After installing your provider, verify it’s running correctly and accepting bids.
Quick Health Checks
1. Check Provider Pod Status
kubectl -n akash-services get podsExpected output:
NAME READY STATUS RESTARTS AGEakash-provider-0 1/1 Running 0 5mStatus should be Running with 1/1 Ready.
2. View Provider Logs
kubectl -n akash-services logs -l app=akash-provider --tail=100 -fLook for:
- **
"bidding on order"- Provider is creating bids - **
"bid complete"- Bids are being submitted - **
"error"or"failed"- Investigate errors
3. Verify On-Chain Registration
provider-services query provider get <your-provider-address>Verify:
- Provider address is correct
- Host URI is accessible
- Attributes are set correctly
4. Check Provider Status Page
curl -k https://<your-provider-domain>:8443/statusExpected: JSON response with provider status information.
5. Check Provider gRPC Endpoint
grpcurl -insecure <your-provider-domain>:8444 akash.provider.v1.ProviderRPC.GetStatusExpected: JSON response with provider status via gRPC.
Install grpcurl (if needed):
# Ubuntu/Debianapt install -y grpcurl
# macOSbrew install grpcurlBid Activity Monitoring
Watch for Bids
kubectl -n akash-services logs -l app=akash-provider --tail=100 -f | grep bidIf you see no bids:
- Check your pricing is competitive
- Verify provider attributes are set
- Ensure firewall allows inbound connections
- Check wallet has sufficient balance
Common Issues
Provider Pod Not Running
Check pod status:
kubectl -n akash-services describe pod akash-provider-0Common causes:
- Insufficient resources
- Configuration errors in
provider.yaml - Missing secrets or certificates
Fix: Review pod events and logs for specific errors.
No Bids Showing
Possible causes:
- Pricing too high - Check your bid pricing script
- Attributes missing - Verify GPU and feature attributes
- Wallet balance low - Provider needs AKT for transaction fees
- Network issues - Verify connectivity to RPC nodes
Debug:
# Check provider account balanceprovider-services query bank balances <provider-address>Provider Not Accessible Externally
Test connectivity:
# From external machinecurl -k https://<provider-domain>:8443/statusgrpcurl -insecure <provider-domain>:8444 akash.provider.v1.ProviderRPC.GetStatusIf it fails:
- Verify DNS points to your provider IP
- Check firewall allows ports 8443 (HTTP) and 8444 (gRPC)
- Ensure provider domain is in certificate
Deployment Pods Failing
Check pod status:
kubectl -n lease get podskubectl -n lease describe pod <pod-name>Common issues:
- GPU not available (for GPU deployments)
- Insufficient resources
- Image pull errors
- Network policies blocking traffic
Verification Checklist
After provider installation, verify:
- Provider pod is running (
kubectl -n akash-services get pods) - Provider is registered on-chain (
provider-services query provider get) - Provider status endpoint is accessible (
curl https://provider:8443/status) - Provider gRPC endpoint is accessible (
grpcurl provider:8444 akash.provider.v1.ProviderRPC.GetStatus) - Provider is bidding on orders (check logs)
- Ingress controller is running (
kubectl -n ingress-nginx get pods) - GPU device plugin is running (GPU providers only)
- Wallet has sufficient balance
- Firewall allows required ports (8443, 8444)
- Domain DNS is configured
Related Resources
- Monitoring - Provider logs and GPU troubleshooting
- Lease Management - Managing active deployments
- Provider Installation
- Hardware Requirements
Need Help? Join #provider-support on Discord