[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6994CDM49JY2CQ9S6DREC3W":3,"skill-same:01M6994CDM49JY2CQ9S6DREC3W":76},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":47,"commitSha":48,"skillDirectory":49,"forksCount":50,"repositoryHtmlUrl":51,"repositoryHomepageUrl":6,"ownerHtmlUrl":52,"ownerBlogUrl":53,"ownerDescription":54,"updatedAt":55,"install":56,"resources":69},"01M6994CDM49JY2CQ9S6DREC3W","gke-node-notready",null,"Diagnoses GKE nodes reporting NotReady or Unknown status by inspecting node conditions, events, kubelet\u002Fcontainerd logs, and node metrics, then proposing safe remediations. Use when nodes show NotReady, when the kubelet stops posting node status, or when workloads are evicted or stuck Pending due to node health. Don't use for pod-level application failures (use gke-workload-troubleshooting), autoscaler scale-up\u002Fscale-down decisions (use gke-cluster-autoscaler), or non-GKE compute.","google\u002Fskills","google","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1342004?v=4","01M698VYEFDM0VJBZDYYJ2T83S","Google","01M698WR0H76ENHE587T9EJT89",24235,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgke-node-notready\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"gke node notready","gke","node","up","when","down","unknown","for","cluster","events","containerd","posting","show","safe","reporting","logs","metrics","workload","status","containers","2026-11-10 10:12:30",false,"---\nname: gke-node-notready\nmetadata:\n  version: \"1.0.1\"\n  category: Containers\ndescription: >-\n  Diagnoses GKE nodes reporting NotReady or Unknown status by inspecting node conditions, events, kubelet\u002Fcontainerd logs, and node metrics, then proposing safe remediations. Use when nodes show NotReady, when the kubelet stops posting node status, or when workloads are evicted or stuck Pending due to node health. Don't use for pod-level application failures (use gke-workload-troubleshooting), autoscaler scale-up\u002Fscale-down decisions (use gke-cluster-autoscaler), or non-GKE compute.\n---\n\n# GKE Node NotReady Troubleshooting Skill\n\nUse this skill to systematically diagnose why one or more GKE nodes report a\n`NotReady` (or `Ready: Unknown`) status and to propose safe remediations. A\n`NotReady` status means the node's kubelet is not reporting to the control plane\ncorrectly, so Kubernetes stops scheduling new Pods on the node, which can reduce\napplication capacity and cause downtime.\n\nThis skill operates **non-interactively** and enforces a **read-only diagnostics\nboundary**: gather evidence first, then propose a fix (a `kubectl`\u002F`gcloud`\ncommand or a GitOps manifest change) for a human to apply. **Never** mutate the\ncluster, drain, delete, or recreate nodes automatically.\n\n> [!IMPORTANT]\n> First rule out an **expected** `NotReady`: a node that is newly provisioning,\n> upgrading, being repaired, cordoned, or scaling down will transiently report\n> `NotReady`. Only treat it as a fault if it persists beyond the expected window.\n\n## Attribution\n\n**MUST READ**: Before executing any `gcloud`, REST API, or MCP command, you\nMUST read [Usage Attribution](references\u002Fattribution.md). For commands executed\ndirectly in the terminal (never in generated scripts, Terraform, or user code),\nprefix `gcloud` with\n`CLOUDSDK_METRICS_REQUEST_ATTRIBUTION=\"agent-skills\u002F1.0 (skill:gke-node-notready)\"`\nand append `agent-skills\u002F1.0 (skill:gke-node-notready)` to `User-Agent` headers.\n\n## ? Diagnostic Workflow\n\n### Step 0: Context discovery & time window\n\n1.  **Parameter extraction** — obtain `project_id`, `cluster_name`,\n    `cluster_location`, and `node_name` non-interactively from the user prompt,\n    active `SETTINGS.md`, or environment defaults (`kubectl config current-context`,\n    `gcloud config get-value project`).\n2.  **Credentials & fallback** — attempt\n    `gcloud container clusters get-credentials {cluster_name} --location {cluster_location} --project {project_id}`.\n    If the cluster is unreachable or commands fail (sandbox\u002Fdry-run\u002Foffline),\n    present the exact diagnostic commands for a human to run and continue the\n    analysis from the reported symptoms.