Huawei Cloud CCE Metric Analyzer

Overview

Query and analyze metrics for CCE clusters (Pod/Node CPU/memory/disk) and cloud resources (ECS, ELB, EIP, NAT). Supports threshold-based anomaly detection, status classification (critical/warning/normal), and full-cluster monitoring aggregation.

Architecture: python3 scripts/huawei-cloud.py dispatcher → hcloud (KooCLI) cloud service queries + signed AOM Prometheus HTTP queries + limited kubectl reads only when Kubernetes resource relationships are required → Pod/Node metrics, ECS/ELB/EIP/NAT metrics → Threshold classification → Anomaly detection

Execution method: Cloud service queries are executed through the local hcloud CLI. AOM Prometheus query_range calls are the only exception and use signed HTTPS requests because the required Prometheus range-query path is not compatible with hcloud. Do not call Huawei Cloud SDKs, curl IAM flows, openstack, or hand-written cloud APIs outside the bundled dispatcher.

Related Skills: use pod/node diagnosers, Kubernetes event analyzer, capacity/cost skills, or auto-remediation runner for follow-up diagnosis or explicitly requested remediation.

Capabilities:

Pod CPU/memory TopN ranking and single Pod time-series metrics
Node CPU/memory/disk TopN ranking and single Node time-series metrics
Node GPU and xGPU metrics, including GPU utilization, memory, temperature, power, schedule policy, xGPU allocation, usage, and health
CoreDNS QPS, error rate excluding NXDOMAIN, NXDOMAIN rate, P95 latency, replica count, and per-Pod CPU/memory metrics
nginx-ingress QPS, 4xx/5xx rate, success rate, P95 latency, active connections, per-Pod CPU/memory, and Ingress TLS certificate expiration status
Autoscaler unschedulable Pods, node state count, scale-up/down events, errors, node groups, HPA current/desired replicas, and per-Pod CPU/memory metrics
Kubernetes control-plane metrics for apiserver, etcd, controller-manager, and scheduler
ECS instance CPU/memory/disk/network metrics
ELB connection, bandwidth, QPS metrics
EIP bandwidth, traffic, packet loss metrics
NAT Gateway SNAT connection metrics
Full-cluster monitoring aggregation with anomaly detection (80% threshold)
Threshold-based status classification (critical/warning/normal/unknown)

Typical Use Cases: query Pod/Node TopN, GPU/xGPU, CoreDNS, nginx-ingress, autoscaler, control-plane, ECS/ELB/EIP/NAT metrics, full-cluster aggregation, and threshold-based anomaly detection.

Prerequisites

1. Runtime Dependencies

Python 3.8+ for the dispatcher and result processing
hcloud (KooCLI) 7.2.2+ for CCE/ECS/ELB/VPC/EIP/NAT/CES/IAM cloud service queries
kubectl only for Kubernetes resource reads that cannot be derived from AOM/hcloud, such as Pod label_selector filtering, Ingress TLS certificate checks, and LoadBalancer Service discovery for ELB/EIP association; clusters without external EIP require the kubectl cce plugin from kubectl-cce-plugin/README.md
Prometheus-related monitoring data is queried from AOM Prometheus with signed HTTPS requests; the cluster must have the Prometheus add-on integrated with AOM, otherwise these tools may return empty metric series
Controller-manager, scheduler, and etcd metrics require the kube-controller-manager, kube-scheduler, and etcd-server ServiceMonitors to be enabled separately in AOM; otherwise these tools may return empty metric series
Autoscaler, ingress-controller, and NVIDIA GPU metrics require the corresponding autoscaler, ingress-controller, and nvidia-gpu-device-plugin PodMonitors to be enabled separately in AOM; ingress request metrics also require nginx_ingress_controller_requests to be explicitly allowed in the ingress-controller PodMonitor
Run environment check before first use (see Verification section)
Install and use kubectl-cce according to references/kubectl-cce.md

