/cilium-cli

CLI to install, manage & troubleshoot Kubernetes clusters running Cilium

Primary LanguageGoApache License 2.0Apache-2.0

Next-Gen Cilium CLI (Experimental)

Go Kind EKS (ENI) EKS (tunnel) GKE AKS (BYOCNI) Multicluster External Workloads

Installation

To build and install, use the install target:

make install

You may set the BINDIR environment variable to install the binary in a specific location instead of /usr/local/bin, e.g.

BINDIR=~/.local/bin make install

Alternatively, to install the latest binary release:

CILIUM_CLI_VERSION=$(curl -s https://raw.githubusercontent.com/cilium/cilium-cli/main/stable.txt)
GOOS=$(go env GOOS)
GOARCH=$(go env GOARCH)
curl -L --remote-name-all https://github.com/cilium/cilium-cli/releases/download/${CILIUM_CLI_VERSION}/cilium-${GOOS}-${GOARCH}.tar.gz{,.sha256sum}
sha256sum --check cilium-${GOOS}-${GOARCH}.tar.gz.sha256sum
sudo tar -C /usr/local/bin -xzvf cilium-${GOOS}-${GOARCH}.tar.gz
rm cilium-${GOOS}-${GOARCH}.tar.gz{,.sha256sum}

See https://github.com/cilium/cilium-cli/releases for supported GOOS/GOARCH binary releases.

Releases

Release Maintained Supported Cilium Versions
v0.14.2 Yes Cilium 1.11 and newer
v0.10.7 No Cilium 1.10

Please see Experimental helm installation mode section regarding our plan to migrate to the new helm installation mode and deprecate the current implementation.

Capabilities

Install Cilium

To install Cilium while automatically detected:

cilium install
🔮 Auto-detected Kubernetes kind: minikube
✨ Running "minikube" validation checks
✅ Detected minikube version "1.5.2"
ℹ️  Cilium version not set, using default version "v1.9.1"
🔮 Auto-detected cluster name: minikube
🔮 Auto-detected datapath mode: tunnel
🔑 Found existing CA in secret cilium-ca
🔑 Generating certificates for Hubble...
🚀 Creating service accounts...
🚀 Creating cluster roles...
🚀 Creating ConfigMap...
🚀 Creating agent DaemonSet...
🚀 Creating operator Deployment...

Supported Environments

  • minikube
  • kind
  • EKS
  • self-managed
  • GKE
  • AKS BYOCNI
  • k3s
  • Rancher

Cluster Context Management

cilium context
Context: minikube
Cluster: minikube
Auth: minikube
Host: https://192.168.64.25:8443
TLS server name:
CA path: /Users/tgraf/.minikube/ca.crt

Hubble

cilium hubble enable
🔑 Generating certificates for Relay...
✨ Deploying Relay...

Status

cilium status
    /¯¯\
 /¯¯\__/¯¯\    Cilium:      OK
 \__/¯¯\__/    Operator:    OK
 /¯¯\__/¯¯\    Hubble:      OK
 \__/¯¯\__/
    \__/
DaemonSet         cilium             Desired: 1, Ready: 1/1, Available: 1/1
Deployment        cilium-operator    Desired: 1, Ready: 1/1, Available: 1/1
Deployment        hubble-relay       Desired: 1, Ready: 1/1, Available: 1/1
Containers:       cilium             Running: 1
                  cilium-operator    Running: 1
                  hubble-relay       Running: 1
Image versions    cilium             quay.io/cilium/cilium:v1.9.1: 1
                  cilium-operator    quay.io/cilium/operator-generic:v1.9.1: 1
                  hubble-relay       quay.io/cilium/hubble-relay:v1.9.1: 1

