A Datastore is a pointer to remote storage from which ZEDEDA Cloud retrieves images and artifacts for deployment to edge nodes. Datastores are project-scoped and reusable across multiple image definitions. Credentials are stored encrypted in ZEDEDA Cloud and never transmitted to edge nodes in plain text.

• Supported types: S3 (AWS S3 and S3-compatible), Azure Blob Storage, HTTP/HTTPS, SFTP
• Credentials (access key, secret, SAS token, or password) stored encrypted per datastore record
• Scoped to a project; can be referenced by multiple image definitions
• Used at image download time when EVE-OS pulls an image to the local node storage

1. Navigate to Objects > Datastores
2. Click Add Datastore
3. Select the datastore type (S3, Azure, HTTP, SFTP)
4. Enter the FQDN or bucket URL, path prefix, and credentials
5. Click Save

resource "zedcloud_datastore" "s3_store" {
  name         = "workshop-s3"
  ds_type      = "DATASTORE_TYPE_S3"
  fqdn         = "s3.us-east-1.amazonaws.com"
  api_key      = var.aws_access_key
  password     = var.aws_secret_key
  dpath        = "my-bucket/images"
  project_id   = zedcloud_project.workshop.id
}

POST /v1/datastores
{
  "name": "workshop-s3",
  "dType": "DATASTORE_TYPE_S3",
  "fqdn": "s3.us-east-1.amazonaws.com",
  "apiKey": "<access_key>",
  "password": "<secret_key>",
  "dPath": "my-bucket/images",
  "projectId": "<project_id>"
}

• The datastore itself is not pushed to the edge node at creation time
• When an image referencing this datastore is assigned to an app instance on the node, EVE-OS receives the image download config including the resolved datastore URL and auth token
• EVE-OS authenticates to the datastore and fetches the image directly, verifying integrity via SHA256 hash

• Projects
• Images
• Persistent Volumes and Content Trees