2. Credential Configuration

Valid Huawei Cloud credentials via hcloud profile or AK/SK mode
Security Rules:
- 🚫 Never expose AK/SK values in code, conversation, or commands
- 🚫 Never use echo $HUAWEI_AK or echo $HUAWEI_SK to check credentials
- ✅ Credential priority for hcloud calls is: explicit tool parameters > local hcloud profile > environment variables
- ✅ AOM Prometheus signed HTTP and Kubernetes certificate setup cannot use encrypted hcloud profile material, so they use explicit tool parameters first and environment variables as the signing fallback
- ✅ Prefer IAM users over root account for cloud operations
- ✅ Enable MFA for sensitive operations

Configuration Method:

hcloud configure list

export HUAWEI_AK=<your-ak>
export HUAWEI_SK=<your-sk>
export HUAWEI_REGION=cn-north-4

3. IAM Permission Requirements

API Action	Permission	Purpose
`cce:cluster:get`	Get cluster	View CCE cluster details
`aom:instance:list`	List AOM instances	Discover AOM Prometheus instance for metrics
`aom:metricsData:get`	Get metrics data	Query Pod/Node CPU/memory/disk metrics
`ces:metricsData:get`	Get CES metrics	Query ECS/ELB/EIP/NAT cloud resource metrics
`ecs:cloudServers:list`	List ECS servers	Correlate ECS instance IDs
`elb:loadbalancers:list`	List ELB instances	Correlate ELB IDs
`vpc:eips:list`	List EIPs	Correlate EIP IDs
`nat:natGateways:list`	List NAT Gateways	Correlate NAT Gateway IDs

Permission Failure Handling:

When any command fails due to IAM permission errors, display the required permission list
Guide the user to create a custom policy in the IAM console and grant authorization
Pause execution and wait for user confirmation that permissions have been granted

Core Commands

All commands use the Python dispatcher script: python3 scripts/huawei-cloud.py <action> <key=value>...

KooCLI命令格式标准

Do not ask users to run raw hcloud commands directly. Use the dispatcher format:

python3 scripts/huawei-cloud.py <tool-name> key=value key=value

The dispatcher converts cloud service queries to KooCLI calls. AOM Prometheus range queries use signed HTTPS requests because that path is not compatible with hcloud. Avoid Kubernetes resource reads unless the tool explicitly needs Pod labels, Ingress TLS Secrets, or LoadBalancer Services. Quote values containing spaces, >, <, |, JSON, or PromQL; never print or persist AK/SK, security tokens, kubeconfig files, or temporary payloads; keep Kubernetes/AOM PromQL scoped with cluster="<cluster_id>".

1. CCE Pod Metrics

# Pod TopN — cluster-wide CPU/memory ranking
python3 scripts/huawei-cloud.py huawei_get_cce_pod_metrics_topN \
  region=cn-north-4 cluster_id=<cluster-id> \
  namespace=default top_n=10 hours=1

# Pod TopN with label selector
python3 scripts/huawei-cloud.py huawei_get_cce_pod_metrics_topN \
  region=cn-north-4 cluster_id=<cluster-id> \
  namespace=default label_selector="app=nginx,version=v1" top_n=10 hours=1

# Single Pod time-series
python3 scripts/huawei-cloud.py huawei_get_cce_pod_metrics \
  region=cn-north-4 cluster_id=<cluster-id> \
  pod_name=my-app-xxx namespace=default hours=1

# Single Pod GPU and xGPU metrics
python3 scripts/huawei-cloud.py huawei_get_cce_pod_gpu_metrics \
  region=cn-north-4 cluster_id=<cluster-id> \
  pod_name=my-gpu-app-xxx namespace=default hours=1

2. CCE Node Metrics

# Node TopN — cluster-wide CPU/memory/disk ranking
python3 scripts/huawei-cloud.py huawei_get_cce_node_metrics_topN \
  region=cn-north-4 cluster_id=<cluster-id> \
  top_n=10 hours=1

