Talos Commands

Common talosctl commands for debugging the Talos Linux cluster. Cluster lifecycle (apply, upgrade, reset) is handled by topf (mise exec -- topf ... from talos/pitower); talosctl complements it for read-only inspection and low-level operations.

Talosconfig

Generate the talosconfig from the secrets bundle and merge it into ~/.talos/config:

bash

cd talos/pitower
just talosconfig && just merge-config
talosctl config endpoint 10.20.10.1 10.20.10.2 10.20.10.3
talosctl config node 10.20.10.1

Health Checks

Cluster Health

Check the overall health of the cluster, including etcd, kubelet, and API server status.

bash

talosctl health

Target a specific node:

bash

talosctl health --nodes 10.20.10.1

Node Dashboard

Open an interactive dashboard showing real-time CPU, memory, and service status for a node.

bash

talosctl dashboard

bash

talosctl dashboard --nodes 10.20.10.4

Member List

List all cluster members as seen by Talos.

bash

talosctl get members

JSON output (useful for scripting):

bash

talosctl get members -o json | jq '.spec'

Service Management

List Services

Show the status of all Talos services on a node.

bash

talosctl services --nodes 10.20.10.1

Service Logs

View logs for a specific Talos service (e.g., etcd, kubelet, apid, containerd).

bash

# etcd logs
talosctl logs etcd --nodes 10.20.10.1

# kubelet logs
talosctl logs kubelet --nodes 10.20.10.1

# Follow logs in real time
talosctl logs etcd --nodes 10.20.10.1 -f

Service Status

Get detailed status for a specific service.

bash

talosctl service etcd --nodes 10.20.10.1

etcd Operations

Membership

List etcd cluster members and their status.

bash

talosctl etcd members --nodes 10.20.10.1

Etcd Status

Check etcd health and leadership.

bash

talosctl etcd status --nodes 10.20.10.1

Remove a Member

Remove a failed etcd member (use with caution).

bash

talosctl etcd remove-member <member-id> --nodes 10.20.10.1

Etcd Snapshot

Take a snapshot of the etcd database.

bash

talosctl etcd snapshot etcd-backup.snapshot --nodes 10.20.10.1

Kubeconfig

Refresh Kubeconfig

Generate a kubeconfig for kubectl access. mise.toml points KUBECONFIG at ~/.kube/pitower.yaml; do not write to ~/.kube/config, which is a different cluster.

bash

talosctl kubeconfig ~/.kube/pitower.yaml --nodes 10.20.10.1

just kubeconfig (topf) writes a short-lived (12h) admin kubeconfig to output/kubeconfig instead.

Configuration Inspection

View Running Config

Inspect the current machine configuration running on a node.

bash

talosctl get machineconfig --nodes 10.20.10.1 -o yaml

Compare Configs

Diff the running config against the declared state (redacted; exit 2 = changes pending):

bash

mise exec -- topf apply --dry-run                              # all nodes
mise exec -- topf apply --dry-run --nodes-filter '^worker-01$'  # one node

topf render writes the full machine configs to output/ (gitignored) for inspection.

Version Information

Check the Talos version on a node.

bash

talosctl version --nodes 10.20.10.1

Resource Inspection

Kernel Logs (dmesg)

View kernel messages from a node.

bash

talosctl dmesg --nodes 10.20.10.4

Follow kernel messages:

bash

talosctl dmesg --nodes 10.20.10.4 -f

Process List

List running processes on a node.

bash

talosctl processes --nodes 10.20.10.1

Disk Usage

Check disk usage on a node.

bash

talosctl usage /var --nodes 10.20.10.1

Container Images

List all container images on a node.

bash

talosctl image list --nodes 10.20.10.1

Network Interfaces

Show network interfaces and addresses.

bash

talosctl get addresses --nodes 10.20.10.1

Routes

Show routing table.

bash

talosctl get routes --nodes 10.20.10.1

Node Operations

Reboot

Reboot a node (with wait for it to come back).

bash

talosctl reboot --nodes 10.20.10.4 --wait

Shutdown

Shut down a node.

bash

talosctl shutdown --nodes 10.20.10.4

Reset and Upgrade

Use topf for both, so the node ends up matching topf.yaml:

bash

cd talos/pitower
mise exec -- topf reset --nodes-filter '^worker-04$' --full=false   # or: just reset worker-04
mise exec -- topf upgrade --nodes-filter '^worker-04$'

See Upgrades.

Troubleshooting Commands

Check Pod CIDR and Service CIDR

bash

talosctl get machineconfig --nodes 10.20.10.1 -o yaml | grep -A3 -E 'podSubnets|serviceSubnets'
kubectl get nodes -o custom-columns=NAME:.metadata.name,PODCIDRS:.spec.podCIDRs

Check Certificate SANs

bash

talosctl get certsan --nodes 10.20.10.1 -o yaml

Read Machine Config Patches

bash

talosctl get machineconfig --nodes 10.20.10.1 -o yaml

Check Time Sync

bash

talosctl get timestatus --nodes 10.20.10.1

Useful Aliases

Consider adding these aliases to your shell configuration:

bash

alias tc='talosctl'
alias tcd='talosctl dashboard'
alias tch='talosctl health'
alias tcl='talosctl logs'
alias tcs='talosctl services'
alias tcm='talosctl get members'