Dashboard Settings
Storage
Section titled “Storage”Each account has its own S3 bucket. Files land exactly where the gateway put them
(<upload path>/<file>); anything the platform writes for itself (rollups, exports,
quarantine, the dataset registry) lives under .hublink/ and is hidden from the data
browser, from files.json and from HubLink-Client.
- Bucket: the account’s bucket name (read-only)
- Credential mode:
sts— the default. Gateways and people receive short-lived (1 hour), bucket-scoped credentials from the settings endpoint; nothing permanent is stored.static— a permanent key pair the account was migrated with, kept encrypted, handed out until every gateway on the account runs an image that can use session credentials. An administrator flips the account tostsafterwards.
Credentials
Section titled “Credentials”Settings → Credentials → Get credentials returns a short-lived key pair (with session token) for the AWS CLI or HubLink-Client:
export AWS_ACCESS_KEY_ID=…export AWS_SECRET_ACCESS_KEY=…export AWS_SESSION_TOKEN=…aws s3 sync s3://<bucket> ./data --exclude ".hublink/*"Gateway Settings
Section titled “Gateway Settings”- Device Filter by Name: the gateway only connects to nodes whose Bluetooth name contains this string; cannot be blank
- Max File Size: files above this size (bytes, at most 1 GB) are not transferred
- Timezone: used for the gateway’s local clock and the dashboard’s daily charts
(e.g.
America/Chicago) - Gateway manages memory: when the gateway’s storage passes 95 % it deletes the oldest files it has already uploaded (never files the cloud has not confirmed)
- Use cloud: turn uploads off; the gateway keeps collecting locally
Gateways
Section titled “Gateways”Every gateway that has fetched settings appears here with its last check-in, IP and contract version. A gateway on the v2 image enrolls on first boot and then holds its own token; revoke invalidates that token (the gateway re-enrolls with the secret URL from its USB stick), remove forgets the gateway entirely.
Datasets
Section titled “Datasets”Datasets register themselves: the first upload into a top-level directory creates
one (RAVEN/001/HUB_….csv → dataset raven). Plots and the Explore tab work over a
dataset’s rollup, rebuilt shortly after each upload.