# Single Node time-series
python3 scripts/huawei-cloud.py huawei_get_cce_node_metrics \
  region=cn-north-4 cluster_id=<cluster-id> \
  node_ip=10.0.0.1 hours=1

# Node GPU and xGPU metrics
python3 scripts/huawei-cloud.py huawei_get_cce_node_gpu_metrics \
  region=cn-north-4 cluster_id=<cluster-id> \
  node_ip=10.0.0.1 hours=1

3. CCE CoreDNS Metrics

# CoreDNS key metrics: QPS, error rate excluding NXDOMAIN, NXDOMAIN rate, P95 latency, replicas, CPU, and memory
python3 scripts/huawei-cloud.py huawei_get_cce_coredns_metrics \
  region=cn-north-4 cluster_id=<cluster-id> \
  namespace=kube-system pod_regex=".*coredns.*" hours=1

4. CCE nginx-ingress Metrics

# nginx-ingress request processing and Ingress TLS certificate expiration
python3 scripts/huawei-cloud.py huawei_get_cce_nginx_ingress_metrics \
  region=cn-north-4 cluster_id=<cluster-id> \
  namespace=kube-system pod_regex=".*nginx.*ingress.*|.*ingress.*nginx.*" \
  ingress_namespace=default cert_expire_warning_days=30 hours=1

5. CCE Autoscaler Metrics

# Cluster Autoscaler and HPA metrics
python3 scripts/huawei-cloud.py huawei_get_cce_autoscaler_metrics \
  region=cn-north-4 cluster_id=<cluster-id> \
  namespace=kube-system pod_regex=".*cluster.*autoscaler.*|.*autoscaler.*" \
  include_hpa=true hours=1

6. Kubernetes Control Plane Metrics

python3 scripts/huawei-cloud.py huawei_get_cce_apiserver_metrics \
  region=cn-north-4 cluster_id=<cluster-id> hours=1

python3 scripts/huawei-cloud.py huawei_get_cce_etcd_metrics \
  region=cn-north-4 cluster_id=<cluster-id> hours=1

python3 scripts/huawei-cloud.py huawei_get_cce_controller_manager_metrics \
  region=cn-north-4 cluster_id=<cluster-id> namespace=kube-system hours=1

python3 scripts/huawei-cloud.py huawei_get_cce_scheduler_metrics \
  region=cn-north-4 cluster_id=<cluster-id> namespace=kube-system hours=1

7. Cloud Resource Metrics

# ECS instance metrics
python3 scripts/huawei-cloud.py huawei_get_ecs_metrics \
  region=cn-north-4 instance_id=<instance-id>

# ELB metrics
python3 scripts/huawei-cloud.py huawei_get_elb_metrics \
  region=cn-north-4 elb_id=<loadbalancer-id> hours=1

# EIP metrics
python3 scripts/huawei-cloud.py huawei_get_eip_metrics \
  region=cn-north-4 eip_id=<eip-id> hours=1

# NAT Gateway metrics
python3 scripts/huawei-cloud.py huawei_get_nat_gateway_metrics \
  region=cn-north-4 nat_gateway_id=<nat-gateway-id> hours=1

8. Cluster Monitoring Aggregation

# Aggregate all monitoring data with anomaly detection
python3 scripts/huawei-cloud.py huawei_cce_cluster_monitoring_aggregation \
  region=cn-north-4 cluster_id=<cluster-id> \
  start_time="2026-05-30 00:00:00" end_time="2026-05-30 23:59:59" \
  namespace=default top_n=10

This tool aggregates: Pod TopN CPU/memory, Node TopN CPU/memory/disk, ELB metrics (matched through LoadBalancer Services fetched by kubectl), NAT Gateway metrics, EIP metrics (bandwidth, packet loss), and anomaly detection using 80% threshold.

It also includes CoreDNS, nginx-ingress, and autoscaler summaries. Cloud resources are scoped to the current cluster when an association can be proven: ELB is matched through LoadBalancer Service IP/EIP, NAT Gateway is filtered by the cluster VPC, and EIP is limited to associated ELB/NAT/Service IPs.