\n3.  **Time window** — determine `{issue_time}` (explicit, relative, or now) and\n    center a 1-hour window around it (`start = issue_time - 30m`,\n    `end = issue_time + 30m`) for all log\u002Fmetric queries.\n\n--------------------------------------------------------------------------------\n\n### Step 1: Identify NotReady nodes and gather initial status\n\n```bash\n# List nodes and spot NotReady status, node IPs, and container-runtime version.\nkubectl get nodes -o wide\n\n# Inspect the affected node's Conditions and Events (the primary clues).\nkubectl describe node \"{node_name}\"\n```\n\nEquivalent via Cloud Logging (preferred when kubectl access is limited or for\nhistorical events). Open it as a **Logs Explorer deep link** — URL-encode the\nquery and append the project and Step 0 time window:\n`https:\u002F\u002Fconsole.cloud.google.com\u002Flogs\u002Fquery;query={URL_ENCODED_QUERY};timeRange={start}%2F{end}?project={project_id}`\n(encode `\u002F` as `%2F`, or use `;duration=PT1H` for a rolling hour):\n\n```\nresource.type=\"k8s_node\"\nlog_id(\"events\")\nresource.labels.node_name=\"{node_name}\"\nresource.labels.cluster_name=\"{cluster_name}\"\nresource.labels.location=\"{cluster_location}\"\n```\n\n**Interpret the `Conditions` table:**\n\n- `Ready: False` \u002F `Ready: Unknown` with reason `KubeletNotReady` \u002F\n  `NodeStatusUnknown` (\"Kubelet stopped posting node status\") → kubelet or\n  runtime problem; continue to Step 2.\n- `MemoryPressure: True`, `DiskPressure: True`, `PIDPressure: True` → resource\n  exhaustion; go to Step 4b.\n- `NetworkUnavailable: True` → networking\u002FCNI problem; go to Step 4d.\n\n--------------------------------------------------------------------------------\n\n### Step 2: Scan kubelet logs for error signatures\n\nOpen these kubelet logs as a **Logs Explorer deep link** using the same\n`logs\u002Fquery;query={URL_ENCODED_QUERY};timeRange=...?project=...` pattern as Step 1.\n\n```\nresource.type=\"k8s_node\"\nresource.labels.node_name=\"{node_name}\"\nresource.labels.cluster_name=\"{cluster_name}\"\nresource.labels.location=\"{cluster_location}\"\nlog_id(\"kubelet\")\nseverity>=WARNING\n```\n\nAlso review the node's serial-console logs (`log_id(\"serialconsole.googleapis.com\u002Fserial_port_1_output\")`\nor the `resource.type=\"gce_instance\"` serial logs) for kernel `TaskHung`,\nOOM-killer, or disk I\u002FO errors that correlate with the kubelet failures.\n\n--------------------------------------------------------------------------------\n\n### Step 3: Map the signature to a root cause (decision table)\n\n| Kubelet \u002F event signature | Likely root cause | Go to |\n| --- | --- | --- |\n| `runtime is down`, `Container runtime not ready`, errors on `\u002Frun\u002Fcontainerd\u002Fcontainerd.sock` (connection refused \u002F DeadlineExceeded) | Container runtime (`containerd`) down or unresponsive | Step 4a |\n| `Got sys oom event from cadvisor` \u002F kernel OOM-killer in serial logs | System (node-level) OOM killed critical processes | Step 4b |\n| `PLEG is not healthy` | PLEG stalled, usually node overload (CPU\u002Fdisk) | Step 4c |\n| `TaskHung` for `containerd`\u002F`kubelet`, high disk latency | Disk throttling \u002F I\u002FO starvation | Step 4b |\n| `failed to ensure lease`, `leases.coordination.k8s.io ... namespace kube-node-lease ... terminating` | `kube-node-lease` termination → NotReady flapping | Step 4f |\n| Kubelet cannot reach API server, TLS\u002Fdial timeouts | Kubelet ↔ control-plane connectivity | Step 4d |\n| `NetworkPluginNotReady`, `cni plugin not initialized`, `NetworkUnavailable` | CNI plugin failure | Step 4d |\n| Node-critical DaemonSet Pods (CNI, kube-proxy, metadata) blocked from admission | Admission webhook interference | Step 4e |\n| Only generic `NodeNotReady`, no other signature | Cause unclear — widen to Step 4d, then escalate | Escalation |\n\n--------------------------------------------------------------------------------\n\n### Step 4: Branch investigations\n\n#### 4a. Container runtime (`containerd`) down\n\nConfirm the kubelet cannot talk to containerd (socket errors above). Check for\n`containerd` restarts\u002Fcrashes in serial logs. **Remediation (propose, don't run):**\nrecreate\u002Frepair the node (`kubectl drain` then let the node pool recreate it, or\n`gcloud container clusters upgrade`\u002Fnode auto-repair); if it recurs across nodes,\nsuspect a node image or custom DaemonSet interfering with containerd.