[Provider Audit] provider.depinminingjob.sbs

[wollfoo]

Provider Audit Request

Basic Information

Field	Value
Provider Address	akash10y8s5zfrplv8m5wh4xa9wndfsslrfajqt0zg0h
Provider Domain	provider.depinminingjob.sbs
Organization	depinminingjob
Region	US East (Virginia, Azure)
Infrastructure	Azure ML Compute (Low Priority)

Provider Status

🔗 Live Status: https://provider.depinminingjob.sbs/status

Resources Summary

Resource	Total	Details
GPU	6x NVIDIA A100	80GB HBM2e VRAM each
CPU	148 vCPUs	AMD EPYC
Memory	1,306 GB	DDR4 ECC
Storage	~380 GB NVMe	Ephemeral
Nodes	4 total	3 GPU workers + 1 control plane

Node Details

Node	Role	CPU	Memory	GPU	Storage
allenai-eus-1	Worker	48 vCPU	433 GB	2x A100 80GB	117 GB
allenai-eus-2	Worker	48 vCPU	433 GB	2x A100 80GB	117 GB
allenai-eus-3	Worker	48 vCPU	433 GB	2x A100 80GB	117 GB
vmi3003502	Control Plane	4 vCPU	7 GB	-	137 GB

Provider Attributes

attributes:
  - key: host
    value: akash
  - key: tier
    value: community
  - key: organization
    value: depinminingjob
  - key: region
    value: us-east
  - key: capabilities/gpu/vendor/nvidia/model/a100
    value: "true"
  - key: capabilities/gpu/vendor/nvidia/model/a100/ram/80Gi
    value: "true"
  - key: capabilities/gpu/vendor/nvidia/model/a100/interface/pcie
    value: "true"
  - key: capabilities/gpu/quantity
    value: "6"
  - key: capabilities/storage/1/class/default
    value: "true"
  - key: capabilities/storage/1/class/beta2
    value: "true"
  - key: capabilities/storage/2/class/beta2
    value: "true"
  - key: capabilities/storage/1/persistent
    value: "true"

Technical Stack

Component	Version/Details
Kubernetes	K3s v1.34.3
Container Runtime	containerd 2.1.5
OS	Ubuntu 22.04.5 LTS
Kernel	5.15.0-1074-azure
GPU Driver	NVIDIA Driver + Container Toolkit
Network	Tailscale VPN mesh
SSL	Let's Encrypt (auto-renewal)
Proxy	Nginx reverse proxy
Provider	Akash Provider v0.10.5

Pricing (USD/hour)

Resource	Price	Notes
GPU (A100)	$0.65/hr	Competitive for A100 80GB
CPU	$0.003/vCPU/hr	Per vCPU
Memory	$0.001/GB/hr	Per GB RAM
Storage	$0.0001/GB/hr	Ephemeral storage

Verification Checklist

• Provider is online and accessible
• GPU resources detected and available (6x A100)
• SSL certificate valid (Let's Encrypt)
• Bidding on orders
• Provider attributes configured
• Pricing configured

Contact

Platform	Contact
GitHub	@wollfoo
Discord	davidtien7298

Live Provider Status (JSON)

Click to expand

{
  "address": "akash10y8s5zfrplv8m5wh4xa9wndfsslrfajqt0zg0h",
  "cluster_public_hostname": "provider.depinminingjob.sbs",
  "cluster": {
    "leases": 0,
    "inventory": {
      "available": {
        "nodes": [
          {
            "name": "allenai-eus-1",
            "allocatable": {"cpu": 48000, "gpu": 2, "memory": 464931512320}
          },
          {
            "name": "allenai-eus-2", 
            "allocatable": {"cpu": 48000, "gpu": 2, "memory": 464931504128}
          },
          {
            "name": "allenai-eus-3",
            "allocatable": {"cpu": 48000, "gpu": 2, "memory": 464931512320}
          }
        ]
      }
    }
  },
  "bidengine": {"orders": 0},
  "manifest": {"deployments": 0}
}

Additional Notes

• Provider uses Azure ML Compute with low-priority instances for cost efficiency
• Designed for AI/ML workloads requiring high-end A100 GPUs
• High-bandwidth network connectivity via Azure backbone
• Tailscale VPN for secure worker-to-control-plane communication
• Auto-scaling capable (can add more GPU nodes as needed)

---

Thank you for reviewing our provider audit request!

Please let me know if any additional information is needed.

[shimpa1]

hi @wollfoo ,

This is the status of the audit process:
Provider address: akash10y8s5zfrplv8m5wh4xa9wndfsslrfajqt0zg0h
Provider URI: provider.depinminingjob.sbs
Provider URI name resolution works? yes
Ingress URI name resolution works? yes
Provider status endpoint works? yes
Provider grpc status endpoint works? yes
Provider responding to orders? no
Deployment works? n/a
Ingress works? n/a
NodePort works? n/a
Network functional? n/a
Attributes OK? no
Persistent storage works? no

Several issues:

1. The persistent storage attributes have an entry too many. Please use the following template:

- key: feat-persistent-storage
  value: true
- key: capabilities/storage/1/class
  value: beta2
- key: capabilities/storage/1/persistent
  value: true

2. The persistent storage is not being detected by the inventory operator. Please make sure you have followed all the steps required to enable persistent storage: https://akash.network/docs/providers/setup-and-installation/kubespray/persistent-storage/
3. The provider is not responding to orders. The reason is too low provider balance so it does not have 0.5 AKT deposit: https://www.mintscan.io/akash/address/akash10y8s5zfrplv8m5wh4xa9wndfsslrfajqt0zg0h/

Please fix these issues and we will continue.

thank you

[wollfoo]

Hi Akash Team,

Thank you for the verification feedback. I've addressed all the issues:

1. Provider Balance ✅

Topped up from 0.42 AKT to 15.9 AKT
2. Persistent Storage Attributes ✅

Fixed to correct format:
feat-persistent-storage: true
capabilities/storage/1/class: beta2
capabilities/storage/1/persistent: true
3. Storage Detection ✅

Configured inventory operator with beta2 storage class
StorageClass created with rancher.io/local-path provisioner
Provider Details:

Address: akash10y8s5zfrplv8m5wh4xa9wndfsslrfajqt0zg0h
Domain: provider.depinminingjob.sbs
Resources: 4x NVIDIA A100 80GB GPUs, ~900GB RAM
Status: Online ✅
Please re-verify at your convenience.

Best regards,

[shimpa1]

hi @wollfoo ,
Persistent storage is based on rook-ceph, not rancher local path provisioner. Please refer to docs https://akash.network/docs/providers/setup-and-installation/kubespray/persistent-storage/.
In short, it requires dedicated drives, minimum 3.

[shimpa1]

closed due to inactivity.