LoadBalancer Service discovery uses kubectl with generated kubeconfig through the cluster EIP when external access is available. If the cluster has no EIP, it uses the kubectl cce plugin. If neither path works, aggregation fails.

Risk Levels

This skill is read-only. It does not create, update, delete, restart, scale, or modify Huawei Cloud or Kubernetes resources.

Level	Meaning	Execution Guidance
R3	No-risk read-only query or local analysis	May run automatically
R2	Low-risk change, such as creating monitoring configuration without deleting resources or increasing service capacity/cost	Not used by current tools
R1	Risky operation, such as restart-like impact, disabling protection, or changes that may increase cost or reduce observability	Not used by current tools
R0	Critical operation, such as deleting clusters, applications, or broad-impact monitoring protections	Not used by current tools

Tool	Operation Type	Risk Level	Description
`huawei_get_cce_pod_metrics_topN`	Query	R3	Read Pod CPU/memory/disk TopN metrics from AOM Prometheus
`huawei_get_cce_pod_metrics`	Query	R3	Read single Pod CPU/memory/disk time-series metrics
`huawei_get_cce_node_metrics_topN`	Query	R3	Read Node CPU/memory/disk TopN metrics from AOM Prometheus
`huawei_get_cce_node_metrics`	Query	R3	Read single Node CPU/memory/disk time-series metrics
`huawei_get_cce_node_gpu_metrics`	Query	R3	Read single Node GPU and xGPU metrics from AOM Prometheus
`huawei_get_cce_pod_gpu_metrics`	Query	R3	Read single Pod GPU and xGPU metrics from AOM Prometheus
`huawei_get_cce_coredns_metrics`	Query	R3	Read CoreDNS QPS, error rate excluding NXDOMAIN, NXDOMAIN rate, P95 latency, replicas, and per-Pod CPU/memory metrics
`huawei_get_cce_nginx_ingress_metrics`	Query	R3	Read nginx-ingress request-processing metrics and Ingress TLS certificate expiration status; QPS falls back to nginx process request counters when request-dimension metrics are absent
`huawei_get_cce_autoscaler_metrics`	Query	R3	Read Cluster Autoscaler scaling metrics, HPA replica state, and autoscaler Pod CPU/memory metrics
`huawei_get_cce_apiserver_metrics`	Query	R3	Read kube-apiserver QPS, error rate, latency, and inflight request metrics
`huawei_get_cce_etcd_metrics`	Query	R3	Read etcd leader, proposal, DB size, disk latency, CPU, and memory metrics
`huawei_get_cce_controller_manager_metrics`	Query	R3	Read control-plane workqueue depth, adds, retries, queue latency, and work duration metrics
`huawei_get_cce_scheduler_metrics`	Query	R3	Read scheduler attempts, pending Pods, scheduling latency, and queue metrics
`huawei_get_ecs_metrics`	Query	R3	Read ECS monitoring data through hcloud/CES
`huawei_get_elb_metrics`	Query	R3	Read ELB monitoring data through hcloud/CES
`huawei_get_eip_metrics`	Query	R3	Read EIP monitoring data through hcloud/CES
`huawei_get_nat_gateway_metrics`	Query	R3	Read NAT Gateway monitoring data through hcloud/CES
`huawei_cce_cluster_monitoring_aggregation`	Query + local analysis	R3	Aggregate Pod/Node/cloud-resource metrics and classify anomalies locally

Parameter Reference

Common Parameters

Parameter	Required/Optional	Description	Default
`region`	Required	Huawei Cloud region	`HUAWEI_REGION`
`cluster_id`	Required	CCE cluster ID	N/A
`namespace`	Recommended	Kubernetes namespace	`default`
`ak`	Optional	Explicit AK; highest priority for all calls	profile/env fallback
`sk`	Optional	Explicit SK; highest priority for all calls	profile/env fallback
`project_id`	Optional	Explicit Project ID; hcloud uses profile before env fallback	Auto from IAM/profile

