[Provider Audit] provider.clomd.com

kevparang commented 1 month ago

Prerequisite Steps:

1. Make sure your provider has community provider attributes and your contact details (email, website):

attributes:
- key: organization
  value: Clomd
- key: capabilities/gpu/vendor/nvidia/model/a100
  value: true
- key: arch
  value: amd64
- key: host
  value: akash
- key: tier
  value: community
  info:
    email: "kevan@clomd.com"
    website: "www.clomd.com"

2. Make sure your provider *.ingress resolves to your provider IP (ideally worker node IP)

host <anything>.ingress.<yourdomain>

Example:

:~$ host provider.clomd.com
provider.clomd.com has address 129.146.97.136
:~$ host anything.ingress.clomd.com
anything.ingress.clomd.com has address 129.146.97.136

3. Please make sure your Akash provider doesn't block any Akash specific ports.

It doesn't block any ports.

Leave contact information (optional)

Kevan
kevan6721
kevan@clomd.com

shimpa1 commented 1 month ago

hi @kevparang ,

provider-services query provider get akash194vhk23ee3k3d6tcnz36wwq342lsv32qqnfj7z -o text attributes:

key: organization value: Clomd
key: capabilities/gpu/vendor/nvidia/model/a100 value: "true"
key: arch value: amd64
key: tier value: community
key: host value: akash
key: email value: kevan@clomd.com
key: website value: www.clomd.com host_uri: https://provider.clomd.com:8443 info: email: "" website: "" owner: akash194vhk23ee3k3d6tcnz36wwq342lsv32qqnfj7z

Your provider is missing some attributes. This is an excerpt from one of the a100 providers. Please adjust to match your GPU memory size and interface type.

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/ram/80Gi/interface/pcie value: "true"
key: capabilities/gpu/vendor/nvidia/model/a100/interface/pcie value: "true"

Next, the info section requires updating: info: email: "" website: ""

When the information has been entered, hen do helm upgrade --install akash-provider akash/provider -n akash-services -f ~/provider/provider.yaml --set bidpricescript="$(cat ~/provider/price_script_generic.sh | openssl base64 -A)"

We will continue with the process.

kevparang commented 1 month ago

I have completed it, please check again.

andy108369 commented 1 month ago

provider-services query provider get akash194vhk23ee3k3d6tcnz36wwq342lsv32qqnfj7z -o text

@kevparang your provider is still missing the necessary info fields and is now also offline:

$ ping provider.clomd.com
PING provider.clomd.com (129.146.97.136) 56(84) bytes of data.
^C
--- provider.clomd.com ping statistics ---
4 packets transmitted, 0 received, 100% packet loss, time 3064ms

$ nc -vz provider.clomd.com 8443
<hangs here>

missing info email & website fields:

$ provider-services query provider get akash194vhk23ee3k3d6tcnz36wwq342lsv32qqnfj7z -o text
attributes:
- key: organization
  value: Clomd
- key: capabilities/gpu/vendor/nvidia/model/a100
  value: "true"
- key: arch
  value: amd64
- key: tier
  value: community
- key: host
  value: akash
- key: email
  value: kevan@clomd.com
- key: website
  value: www.clomd.com
host_uri: https://provider.clomd.com:8443
info:
  email: ""
  website: ""
owner: akash194vhk23ee3k3d6tcnz36wwq342lsv32qqnfj7z

Please do not add them as the provider attributes but as the info fields, for that simply add them to your provider.yaml file as follows:

email: kevan@clomd.com
website: https://www.clomd.com

And for the GPU model, please add more fields, e.g.:

make sure to set the RAM & interface correctly of your A100

- 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/ram/80Gi/interface/pcie
value: true
- key: capabilities/gpu/vendor/nvidia/model/a100/interface/pcie
value: true

akash-network / community