Connectivity Check

cilium connectivity test --single-node
⌛ Waiting for deployments to become ready
🔭 Enabling Hubble telescope...
⚠️  Unable to contact Hubble Relay: rpc error: code = Unavailable desc = connection error: desc = "transport: Error while dialing dial tcp [::1]:4245: connect: connection refused"
⚠️  Did you enable and expose Hubble + Relay?
ℹ️  You can export Relay with a port-forward: kubectl port-forward -n kube-system deployment/hubble-relay 4245:4245
ℹ️  Disabling Hubble telescope and flow validation...
-------------------------------------------------------------------------------------------
🔌 Validating from pod cilium-test/client-9f579495f-b2pcq to pod cilium-test/echo-same-node-7f877bbf9-p2xg8...
-------------------------------------------------------------------------------------------
✅ client pod client-9f579495f-b2pcq was able to communicate with echo pod echo-same-node-7f877bbf9-p2xg8 (10.0.0.166)
-------------------------------------------------------------------------------------------
🔌 Validating from pod cilium-test/client-9f579495f-b2pcq to outside of cluster...
-------------------------------------------------------------------------------------------
✅ client pod client-9f579495f-b2pcq was able to communicate with cilium.io
-------------------------------------------------------------------------------------------
🔌 Validating from pod cilium-test/client-9f579495f-b2pcq to local host...
-------------------------------------------------------------------------------------------
✅ client pod client-9f579495f-b2pcq was able to communicate with local host
-------------------------------------------------------------------------------------------
🔌 Validating from pod cilium-test/client-9f579495f-b2pcq to service echo-same-node...
-------------------------------------------------------------------------------------------
✅ client pod client-9f579495f-b2pcq was able to communicate with service echo-same-node

