster.
Assign the cluster pipeline from Fleet Management.
Pods, nodes and workloads report inside a minute.

Get an endpoint Read the setup guide

terminal · helm

# 01 · install the agent chart

helm repo add xscaler <xscaler-chart-repo>

helm install xscaler-agent xscaler/agent \
  --namespace observability --create-namespace \
  --set opamp.endpoint=wss://<your-xscaler-endpoint>/v1/opamp

# 02 · assign the cluster pipeline in Fleet Management

cluster-default v1 → 18 agents · applied

# 03 · cluster reportingreporting

k8s.cluster.name "prod-eu-1"

k8s.node.count 18

k8s.pod.count 214

k8s.pod.phase Running 209

What you can ask

What that looks like as a query.

Paste any of these into Insights, or use them as an alert rule.

sum by (namespace) (container_memory_working_set_bytes)

Which namespace is holding the memory.

Working set, the figure the kubelet uses for eviction.

kube_pod_container_status_restarts_total > 3

Containers stuck in a restart loop.

From kube-state-metrics, unchanged.

sum by (node) (kube_pod_info)
  / on(node) kube_node_status_allocatable{resource="pods"}

Nodes approaching their pod ceiling.

Scheduled pods against what each node can hold.

Questions

What platform teams ask before the rollout.

Something not here? Talk to an engineer who works on the platform.

Book thirty minutes

No. Point its remote_write at us and kube-state-metrics plus kubelet cAdvisor keep scraping exactly as they do today. ServiceMonitors, recording rules and PromQL are unchanged, and Grafana Alloy forwards the same way.

Start on the free tier.

20,000 active series, 50 GB of logs and 50 GB of traces a month, 90-day retention, and the whole of Insights.

Book a demo Start free

Kubernetes without a sidecar per service.

Three steps, then it is reporting.

Which namespace is holding the memory.

Containers stuck in a restart loop.

Nodes approaching their pod ceiling.

Related use cases

Infrastructure monitoring

Application performance

Log management

What platform teams ask before the rollout.

01Do we have to replace Prometheus?

02What does eBPF need on the node?

03Do pods restart when we turn tracing on?

04Will pod labels wreck our cardinality?

Start on the free tier.