Skip to content
Riversoft STCP DirectLink
Integration
Cloud storage

Cloud storage

Riversoft® STCP DirectLink exchanges files with cloud storage buckets and containers. The connection is always started by Riversoft STCP DirectLink, which writes and fetches objects in the repository set for each user.

ConnectorRepository
Amazon S3Amazon S3 buckets and S3 API-compatible services.
Azure Blob StorageContainers of an Azure storage account.
Google Cloud StorageGoogle Cloud Storage buckets.
Oracle Cloud Object StorageOracle Cloud Infrastructure (OCI) Object Storage buckets.

Outbound connections

Send

    flowchart LR
    subgraph DL[Riversoft STCP DirectLink]
      SA["saida directory"] --> C[Cloud connector]
    end
    C -->|send| P[Upload bucket or container]

Fetch

    flowchart LR
    subgraph DL[Riversoft STCP DirectLink]
      C[Cloud connector] --> EN["entrada directory"]
    end
    P[Download bucket or container] -->|fetch| C

Note

Sending and fetching can happen in the same connection: Riversoft STCP DirectLink first sends the files, then fetches the partner’s files. Depending on the mode configured in the user, the connection can also only send or only fetch.

Riversoft STCP DirectLink connects to the repository by schedule, or when the operator starts the connection in the Riversoft Console Portal. In the same connection, it:

  1. sends the files available in the user’s mailbox (saida directory) to the upload bucket or container;
  2. fetches the objects of the download bucket or container and writes them to the entrada directory of the mailbox.
StepBehavior
SendEach file is written as an object, with the configured prefix (folder). In Google Cloud Storage, the upload is refused if an object with the same name already exists.
FetchOnly objects in the configured prefix and matching the mask are fetched.
After fetchingThe object can be copied to a backup prefix and, if configured, removed from the source.

The upload and download buckets or containers can be the same or different, and each has its own prefix.

Authentication by provider

ProviderAuthentication options
Amazon S3Access key (access key and secret key, with an optional session token); environment or profile credentials; assumed role (AssumeRole); IAM Roles Anywhere, with a certificate.
Azure Blob StorageStorage account name and key (Shared Key) or SAS token.
Google Cloud StorageService account JSON key, or the environment default credentials.
Oracle Cloud Object StorageAPI key: tenancy, user, fingerprint and private key, with the bucket region.

In Amazon S3, a custom endpoint allows S3 API-compatible services, including in path-style format. The connector can also check that the buckets exist before transferring.

What to agree with the repository owner

ItemDescription
RepositoryUpload and download bucket or container, region and endpoint, when not the provider default.
PrefixesUpload, download and backup folders.
CredentialThe provider authentication option, with permission only on the buckets or containers used.
Post-processingWhether fetched objects must be removed or copied to backup.

These details are registered in the user, in the Riversoft Console Portal, together with the connection schedule.

Security

  • Give the connector credential only the permissions needed on the buckets or containers used (read, write and delete, when there is removal).
  • Prefer temporary credentials (assumed role or IAM Roles Anywhere in Amazon S3, SAS token with an expiration in Azure Blob Storage).
  • Keep the credentials in a secrets provider. See KMS and secrets.
  • Connections use HTTPS. See Security: connectors.
REST APIAS2