With Flow Validation

cilium hubble port-forward&
cilium connectivity test --single-node
⌛ Waiting for deployments to become ready
🔭 Enabling Hubble telescope...
Handling connection for 4245
ℹ️  Hubble is OK, flows: 405/4096
-------------------------------------------------------------------------------------------
🔌 Validating from pod cilium-test/client-9f579495f-b2pcq to pod cilium-test/echo-same-node-7f877bbf9-p2xg8...
-------------------------------------------------------------------------------------------
📄 Flow logs of pod cilium-test/client-9f579495f-b2pcq:
Jan  6 13:41:17.739: 10.0.0.11:43876 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: SYN)
Jan  6 13:41:17.739: 10.0.0.166:8080 -> 10.0.0.11:43876 to-endpoint FORWARDED (TCP Flags: SYN, ACK)
Jan  6 13:41:17.739: 10.0.0.11:43876 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK)
Jan  6 13:41:17.739: 10.0.0.11:43876 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK, PSH)
Jan  6 13:41:17.755: 10.0.0.166:8080 -> 10.0.0.11:43876 to-endpoint FORWARDED (TCP Flags: ACK, PSH)
Jan  6 13:41:17.756: 10.0.0.11:43876 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK, FIN)
Jan  6 13:41:17.757: 10.0.0.166:8080 -> 10.0.0.11:43876 to-endpoint FORWARDED (TCP Flags: ACK, FIN)
Jan  6 13:41:17.757: 10.0.0.11:43876 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK)
📄 Flow logs of pod cilium-test/echo-same-node-7f877bbf9-p2xg8:
Jan  6 13:41:17.739: 10.0.0.11:43876 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: SYN)
Jan  6 13:41:17.739: 10.0.0.166:8080 -> 10.0.0.11:43876 to-endpoint FORWARDED (TCP Flags: SYN, ACK)
Jan  6 13:41:17.739: 10.0.0.11:43876 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK)
Jan  6 13:41:17.739: 10.0.0.11:43876 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK, PSH)
Jan  6 13:41:17.755: 10.0.0.166:8080 -> 10.0.0.11:43876 to-endpoint FORWARDED (TCP Flags: ACK, PSH)
Jan  6 13:41:17.756: 10.0.0.11:43876 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK, FIN)
Jan  6 13:41:17.757: 10.0.0.166:8080 -> 10.0.0.11:43876 to-endpoint FORWARDED (TCP Flags: ACK, FIN)
Jan  6 13:41:17.757: 10.0.0.11:43876 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK)
✅ client pod client-9f579495f-b2pcq was able to communicate with echo pod echo-same-node-7f877bbf9-p2xg8 (10.0.0.166)
-------------------------------------------------------------------------------------------
🔌 Validating from pod cilium-test/client-9f579495f-b2pcq to outside of cluster...
-------------------------------------------------------------------------------------------
❌ Found RST in pod cilium-test/client-9f579495f-b2pcq
❌ FIN not found in pod cilium-test/client-9f579495f-b2pcq
📄 Flow logs of pod cilium-test/client-9f579495f-b2pcq:
Jan  6 13:41:22.025: 10.0.0.11:55334 -> 10.0.0.243:53 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.025: 10.0.0.11:55334 -> 10.0.0.243:53 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.027: 10.0.0.243:53 -> 10.0.0.11:55334 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.028: 10.0.0.243:53 -> 10.0.0.11:55334 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.028: 10.0.0.11:56466 -> 10.0.0.104:53 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.028: 10.0.0.11:56466 -> 10.0.0.104:53 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.029: 10.0.0.104:53 -> 10.0.0.11:56466 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.029: 10.0.0.104:53 -> 10.0.0.11:56466 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.030: 10.0.0.11:57691 -> 10.0.0.243:53 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.030: 10.0.0.243:53 -> 10.0.0.11:57691 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.030: 10.0.0.11:57691 -> 10.0.0.243:53 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.031: 10.0.0.243:53 -> 10.0.0.11:57691 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.031: 10.0.0.11:52849 -> 10.0.0.104:53 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.032: 10.0.0.104:53 -> 10.0.0.11:52849 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.033: 10.0.0.11:52849 -> 10.0.0.104:53 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.037: 10.0.0.104:53 -> 10.0.0.11:52849 to-endpoint FORWARDED (UDP)
Jan  6 13:41:22.038: 10.0.0.11:45040 -> 172.217.168.46:443 to-stack FORWARDED (TCP Flags: SYN)
Jan  6 13:41:22.041: 172.217.168.46:443 -> 10.0.0.11:45040 to-endpoint FORWARDED (TCP Flags: SYN, ACK)
Jan  6 13:41:22.041: 10.0.0.11:45040 -> 172.217.168.46:443 to-stack FORWARDED (TCP Flags: ACK)
Jan  6 13:41:22.059: 10.0.0.11:45040 -> 172.217.168.46:443 to-stack FORWARDED (TCP Flags: ACK, PSH)
Jan  6 13:41:22.073: 172.217.168.46:443 -> 10.0.0.11:45040 to-endpoint FORWARDED (TCP Flags: ACK, PSH)
Jan  6 13:41:22.096: 10.0.0.11:45040 -> 172.217.168.46:443 to-stack FORWARDED (TCP Flags: ACK, RST)
Jan  6 13:41:22.097: 172.217.168.46:443 -> 10.0.0.11:45040 to-endpoint FORWARDED (TCP Flags: ACK, FIN)
Jan  6 13:41:22.097: 10.0.0.11:45040 -> 172.217.168.46:443 to-stack FORWARDED (TCP Flags: RST)
✅ client pod client-9f579495f-b2pcq was able to communicate with cilium.io
-------------------------------------------------------------------------------------------
🔌 Validating from pod cilium-test/client-9f579495f-b2pcq to local host...
-------------------------------------------------------------------------------------------
📄 Flow logs of pod cilium-test/client-9f579495f-b2pcq:
Jan  6 13:41:25.305: 10.0.0.11 -> 192.168.64.25 to-stack FORWARDED (ICMPv4 EchoRequest)
Jan  6 13:41:25.305: 192.168.64.25 -> 10.0.0.11 to-endpoint FORWARDED (ICMPv4 EchoReply)
✅ client pod client-9f579495f-b2pcq was able to communicate with local host
-------------------------------------------------------------------------------------------
🔌 Validating from pod cilium-test/client-9f579495f-b2pcq to service echo-same-node...
-------------------------------------------------------------------------------------------
📄 Flow logs of pod cilium-test/client-9f579495f-b2pcq:
Jan  6 13:41:30.499: 10.0.0.11:39559 -> 10.0.0.104:53 to-endpoint FORWARDED (UDP)
Jan  6 13:41:30.499: 10.0.0.11:39559 -> 10.0.0.104:53 to-endpoint FORWARDED (UDP)
Jan  6 13:41:30.500: 10.0.0.104:53 -> 10.0.0.11:39559 to-endpoint FORWARDED (UDP)
Jan  6 13:41:30.500: 10.0.0.104:53 -> 10.0.0.11:39559 to-endpoint FORWARDED (UDP)
Jan  6 13:41:30.503: 10.0.0.11:59414 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: SYN)
Jan  6 13:41:30.503: 10.0.0.166:8080 -> 10.0.0.11:59414 to-endpoint FORWARDED (TCP Flags: SYN, ACK)
Jan  6 13:41:30.503: 10.0.0.11:59414 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK)
Jan  6 13:41:30.503: 10.0.0.11:59414 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK, PSH)
Jan  6 13:41:30.505: 10.0.0.166:8080 -> 10.0.0.11:59414 to-endpoint FORWARDED (TCP Flags: ACK, PSH)
Jan  6 13:41:30.509: 10.0.0.11:59414 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK, FIN)
Jan  6 13:41:30.509: 10.0.0.166:8080 -> 10.0.0.11:59414 to-endpoint FORWARDED (TCP Flags: ACK, FIN)
Jan  6 13:41:30.509: 10.0.0.11:59414 -> 10.0.0.166:8080 to-endpoint FORWARDED (TCP Flags: ACK)
✅ client pod client-9f579495f-b2pcq was able to communicate with service echo-same-node