\n\n#### 4b. Resource pressure & OOM\n\n```bash\n# Node allocatable vs. usage.\nkubectl describe node \"{node_name}\" | sed -n '\u002FAllocated resources\u002F,\u002FEvents\u002Fp'\n```\nCloud Monitoring metrics to inspect (read-only): `kubernetes.io\u002Fnode\u002Fmemory\u002Fused_bytes`,\n`kubernetes.io\u002Fnode\u002Fcpu\u002Fcore_usage_time`, `kubernetes.io\u002Fnode\u002Fephemeral_storage\u002Fused_bytes`.\n- **DiskPressure \u002F disk throttling**: full boot disk or slow PD → increase disk\n  size \u002F use a faster PD type; reduce image\u002Flog churn.\n- **System OOM**: node memory exhausted → set\u002Fraise Pod memory `requests`\u002F`limits`,\n  reduce over-commit, or use larger machine types. Distinguish **system OOM**\n  (node-wide, kills kubelet\u002Fruntime) from **cgroup OOM** (single container).\n- **PIDPressure**: too many processes → cap Pod PIDs \u002F reduce workload density.\n\n#### 4c. PLEG is not healthy\n\n`PLEG is not healthy` almost always means the node is overloaded (CPU saturation,\ndisk latency, or too many Pods\u002Fcontainers per node) so the runtime can't relist\nin time. Correlate with 4b metrics. **Remediation:** reduce node density, add\nCPU\u002Fdisk headroom, or spread workloads.\n\n#### 4d. Networking\n\n```bash\n# Are node-critical networking Pods healthy on this node?\nkubectl get pods -n kube-system -o wide --field-selector spec.nodeName={node_name}\n```\n- **Kubelet ↔ control-plane**: dial\u002FTLS timeouts to the API server → check\n  firewall rules, Private Google Access, authorized networks, and route\u002FNAT\n  changes.\n- **CNI failure** (`NetworkPluginNotReady`): the CNI DaemonSet\n  (`netd`\u002F`calico`\u002Fdataplane) is not running on the node → inspect those Pods'\n  logs\u002Fevents.\n\n#### 4e. Admission webhook interference\n\nA misconfigured\u002Ffailing validating or mutating webhook with a broad scope can\nblock node-critical system Pods from being admitted, keeping the node NotReady.\n\n```bash\nkubectl get validatingwebhookconfigurations,mutatingwebhookconfigurations\n```\nLook for webhooks that intercept `kube-system` \u002F node-critical objects with\n`failurePolicy: Fail`. **Remediation (propose):** scope the webhook out of\n`kube-system`\u002Fnode-critical namespaces or set an appropriate `namespaceSelector`.\n\n#### 4f. `kube-node-lease` termination flapping\n\nIf the node flaps NotReady with `leases.coordination.k8s.io ... namespace\nkube-node-lease ... is being terminated`, the `kube-node-lease` namespace was\ndeleted\u002Fterminating. **Remediation (propose):** do not delete the\n`kube-node-lease` namespace; if terminating, identify the finalizer\u002Factor holding\nit and restore the namespace.\n\n--------------------------------------------------------------------------------\n\n### Step 5: Remediation boundary & escalation\n\n- Present the **root cause + evidence** (the exact conditions, events, log lines,\n  or metrics observed). Provide **Cloud Logging deep links** (and Cloud Monitoring\n  links for the Step 4b metrics) to the supporting entries — using the deep-link\n  pattern from Steps 1-2 — so a human can open the evidence directly.\n- Propose the fix as a **command or GitOps manifest change** for a human to apply\n  — never apply, drain, or recreate nodes automatically.\n**When to escalate (do this instead of proposing more self-service diagnostics):**\n\nEscalate when either:\n\n- the relevant logs are **unavailable** — excluded by a logging filter, or older\n  than the log bucket's retention (the `_Default` bucket defaults to 30 days, so\n  incidents older than that are permanently deleted); or\n- the kubelet\u002Fevent signature is **not in the Step 3 table** and the root cause\n  remains **undetermined** after the branch investigations.