`huawei_get_cce_pod_metrics_topN` Parameters

Parameter	Required	Description	Default
`namespace`	No	Namespace filter	all
`label_selector`	No	Label selector (e.g. app=web)	N/A
`top_n`	No	Number of top items	10
`hours`	No	Metrics lookback hours	1
`node_ip`	No	Filter Pods on specific node	N/A
`cpu_query`	No	Custom CPU PromQL	Auto
`memory_query`	No	Custom memory PromQL	Auto
`disk_query`	No	Custom disk PromQL	Auto

`huawei_get_cce_pod_metrics` Parameters

Parameter	Required	Description	Default
`pod_name`	Yes	Target Pod name	N/A
`namespace`	No	Namespace	`default`
`hours`	No	Metrics lookback hours	1
`cpu_query`	No	Custom CPU PromQL	Auto
`memory_query`	No	Custom memory PromQL	Auto
`disk_query`	No	Custom disk PromQL	Auto

`huawei_get_cce_pod_gpu_metrics` Parameters

Parameter	Required	Description	Default
`pod_name`	Yes	Target Pod name	N/A
`namespace`	No	Target Pod namespace	all
`hours`	No	Metrics lookback hours	1
`gpu_selector`	No	Custom GPU metric label selector. Use this when GPU metrics do not use the `pod` or `namespace` labels	`pod="<pod_name>",namespace="<namespace>"`

Optional custom PromQL overrides are supported for GPU utilization, memory, schedule policy, xGPU allocation/usage, and xGPU health metrics.

`huawei_get_cce_node_metrics_topN` Parameters

Parameter	Required	Description	Default
`top_n`	No	Number of top items	10
`hours`	No	Metrics lookback hours	1

`huawei_get_cce_node_metrics` Parameters

Parameter	Required	Description	Default
`node_ip`	Yes	Target Node IP	N/A
`hours`	No	Metrics lookback hours	1

`huawei_get_cce_node_gpu_metrics` Parameters

Parameter	Required	Description	Default
`node_ip`	Yes	Target Node IP or node name	N/A
`hours`	No	Metrics lookback hours	1
`gpu_selector`	No	Custom GPU metric label selector. Use this when GPU metrics do not use the `node` label	`node=~"<node_ip>

Optional custom PromQL overrides are supported for GPU utilization, memory, temperature, power, schedule policy, xGPU allocation/usage, and xGPU health metrics.

`huawei_get_cce_coredns_metrics` Parameters

Parameter	Required	Description	Default
`namespace`	No	CoreDNS namespace	`kube-system`
`pod_regex`	No	Regex used to match CoreDNS Pods	`.coredns.`
`hours`	No	Metrics lookback hours	1

Optional custom PromQL overrides are supported for QPS, error rate, NXDOMAIN rate, P95 latency, CPU, memory, and replica count.

`huawei_get_cce_nginx_ingress_metrics` Parameters

Parameter	Required	Description	Default
`namespace`	No	Namespace of nginx-ingress controller Pods. Use an empty value to query all namespaces	`kube-system`
`pod_regex`	No	Regex used to match nginx-ingress controller Pods	`.*nginx.ingress.
`ingress_namespace`	No	Namespace filter for Ingress TLS certificate checks	all
`hours`	No	Metrics lookback hours	1
`cert_expire_warning_days`	No	Days before expiry to mark certificates as warning	30
`check_certificates`	No	Whether to inspect Ingress TLS Secrets for expiration status	true

Ingress-controller metrics depend on the corresponding AOM PodMonitor. The nginx_ingress_controller_requests metric must be explicitly allowed in the ingress-controller PodMonitor; otherwise request-dimension metrics such as 4xx/5xx QPS, success rate, and latency may be empty, and QPS may only use the nginx_ingress_controller_nginx_process_requests_total fallback when available.