Network Performance test

cilium connectivity test --perf
🔥 Performance Test Summary
-----------------------------------------------------------------------------------------------------------------------------
📋 Scenario                                           | Test            | Num Samples     | Duration        | Avg value      
-----------------------------------------------------------------------------------------------------------------------------
📋 perf-client-5d7cb4d587-cn8sw                       | TCP_RR          | 1               | 10s             | 29975.37 (OP/s)
📋 perf-client-5d7cb4d587-cn8sw                       | TCP_CRR         | 1               | 10s             | 3926.56 (OP/s)
📋 perf-client-5d7cb4d587-cn8sw                       | TCP_STREAM      | 1               | 10s             | 2275.42 (Mb/s)
📋 perf-client-other-node-7867748554-vfvgt            | TCP_RR          | 1               | 10s             | 964.55 (OP/s)
📋 perf-client-other-node-7867748554-vfvgt            | TCP_STREAM      | 1               | 10s             | 4743.39 (Mb/s)
📋 perf-client-other-node-7867748554-vfvgt            | UDP_RR          | 1               | 10s             | 1134.20 (OP/s)
📋 perf-client-other-node-7867748554-vfvgt            | UDP_STREAM      | 1               | 10s             | 1425.74 (Mb/s)
📋 perf-client-5d7cb4d587-cn8sw                       | UDP_RR          | 1               | 10s             | 31737.62 (OP/s)
📋 perf-client-5d7cb4d587-cn8sw                       | UDP_STREAM      | 1               | 10s             | 865.24 (Mb/s)
📋 perf-client-other-node-7867748554-vfvgt            | TCP_CRR         | 1               | 10s             | 435.80 (OP/s)
-----------------------------------------------------------------------------------------------------------------------------

ClusterMesh

Install Cilium & enable ClusterMesh in Cluster 1

cilium install --helm-set=cluster.id=1
🔮 Auto-detected Kubernetes kind: GKE
ℹ️  Cilium version not set, using default version "v1.9.1"
🔮 Auto-detected cluster name: gke-cilium-dev-us-west2-a-tgraf-cluster1
🔮 Auto-detected datapath mode: gke
✅ Detected GKE native routing CIDR: 10.52.0.0/14
🚀 Creating resource quotas...
🔑 Found existing CA in secret cilium-ca
🔑 Generating certificates for Hubble...
🚀 Creating service accounts...
🚀 Creating cluster roles...
🚀 Creating ConfigMap...
🚀 Creating GKE Node Init DaemonSet...
🚀 Creating agent DaemonSet...
🚀 Creating operator Deployment...

cilium clustermesh enable
✨ Validating cluster configuration...
✅ Valid cluster identification found: name="gke-cilium-dev-us-west2-a-tgraf-cluster1" id="1"
🔑 Found existing CA in secret cilium-ca
🔑 Generating certificates for ClusterMesh...
✨ Deploying clustermesh-apiserver...
🔮 Auto-exposing service within GCP VPC (cloud.google.com/load-balancer-type=internal)

Install Cilium in Cluster 2

cilium install --context gke_cilium-dev_us-west2-a_tgraf-cluster2 --helm-set=cluster.id=2
🔮 Auto-detected Kubernetes kind: GKE
ℹ️  Cilium version not set, using default version "v1.9.1"
🔮 Auto-detected cluster name: gke-cilium-dev-us-west2-a-tgraf-cluster2
🔮 Auto-detected datapath mode: gke
✅ Detected GKE native routing CIDR: 10.4.0.0/14
🚀 Creating resource quotas...
🔑 Found existing CA in secret cilium-ca
🔑 Generating certificates for Hubble...
🚀 Creating service accounts...
🚀 Creating cluster roles...
🚀 Creating ConfigMap...
🚀 Creating GKE Node Init DaemonSet...
🚀 Creating agent DaemonSet...
🚀 Creating operator Deployment...

cilium clustermesh enable --context gke_cilium-dev_us-west2-a_tgraf-cluster2
✨ Validating cluster configuration...
✅ Valid cluster identification found: name="gke-cilium-dev-us-west2-a-tgraf-cluster2" id="2"
🔑 Found existing CA in secret cilium-ca
🔑 Generating certificates for ClusterMesh...
✨ Deploying clustermesh-apiserver...
🔮 Auto-exposing service within GCP VPC (cloud.google.com/load-balancer-type=internal)