\n\nIn those cases, do all three:\n\n1.  **State the limitation plainly** (for example, \"kubelet logs for that date are\n    past the 30-day `_Default` retention window and are permanently deleted\").\n2.  **Summarize the findings you did gather** (node conditions, events, metrics,\n    and any Admin Activity audit logs still in the `_Required` bucket, default\n    400-day retention).\n3.  **Route to GKE support \u002F engineering escalation with those findings.** Do\n    **not** keep proposing further self-service investigation, and do **not**\n    fabricate a diagnosis when the evidence is missing.\n\n--------------------------------------------------------------------------------\n\n## References\n\nThis skill is derived from public Google Cloud documentation:\n\n- [Troubleshoot nodes with the NotReady status](https:\u002F\u002Fcloud.google.com\u002Fkubernetes-engine\u002Fdocs\u002Ftroubleshooting\u002Fnode-notready)\n  — node conditions and the kubelet \u002F PLEG \u002F system-OOM \u002F containerd \u002F\n  `kube-node-lease` \u002F CNI \u002F admission-webhook signatures and their remediations.\n- [Troubleshoot node registration](https:\u002F\u002Fcloud.google.com\u002Fkubernetes-engine\u002Fdocs\u002Ftroubleshooting\u002Fnode-registration)\n  — Node Registration Checker for nodes that never finish registering.\n- [View GKE logs](https:\u002F\u002Fcloud.google.com\u002Fkubernetes-engine\u002Fdocs\u002Fhow-to\u002Fview-logs)\n  and [Cloud Logging routing overview](https:\u002F\u002Fcloud.google.com\u002Flogging\u002Fdocs\u002Frouting\u002Foverview)\n  — log queries (`resource.type=\"k8s_node\"`, `log_id(\"kubelet\")`) and log-bucket\n  retention (`_Default` 30 days, `_Required` 400 days).\n- [Logs Explorer interface](https:\u002F\u002Fcloud.google.com\u002Flogging\u002Fdocs\u002Fview\u002Flogs-explorer-interface)\n  — building and sharing a query by URL (the `logs\u002Fquery;query=...` deep-link\n  format used above).\n","# GKE Node NotReady Troubleshooting Skill\n\nUse this skill to systematically diagnose why one or more GKE nodes report a\n`NotReady` (or `Ready: Unknown`) status and to propose safe remediations. A\n`NotReady` status means the node's kubelet is not reporting to the control plane\ncorrectly, so Kubernetes stops scheduling new Pods on the node, which can reduce\napplication capacity and cause downtime.\n\nThis skill operates **non-interactively** and enforces a **read-only diagnostics\nboundary**: gather evidence first, then propose a fix (a `kubectl`\u002F`gcloud`\ncommand or a GitOps manifest change) for a human to apply. **Never** mutate the\ncluster, drain, delete, or recreate nodes automatically.\n\n> [!IMPORTANT]\n> First rule out an **expected** `NotReady`: a node that is newly provisioning,\n> upgrading, being repaired, cordoned, or scaling down will transiently report\n> `NotReady`. Only treat it as a fault if it persists beyond the expected window.\n\n## Attribution\n\n**MUST READ**: Before executing any `gcloud`, REST API, or MCP command, you\nMUST read [Usage Attribution](references\u002Fattribution.md). For commands executed\ndirectly in the terminal (never in generated scripts, Terraform, or user code),\nprefix `gcloud` with\n`CLOUDSDK_METRICS_REQUEST_ATTRIBUTION=\"agent-skills\u002F1.0 (skill:gke-node-notready)\"`\nand append `agent-skills\u002F1.0 (skill:gke-node-notready)` to `User-Agent` headers.\n\n## ? Diagnostic Workflow\n\n### Step 0: Context discovery & time window\n\n1.  **Parameter extraction** — obtain `project_id`, `cluster_name`,\n    `cluster_location`, and `node_name` non-interactively from the user prompt,\n    active `SETTINGS.md`, or environment defaults (`kubectl config current-context`,\n    `gcloud config get-value project`).\n2.  **Credentials & fallback** — attempt\n    `gcloud container clusters get-credentials {cluster_name} --location {cluster_location} --project {project_id}`.\n    If the cluster is unreachable or commands fail (sandbox\u002Fdry-run\u002Foffline),\n    present the exact diagnostic commands for a human to run and continue the\n    analysis from the reported symptoms.