Optional custom PromQL overrides are supported for QPS, 4xx/5xx, success rate, P95 latency, active connections, CPU, and memory.

`huawei_get_cce_autoscaler_metrics` Parameters

Parameter	Required	Description	Default
`namespace`	No	Namespace of Cluster Autoscaler Pods. Use an empty value to query all namespaces	`kube-system`
`pod_regex`	No	Regex used to match autoscaler Pods	`.*cluster.autoscaler.
`hpa_namespace`	No	Namespace filter for HPA replica metrics	all
`hours`	No	Metrics lookback hours	1
`include_hpa`	No	Whether to query HPA current/desired replica metrics	true

Optional custom PromQL overrides are supported for unschedulable Pods, node states, scale events, errors, node groups, HPA replicas, CPU, and memory.

Kubernetes Control Plane Tool Parameters

Applies to huawei_get_cce_apiserver_metrics, huawei_get_cce_etcd_metrics, huawei_get_cce_controller_manager_metrics, and huawei_get_cce_scheduler_metrics.

huawei_get_cce_apiserver_metrics defaults to cluster="<cluster_id>",component="apiserver" and does not add namespace or Pod labels. Its default P95 latency excludes WATCH|CONNECT requests and also returns latency_p95_by_verb_ms for diagnosis. Use metric_selector only when the Prometheus labels differ.

huawei_get_cce_etcd_metrics defaults to cluster="<cluster_id>" and does not add namespace or Pod labels. Use metric_selector only when the Prometheus labels differ.

huawei_get_cce_controller_manager_metrics defaults to cluster="<cluster_id>" because CCE AOM workqueue metrics may not expose stable controller-manager Pod labels. It returns both aggregate workqueue metrics and per-queue name breakdowns.

huawei_get_cce_scheduler_metrics defaults to cluster="<cluster_id>" and returns aggregate metrics plus result, profile/result, and queue breakdowns.

Controller-manager, scheduler, and etcd metrics depend on AOM ServiceMonitor collection being enabled for the corresponding kube-controller-manager, kube-scheduler, and etcd-server endpoints. If ServiceMonitor is not enabled, the tools can run successfully but return empty series.

Parameter	Required	Description	Default
`namespace`	No	Namespace of control-plane Pods. Use an empty value to query all namespaces	`kube-system`
`pod_regex`	No	Regex used to match target component Pods	component-specific
`metric_selector`	No	Custom apiserver/etcd/controller-manager/scheduler metric label selector	apiserver: `cluster="<cluster_id>",component="apiserver"`; etcd/controller-manager/scheduler: `cluster="<cluster_id>"`
`hours`	No	Metrics lookback hours	1

Cloud Resource Tool Parameters

Tool	Required ID Parameter	Optional Parameters
`huawei_get_ecs_metrics`	`instance_id`	none
`huawei_get_elb_metrics`	`elb_id`	`hours`
`huawei_get_eip_metrics`	`eip_id`	`hours`
`huawei_get_nat_gateway_metrics`	`nat_gateway_id`	`hours`

`huawei_cce_cluster_monitoring_aggregation` Parameters

Parameter	Required	Description	Default
`start_time`	Yes	Start time (YYYY-MM-DD HH:MM:SS)	N/A
`end_time`	Yes	End time (YYYY-MM-DD HH:MM:SS)	N/A
`namespace`	No	Namespace filter	`default`
`top_n`	No	Number of top items	10
`security_token`	No	Temporary security token for AK/SK session credentials	env fallback

Output Format

See Output Schema for the complete JSON response structure.

