sync status (vibe-kanban)

CLAUDE.md

@@ -91,4 +91,54 @@ def to_json_serializable(obj):
 1. Check node sync status with monitoring scripts
 2. Create chunks for historical data processing
 3. Verify 100% data completeness before processing
-4. Use management commands for bulk operations
+4. Use management commands for bulk operations
+
+### Production Ethereum Node Deployment
+
+#### Initial Setup
+```bash
+# Set up HOME cluster credentials
+python manage.py setup_home_cluster --namespace devbox
+
+# Create and start a node
+python manage.py create_node eth-mainnet-01 --start
+```
+
+#### Resource Requirements (CRITICAL)
+**Execution Client (Geth):**
+- Memory: 16Gi limit, 8Gi request
+- CPU: 4 cores limit, 2 cores request
+- Storage: 2TB minimum (8TB recommended for growth)
+
+**Consensus Client (Lighthouse):**
+- Memory: **12Gi limit, 6Gi request** (8Gi causes OOM during sync)
+- CPU: 4 cores limit, 2 cores request  
+- Storage: 200GB minimum
+
+#### Node Selection
+- **Preferred nodes**: vega, nova (have working NFS CSI drivers)
+- **Avoid**: enterprise, ziti (resource constrained, cause pod failures)
+- Use `kubernetes.io/hostname` selector for explicit placement
+
+#### Storage Classes
+- HOME cluster: `nfs-iota-hdd-slush` (NFS-based, good for blockchain data)
+- Avoid `local-path` for production (node-specific, not portable)
+
+#### Common Issues & Fixes
+1. **Consensus OOM kills (exit code 137)**
+   - Increase memory limit to 12Gi minimum
+   - Watch for "Database write failed" errors in logs
+
+2. **Pod stuck in Init phase**
+   - Check PVC mounting issues
+   - Verify NFS CSI driver is running on target node
+   - May need to restart k3s-agent on problematic nodes
+
+3. **Database lock errors**
+   - Scale deployment to 0, then back to 1
+   - Ensures clean shutdown and lock release
+
+4. **Nova node issues**
+   - New nodes may have Cilium/CSI initialization problems
+   - SSH to node and restart k3s-agent if needed
+   - Check for "services have not yet been read" errors

zeroindex/apps/nodes/management/commands/setup_home_cluster.py

@@ -0,0 +1,96 @@
+"""
+Management command to set up HOME cluster credentials for blockchain node deployment.
+"""
+import base64
+import os
+from django.core.management.base import BaseCommand, CommandError
+from zeroindex.apps.nodes.models import KubeCredential
+from zeroindex.apps.chains.models import Chain
+
+
+class Command(BaseCommand):
+    help = 'Set up HOME cluster Kubernetes credentials for blockchain deployments'
+
+    def add_arguments(self, parser):
+        parser.add_argument(
+            '--kubeconfig-path',
+            type=str,
+            default=os.path.expanduser('~/.kube/clusters/home'),
+            help='Path to HOME cluster kubeconfig file'
+        )
+        parser.add_argument(
+            '--namespace',
+            type=str,
+            default='devbox',
+            help='Kubernetes namespace for deployments (default: devbox)'
+        )
+        parser.add_argument(
+            '--name',
+            type=str,
+            default='home-cluster',
+            help='Name for the credential entry'
+        )
+
+    def handle(self, *args, **options):
+        kubeconfig_path = options['kubeconfig_path']
+        namespace = options['namespace']
+        name = options['name']
+
+        # Check if kubeconfig file exists
+        if not os.path.exists(kubeconfig_path):
+            raise CommandError(f'Kubeconfig file not found: {kubeconfig_path}')
+
+        # Read and encode kubeconfig
+        with open(kubeconfig_path, 'r') as f:
+            kubeconfig_content = f.read()
+
+        kubeconfig_b64 = base64.b64encode(kubeconfig_content.encode()).decode()
+
+        # Check if credential already exists
+        existing = KubeCredential.objects.filter(name=name).first()
+        if existing:
+            self.stdout.write(f'Updating existing credential: {name}')
+            existing.namespace = namespace
+            existing.kubeconfig = kubeconfig_b64
+            existing.is_active = True
+            existing.save()
+            credential = existing
+        else:
+            # Create new credential
+            credential = KubeCredential.objects.create(
+                name=name,
+                cluster_name='HOME',
+                namespace=namespace,
+                kubeconfig=kubeconfig_b64,
+                is_active=True
+            )
+            self.stdout.write(self.style.SUCCESS(f'Created new credential: {name}'))
+
+        # Ensure Ethereum mainnet chain exists
+        chain, created = Chain.objects.get_or_create(
+            chain_id=1,
+            defaults={
+                'name': 'Ethereum',
+                'symbol': 'ETH',
+                'is_testnet': False
+            }
+        )
+        if created:
+            self.stdout.write(self.style.SUCCESS('Created Ethereum mainnet chain'))
+
+        # Display summary
+        self.stdout.write(f'\n{self.style.HTTP_INFO("Configuration Summary:")}')
+        self.stdout.write(f'  Credential: {credential.name}')
+        self.stdout.write(f'  Cluster: {credential.cluster_name}')
+        self.stdout.write(f'  Namespace: {credential.namespace}')
+        self.stdout.write(f'  Active: {credential.is_active}')
+        self.stdout.write(f'  Chain: {chain.name} (ID: {chain.chain_id})')
+
+        self.stdout.write(f'\n{self.style.SUCCESS("✓ HOME cluster is ready for blockchain deployments")}')
+        self.stdout.write('\nTo create a node, run:')
+        self.stdout.write('  python manage.py create_node eth-mainnet-01 --start')
+        self.stdout.write('\nRecommended node configuration:')
+        self.stdout.write('  - Deploy to vega or nova nodes (avoid enterprise/ziti)')
+        self.stdout.write('  - Use nfs-iota-hdd-slush storage class')
+        self.stdout.write('  - Consensus needs 12Gi memory limit minimum')
+        self.stdout.write('  - Execution needs 16Gi memory limit for full node')

zeroindex/apps/nodes/models.py

@@ -106,8 +106,8 @@ class Node(models.Model):
     storage_class = models.CharField(
         max_length=100, 
         blank=True, 
-        default="iota-slush",
-        help_text="Kubernetes storage class for PVCs (e.g., iota-slush)"
+        default="nfs-iota-hdd-slush",
+        help_text="Kubernetes storage class for PVCs (e.g., nfs-iota-hdd-slush for HOME cluster)"
     )
 
     # Node targeting for Kubernetes scheduling
@@ -246,9 +246,10 @@ def get_default_consensus_resources(self):
                 'limits': {'cpu': '4', 'memory': '12Gi'}
             }
         else:
+            # Increased from 8Gi to 12Gi to prevent OOM during sync
             return {
-                'requests': {'cpu': '1', 'memory': '4Gi'},
-                'limits': {'cpu': '2', 'memory': '8Gi'}
+                'requests': {'cpu': '2', 'memory': '6Gi'},
+                'limits': {'cpu': '4', 'memory': '12Gi'}
             }
 
     def get_default_resource_requests(self):