\n3.  **Time window** — determine `{issue_time}` (explicit, relative, or now) and\n    center a 1-hour window around it (`start = issue_time - 30m`,\n    `end = issue_time + 30m`) for all log\u002Fmetric queries.\n\n--------------------------------------------------------------------------------\n\n### Step 1: Identify NotReady nodes and gather initial status\n\n```bash\n# List nodes and spot NotReady status, node IPs, and container-runtime version.\nkubectl get nodes -o wide\n\n# Inspect the affected node's Conditions and Events (the primary clues).\nkubectl describe node \"{node_name}\"\n```\n\nEquivalent via Cloud Logging (preferred when kubectl access is limited or for\nhistorical events). Open it as a **Logs Explorer deep link** — URL-encode the\nquery and append the project and Step 0 time window:\n`https:\u002F\u002Fconsole.cloud.google.com\u002Flogs\u002Fquery;query={URL_ENCODED_QUERY};timeRange={start}%2F{end}?project={project_id}`\n(encode `\u002F` as `%2F`, or use `;duration=PT1H` for a rolling hour):\n\n```\nresource.type=\"k8s_node\"\nlog_id(\"events\")\nresource.labels.node_name=\"{node_name}\"\nresource.labels.cluster_name=\"{cluster_name}\"\nresource.labels.location=\"{cluster_location}\"\n```\n\n**Interpret the `Conditions` table:**\n\n- `Ready: False` \u002F `Ready: Unknown` with reason `KubeletNotReady` \u002F\n  `NodeStatusUnknown` (\"Kubelet stopped posting node status\") → kubelet or\n  runtime problem; continue to Step 2.\n- `MemoryPressure: True`, `DiskPressure: True`, `PIDPressure: True` → resource\n  exhaustion; go to Step 4b.\n- `NetworkUnavailable: True` → networking\u002FCNI problem; go to Step 4d.\n\n--------------------------------------------------------------------------------\n\n### Step 2: Scan kubelet logs for error signatures\n\nOpen these kubelet logs as a **Logs Explorer deep link** using the same\n`logs\u002Fquery;query={URL_ENCODED_QUERY};timeRange=...?project=...` pattern as Step 1.\n\n```\nresource.type=\"k8s_node\"\nresource.labels.node_name=\"{node_name}\"\nresource.labels.cluster_name=\"{cluster_name}\"\nresource.labels.location=\"{cluster_location}\"\nlog_id(\"kubelet\")\nseverity>=WARNING\n```\n\nAlso review the node's serial-console logs (`log_id(\"serialconsole.googleapis.com\u002Fserial_port_1_output\")`\nor the `resource.type=\"gce_instance\"` serial logs) for kernel `TaskHung`,\nOOM-killer, or disk I\u002FO errors that correlate with the kubelet failures.\n\n--------------------------------------------------------------------------------\n\n### Step 3: Map the signature to a root cause (decision table)\n\n| Kubelet \u002F event signature | Likely root cause | Go to |\n| --- | --- | --- |\n| `runtime is down`, `Container runtime not ready`, errors on `\u002Frun\u002Fcontainerd\u002Fcontainerd.sock` (connection refused \u002F DeadlineExceeded) | Container runtime (`containerd`) down or unresponsive | Step 4a |\n| `Got sys oom event from cadvisor` \u002F kernel OOM-killer in serial logs | System (node-level) OOM killed critical processes | Step 4b |\n| `PLEG is not healthy` | PLEG stalled, usually node overload (CPU\u002Fdisk) | Step 4c |\n| `TaskHung` for `containerd`\u002F`kubelet`, high disk latency | Disk throttling \u002F I\u002FO starvation | Step 4b |\n| `failed to ensure lease`, `leases.coordination.k8s.io ... namespace kube-node-lease ... terminating` | `kube-node-lease` termination → NotReady flapping | Step 4f |\n| Kubelet cannot reach API server, TLS\u002Fdial timeouts | Kubelet ↔ control-plane connectivity | Step 4d |\n| `NetworkPluginNotReady`, `cni plugin not initialized`, `NetworkUnavailable` | CNI plugin failure | Step 4d |\n| Node-critical DaemonSet Pods (CNI, kube-proxy, metadata) blocked from admission | Admission webhook interference | Step 4e |\n| Only generic `NodeNotReady`, no other signature | Cause unclear — widen to Step 4d, then escalate | Escalation |\n\n--------------------------------------------------------------------------------\n\n### Step 4: Branch investigations\n\n#### 4a. Container runtime (`containerd`) down\n\nConfirm the kubelet cannot talk to containerd (socket errors above). Check for\n`containerd` restarts\u002Fcrashes in serial logs. **Remediation (propose, don't run):**\nrecreate\u002Frepair the node (`kubectl drain` then let the node pool recreate it, or\n`gcloud container clusters upgrade`\u002Fnode auto-repair); if it recurs across nodes,\nsuspect a node image or custom DaemonSet interfering with containerd.