Key output fields:

success — boolean, true if query completed
region — Huawei Cloud region
cluster_id / cluster_name — CCE cluster identity
aom_instance_id — AOM Prometheus instance used for metric queries
metrics — Dict with cpu/memory/disk data per resource, including status classification
certificate_check — nginx-ingress Ingress TLS certificate expiration summary when certificate checking is enabled
time_series — Historical data points with timestamp, time, average, min, max
status — Threshold classification: critical (>80% CPU, >85% memory/disk), warning (>50% CPU/memory, >70% disk), normal (below warning), unknown (no data)

Workflow

Resolve region, cluster ID, and credentials using the documented priority.
Discover the AOM Prometheus instance from the CCE cluster add-on binding.
Start with Pod/Node TopN or aggregation, then drill into a Pod, Node, component, or cloud resource.
Keep PromQL scoped by cluster="<cluster_id>"; add namespace, pod, or resource filters only to reduce noise.
Use status classification as an investigation lead, then correlate anomalies with events or alarm history.

Verification

Run python3 scripts/huawei-cloud.py huawei_get_cce_pod_metrics_topN region=cn-north-4 cluster_id=<cluster-id> namespace=default top_n=5 to verify Pod metric queries
Run python3 scripts/huawei-cloud.py huawei_get_cce_node_metrics_topN region=cn-north-4 cluster_id=<cluster-id> top_n=5 to verify Node metric queries
Run python3 scripts/huawei-cloud.py huawei_get_ecs_metrics region=cn-north-4 instance_id=<instance-id> to verify CES metric connectivity

Best Practices

Start with Pod/Node TopN before drilling into individual resources.
Keep hours small (1-4) for recent analysis; cap historical reviews at 24 hours.
Provide namespace to reduce Pod noise while preserving the cluster filter.
Focus on critical and warning resources first.
Use huawei_cce_cluster_monitoring_aggregation for full-cluster health checks.
Correlate metric anomalies with huawei-cloud-cce-kubernetes-event-analyzer.
Do not expose production Pod names, node IPs, or cluster IDs in public summaries.

Notes

This skill is strictly read-only and never modifies resources or configurations.
Thresholds are predefined baselines; tune them against workload SLOs before making operational decisions.
AK/SK must never be hardcoded; use hcloud profile for normal hcloud calls or environment fallback for signed AOM/Kubernetes calls.
scripts/huawei-cloud.py is the only user-facing execution method
AOM Prometheus instance is auto-discovered; no need to manually specify aom_instance_id
Cloud resource metrics (ECS/ELB/EIP/NAT) use CES (Cloud Eye Service), not AOM
Do not make automatic scaling or remediation decisions based solely on metric analysis.

Troubleshooting

Pitfall	Symptom	Quick Fix
Missing `cluster_id`	Action fails immediately	Provide `cluster_id` from cluster listing
AOM Prometheus instance not found	Metric queries return empty results	Ensure AOM Prom instance is created for the cluster; check `aom:instance:list` permission
Large time window without namespace filter	Slow response, too many results	Narrow `hours` to 1-4 and add `namespace` filter
Cloud resource ID not found	ECS/ELB/EIP/NAT query returns error	Verify resource ID and CES IAM permission
Custom PromQL syntax error	Custom query returns empty	Use default PromQL unless familiar with AOM PromQL
Aggregation missing time range	`start_time` / `end_time` missing	Provide both time boundaries

Limitations

AOM Prometheus data requires the cluster Prometheus add-on to be integrated with AOM.
Control-plane ServiceMonitors and component PodMonitors must be enabled before related metrics appear.
Query results reflect collected monitoring data only; missing series are not proof that the workload is healthy.
This skill does not remediate, scale, restart, create, update, or delete cloud or Kubernetes resources.

References

Document	Description
Workflow	Metric query sequence, threshold detection, next-step handoff
Risk Rules	Read-only constraints, data redaction, time-bounding, threshold caveats
Output Schema	JSON response schema for metric and status output
CLI Installation Guide	hcloud, kubectl, kubectl-cce, and dispatcher setup
IAM Policies	Required read-only Huawei Cloud and Kubernetes permissions
Verification Method	Static checks and smoke tests
Acceptance Criteria	Functional, security, documentation, and quality gates

huawei-cloud-cce-metric-analyzer