Connect Clusters

cilium clustermesh connect --destination-context gke_cilium-dev_us-west2-a_tgraf-cluster2
✨ Extracting access information of cluster gke-cilium-dev-us-west2-a-tgraf-cluster2...
🔑 Extracting secrets from cluster gke-cilium-dev-us-west2-a-tgraf-cluster2...
ℹ️  Found ClusterMesh service IPs: [10.168.15.209]
✨ Extracting access information of cluster gke-cilium-dev-us-west2-a-tgraf-cluster1...
🔑 Extracting secrets from cluster gke-cilium-dev-us-west2-a-tgraf-cluster1...
ℹ️  Found ClusterMesh service IPs: [10.168.15.208]
✨ Connecting cluster gke_cilium-dev_us-west2-a_tgraf-cluster1 -> gke_cilium-dev_us-west2-a_tgraf-cluster2...
🔑 Patching existing secret cilium-clustermesh...
✨ Patching DaemonSet with IP aliases cilium-clustermesh...
✨ Connecting cluster gke_cilium-dev_us-west2-a_tgraf-cluster2 -> gke_cilium-dev_us-west2-a_tgraf-cluster1...
🔑 Patching existing secret cilium-clustermesh...
✨ Patching DaemonSet with IP aliases cilium-clustermesh...

Encryption

Install a Cilium in a cluster and enable encryption with IPsec

cilium install --encryption=ipsec
🔮 Auto-detected Kubernetes kind: kind
✨ Running "kind" validation checks
✅ Detected kind version "0.9.0"
ℹ️  Cilium version not set, using default version "v1.9.2"
🔮 Auto-detected cluster name: kind-chart-testing
🔮 Auto-detected IPAM mode: kubernetes
🔑 Found existing CA in secret cilium-ca
🔑 Generating certificates for Hubble...
🚀 Creating Service accounts...
🚀 Creating Cluster roles...
🔑 Generated encryption secret cilium-ipsec-keys
🚀 Creating ConfigMap...
🚀 Creating Agent DaemonSet...
🚀 Creating Operator Deployment...
⌛ Waiting for Cilium to be installed...

Experimental helm installation mode

cilium-cli v0.14 introduces a new helm installation mode. In the current installation mode (we now call it classic mode), cilium-cli directly calls Kubernetes APIs to manage resources related to Cilium. In the new helm mode, cilium-cli delegates all the installation state management to Helm. This enables you to use cilium-cli and helm interchangeably to manage your Cilium installation, while taking advantage of cilium-cli's advanced features such as Cilium configuration auto-detection.

Our current plan is to:

  • v0.14: Introduce the helm installation mode as an opt-in feature.
  • v0.15: Enable the helm installation mode by default, and deprecate the classic installation mode.
  • v1.0: Remove the classic installation mode.

Examples

To opt in to use the helm installation mode:

export CILIUM_CLI_MODE=helm

install examples

To install the default version of Cilium:

cilium install

To see the Helm release that got deployed:

helm list -n kube-system cilium

To see non-default Helm values that cilium-cli used for this Cilium installation:

helm get values -n kube-system cilium

To see all the Cilium-related resources without installing them to your cluster:

cilium install --dry-run

To see all the non-default Helm values without actually performing the installation:

cilium install --dry-run-helm-values

To install using Cilium's OCI dev chart repository:

cilium install --repository oci://quay.io/cilium-charts-dev/cilium --version 1.14.0-dev-dev.4-main-797347707c

upgrade examples

To upgrade to a specific version of Cilium:

cilium upgrade --version v1.13.2

To upgrade using a local Helm chart:

cilium upgrade --chart-directory ./install/kubernetes/cilium

To upgrade using Cilium's OCI dev chart repository:

cilium upgrade --repository oci://quay.io/cilium-charts-dev/cilium --version 1.14.0-dev-dev.4-main-797347707c

When upgrading Cilium, please refer to the upgrade guide and adjust non-default Helm values accordingly. cilium-cli does not automatically modify non-default Helm values during upgrade. You can use --dry-run and --dry-run-helm-values flags to review Kubernetes resources and non-default Helm values without actually performing upgrade:

To see the difference between the current Kubernetes resources in a live cluster and what would be applied:

cilium upgrade --version v1.13.2 --dry-run | kubectl diff -f -

To see the non-default Helm values that would be used during upgrade:

cilium upgrade --version v1.13.2 --dry-run-helm-values

Note You can use external diff tools such as dyff to make kubectl diff output more readable.

Supported commands as of v0.14

  • clustermesh
  • completion
  • config
  • connectivity
  • context
  • help
  • hubble
  • install
  • status
  • sysdump
  • uninstall
  • upgrade
  • version