\n\n#### 4b. Resource pressure & OOM\n\n```bash\n# Node allocatable vs. usage.\nkubectl describe node \"{node_name}\" | sed -n '\u002FAllocated resources\u002F,\u002FEvents\u002Fp'\n```\nCloud Monitoring metrics to inspect (read-only): `kubernetes.io\u002Fnode\u002Fmemory\u002Fused_bytes`,\n`kubernetes.io\u002Fnode\u002Fcpu\u002Fcore_usage_time`, `kubernetes.io\u002Fnode\u002Fephemeral_storage\u002Fused_bytes`.\n- **DiskPressure \u002F disk throttling**: full boot disk or slow PD → increase disk\n  size \u002F use a faster PD type; reduce image\u002Flog churn.\n- **System OOM**: node memory exhausted → set\u002Fraise Pod memory `requests`\u002F`limits`,\n  reduce over-commit, or use larger machine types. Distinguish **system OOM**\n  (node-wide, kills kubelet\u002Fruntime) from **cgroup OOM** (single container).\n- **PIDPressure**: too many processes → cap Pod PIDs \u002F reduce workload density.\n\n#### 4c. PLEG is not healthy\n\n`PLEG is not healthy` almost always means the node is overloaded (CPU saturation,\ndisk latency, or too many Pods\u002Fcontainers per node) so the runtime can't relist\nin time. Correlate with 4b metrics. **Remediation:** reduce node density, add\nCPU\u002Fdisk headroom, or spread workloads.\n\n#### 4d. Networking\n\n```bash\n# Are node-critical networking Pods healthy on this node?\nkubectl get pods -n kube-system -o wide --field-selector spec.nodeName={node_name}\n```\n- **Kubelet ↔ control-plane**: dial\u002FTLS timeouts to the API server → check\n  firewall rules, Private Google Access, authorized networks, and route\u002FNAT\n  changes.\n- **CNI failure** (`NetworkPluginNotReady`): the CNI DaemonSet\n  (`netd`\u002F`calico`\u002Fdataplane) is not running on the node → inspect those Pods'\n  logs\u002Fevents.\n\n#### 4e. Admission webhook interference\n\nA misconfigured\u002Ffailing validating or mutating webhook with a broad scope can\nblock node-critical system Pods from being admitted, keeping the node NotReady.\n\n```bash\nkubectl get validatingwebhookconfigurations,mutatingwebhookconfigurations\n```\nLook for webhooks that intercept `kube-system` \u002F node-critical objects with\n`failurePolicy: Fail`. **Remediation (propose):** scope the webhook out of\n`kube-system`\u002Fnode-critical namespaces or set an appropriate `namespaceSelector`.\n\n#### 4f. `kube-node-lease` termination flapping\n\nIf the node flaps NotReady with `leases.coordination.k8s.io ... namespace\nkube-node-lease ... is being terminated`, the `kube-node-lease` namespace was\ndeleted\u002Fterminating. **Remediation (propose):** do not delete the\n`kube-node-lease` namespace; if terminating, identify the finalizer\u002Factor holding\nit and restore the namespace.\n\n--------------------------------------------------------------------------------\n\n### Step 5: Remediation boundary & escalation\n\n- Present the **root cause + evidence** (the exact conditions, events, log lines,\n  or metrics observed). Provide **Cloud Logging deep links** (and Cloud Monitoring\n  links for the Step 4b metrics) to the supporting entries — using the deep-link\n  pattern from Steps 1-2 — so a human can open the evidence directly.\n- Propose the fix as a **command or GitOps manifest change** for a human to apply\n  — never apply, drain, or recreate nodes automatically.\n**When to escalate (do this instead of proposing more self-service diagnostics):**\n\nEscalate when either:\n\n- the relevant logs are **unavailable** — excluded by a logging filter, or older\n  than the log bucket's retention (the `_Default` bucket defaults to 30 days, so\n  incidents older than that are permanently deleted); or\n- the kubelet\u002Fevent signature is **not in the Step 3 table** and the root cause\n  remains **undetermined** after the branch investigations.