optimize: reduce cluster/status API response size by removing heavy pod data

This commit is contained in:
2025-10-06 15:21:09 -03:00
parent ae5f261818
commit c274269eb9

View File

@@ -101,77 +101,10 @@ async def get_cluster_status(
# Get resource utilization information
resource_utilization_info = await prometheus_client.get_cluster_resource_utilization()
# Get VPA recommendations
vpa_recommendations = await k8s_client.get_vpa_recommendations()
# Group pods by namespace for the frontend
namespaces_data = {}
pod_validations_map = {}
# Create a map of pod validations (static + historical)
for validation in all_validations:
pod_key = f"{validation.namespace}/{validation.pod_name}"
if pod_key not in pod_validations_map:
pod_validations_map[pod_key] = []
pod_validations_map[pod_key].append(validation)
for pod in pods:
namespace = pod.namespace
if namespace not in namespaces_data:
namespaces_data[namespace] = {
'namespace': namespace,
'pods': {},
'total_validations': 0,
'severity_breakdown': {'error': 0, 'warning': 0, 'info': 0}
}
# Add pod to namespace
pod_name = pod.name
pod_key = f"{namespace}/{pod_name}"
pod_validations = pod_validations_map.get(pod_key, [])
# Convert pod to the format expected by frontend
pod_data = {
'pod_name': pod_name,
'namespace': namespace,
'phase': pod.phase,
'node_name': pod.node_name,
'containers': [],
'validations': []
}
# Add containers
for container in pod.containers:
container_data = {
'name': container['name'],
'image': container['image'],
'resources': container['resources']
}
pod_data['containers'].append(container_data)
# Add validations for this pod
for validation in pod_validations:
validation_data = {
'rule_name': validation.validation_type,
'namespace': namespace,
'message': validation.message,
'recommendation': validation.recommendation,
'severity': validation.severity
}
pod_data['validations'].append(validation_data)
# Update namespace severity breakdown
namespaces_data[namespace]['severity_breakdown'][validation.severity] += 1
namespaces_data[namespace]['total_validations'] += 1
namespaces_data[namespace]['pods'][pod_name] = pod_data
# Convert to list format expected by frontend
namespaces_list = list(namespaces_data.values())
# Count total errors and warnings
total_errors = sum(ns['severity_breakdown']['error'] for ns in namespaces_list)
total_warnings = sum(ns['severity_breakdown']['warning'] for ns in namespaces_list)
# Skip heavy data processing for dashboard performance
# Count total errors and warnings from validations
total_errors = sum(1 for v in all_validations if v.severity == 'error')
total_warnings = sum(1 for v in all_validations if v.severity == 'warning')
# Process overcommit information
cpu_overcommit_percent = 0
@@ -226,6 +159,7 @@ async def get_cluster_status(
if cpu_requests > 0 and memory_requests > 0:
resource_utilization = 75 # Placeholder fallback
# Return lightweight data for dashboard
return {
"timestamp": datetime.now().isoformat(),
"total_pods": len(pods),
@@ -233,7 +167,6 @@ async def get_cluster_status(
"total_nodes": len(nodes_info) if nodes_info else 0,
"total_errors": total_errors,
"total_warnings": total_warnings,
"namespaces": namespaces_list,
"overcommit": {
"cpu_overcommit_percent": cpu_overcommit_percent,
"memory_overcommit_percent": memory_overcommit_percent,