\n\nIn those cases, do all three:\n\n1.  **State the limitation plainly** (for example, \"kubelet logs for that date are\n    past the 30-day `_Default` retention window and are permanently deleted\").\n2.  **Summarize the findings you did gather** (node conditions, events, metrics,\n    and any Admin Activity audit logs still in the `_Required` bucket, default\n    400-day retention).\n3.  **Route to GKE support \u002F engineering escalation with those findings.** Do\n    **not** keep proposing further self-service investigation, and do **not**\n    fabricate a diagnosis when the evidence is missing.\n\n--------------------------------------------------------------------------------\n\n## References\n\nThis skill is derived from public Google Cloud documentation:\n\n- [Troubleshoot nodes with the NotReady status](https:\u002F\u002Fcloud.google.com\u002Fkubernetes-engine\u002Fdocs\u002Ftroubleshooting\u002Fnode-notready)\n  — node conditions and the kubelet \u002F PLEG \u002F system-OOM \u002F containerd \u002F\n  `kube-node-lease` \u002F CNI \u002F admission-webhook signatures and their remediations.\n- [Troubleshoot node registration](https:\u002F\u002Fcloud.google.com\u002Fkubernetes-engine\u002Fdocs\u002Ftroubleshooting\u002Fnode-registration)\n  — Node Registration Checker for nodes that never finish registering.\n- [View GKE logs](https:\u002F\u002Fcloud.google.com\u002Fkubernetes-engine\u002Fdocs\u002Fhow-to\u002Fview-logs)\n  and [Cloud Logging routing overview](https:\u002F\u002Fcloud.google.com\u002Flogging\u002Fdocs\u002Frouting\u002Foverview)\n  — log queries (`resource.type=\"k8s_node\"`, `log_id(\"kubelet\")`) and log-bucket\n  retention (`_Default` 30 days, `_Required` 400 days).\n- [Logs Explorer interface](https:\u002F\u002Fcloud.google.com\u002Flogging\u002Fdocs\u002Fview\u002Flogs-explorer-interface)\n  — building and sharing a query by URL (the `logs\u002Fquery;query=...` deep-link\n  format used above).\n",{"name":5,"metadata":44,"description":7},{"version":45,"category":46},"1.0.1","Containers","main","bb9b3f464aae37af360d079c1d863524955b0363","skills\u002Fcloud\u002Fgke-node-notready",2014,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills","https:\u002F\u002Fgithub.com\u002Fgoogle","https:\u002F\u002Fopensource.google\u002F","Google ❤️ Open Source","2026-11-14 23:31:00",{"source":8,"skillName":5,"commands":57,"prompt":67,"zipAvailable":68},[58,61,64],{"tool":59,"command":60},"npx","npx skills add google\u002Fskills --skill gke-node-notready",{"tool":62,"command":63},"bunx","bunx skills add google\u002Fskills --skill gke-node-notready",{"tool":65,"command":66},"pnpm","pnpm dlx skills add google\u002Fskills --skill gke-node-notready","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgke-node-notready\u002FSKILL.md",true,[70],{"path":71,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Fattribution.md","REFERENCE","参考文档",1891,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgke-node-notready\u002Freferences\u002Fattribution.md",[77,85,92,99,106,113],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":84},"01M7DQ9K7JF4T7NJZ2K1MQZ44Q","cloud-operator-share-context","Shares Google Cloud operating context the user or their team already have — environments, projects, services and regions, product preferences, best practices, architecture guidance, runbooks — found in agent instruction files, local memory, skills and workflows, and the repository's GCP docs, with the operator agent's memory after the user reviews the list. Use when the user runs this skill, accepts the cloud-operator or cloud-operator-setup skill's offer to share their existing Google Cloud context, or asks to share, import or save that context or a GCP fact for the operator agent. Don't use for ordinary GCP requests, which go to the cloud-operator skill.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fcloud-operator-share-context\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":84},"01M7DQ9ERFGGQCFAWQW0KRKYBY","cloud-operator-setup","Sets up the operator agent in a Google Cloud project via the autocloud MCP server: picks the operator project, creates or checks the user's operator agent, saves the project, and offers to share the user's existing Google Cloud operating context with the operator agent's memory. Use when the user runs this skill, asks to set up the operator or create their operator agent, or chooses setup when the cloud-operator skill asks. Don't use for ordinary GCP requests or installing gcloud.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fcloud-operator-setup\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":84},"01M7B4RHQCM0DK2XQ7S7ZXVN8D","developer-device-platform-device-run","Runs automated, non-interactive jobs on remote Android devices with Developer Device Platform (DDP) Device Run using `gcloud beta device-run sessions`: native Android executables or shell scripts (for example CLI tools, custom runners, test harnesses or benchmarks) and Android instrumentation tests, on one or many devices, then waits for completion and collects per-device outputs and logs. Use when running a binary or test suite on remote devices without an interactive connection. Don't use it for interactive device reservations, adb tunnels or screen streaming (use developer-device-platform-basics), or for local devices.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloper-device-platform-device-run\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":84},"01M7B4R9YXMRW7RZP9Q9FMZJ2P","compute-vm-creation","Guides the planning, configuration, and provisioning of standalone Google Compute Engine virtual machines (VMs) and bare metal (-metal) instances using gcloud CLI commands or direct MCP tool execution. Use ONLY when the user wants to create, configure, or provision a new standalone Compute Engine VM or bare metal instance from conversational intents (hardware sizing, region selection, OS image, networking, security defaults). DO NOT read or use for Google Kubernetes Engine (GKE) Autopilot or Standard clusters, Kubernetes node pools or Pods, container orchestration, managing existing VMs, Managed Instance Groups (MIGs), or Instance Templates.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fcompute-vm-creation\u002FSKILL.md",[],[],{"skillNo":107,"name":108,"nameZh":6,"description":109,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":84},"01M7B4QD02N0YBVT0838Z5K5VC","agent-platform-autorater-customization","Customizes an autorater (evaluation metric) for an AI agent so its grades match the user's own judgment, starting from the agent's real trajectories. Use when a user wants a metric tailored to their agent, wants to turn failing trajectories into rubrics, says an existing autorater grades their agent wrong, or wants an autorater to agree with human labels (autorater hillclimbing, judge calibration). Don't use for quick evals with built-in metrics or for improving the agent itself (use agent-platform-eval-flywheel).","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fagent-platform-autorater-customization\u002FSKILL.md",[],[],{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":84},"01M78JP22K73KK96JH0JNDT5C4","iap-agent-platform-troubleshooting","Diagnoses Identity-Aware Proxy (IAP) denials of outbound requests from Gemini Enterprise Agent Platform agents that run on Agent Runtime and send traffic through Agent Gateway. Use when an agent request fails with \"Egress request is not authorized\", when Cloud Audit Logs show an IAP egressViaIAP decision with granted false, when IAP audit entries are missing while debugging agent egress, or when Access policies, IAP allow policies, policy bindings, or Agent Registry entries might block agent egress. Don't use for Gemini model IAM permission errors (such as aiplatform.endpoints.predict), Agent Platform API errors, or Agent Runtime deployment or 500 errors (use agent-platform-troubleshooting instead). Don't use for setting up, creating, or connecting an Agent Gateway, or for IAP web apps, IAP TCP forwarding, or SSH.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fiap-agent-platform-troubleshooting\u002FSKILL.md",[],[]]