First sign-in and setup
First sign-in and who may sign in
Open the UI URL and sign in with an account in the tenant you configured.
By default any user in the tenant can sign in, and every user who can sign in can change every setting: Cloud Connector has one level of access. To restrict sign-in, open the Cloud Connector Enterprise application in Microsoft Entra ID, set Assignment required? to Yes, and assign users or groups under Users and groups. Setup → Access shows whether the setting is on and links to that blade.
Configure email for notifications
Go to Setup → Runtime → Email for notifications and pick a transport:
- Microsoft 365 (recommended) — set Send as to the mailbox notifications should come from. No account signs in and no password is stored: Cloud Connector sends using its own Azure managed identity. An administrator must give that identity permission to send as this one mailbox, and only this one. The two ways to do that are written out in Email for notifications.
- SMTP relay — host, port, credentials and from-address, for tenants with no Exchange Online or with a third-party relay. Office 365 mailboxes cannot use SMTP when MFA is enforced.
Send a test email before relying on production notifications. Until a transport is set up, failure notifications are not delivered.
Allow-list Cloud Connector's outbound IP
If your databases or source systems use IP allow-lists, copy the IPs or the ready-made Azure SQL firewall script from Setup → Access → Cloud Connector external IP. With a NAT Gateway the IP is stable; without one it comes from a shared Azure pool and may change without notice.
Signing in and navigating the UI
Sign-in flow
Cloud Connector signs users in with Microsoft Entra ID through the "Cloud Connector" app registration, using the browser-side redirect flow. No passwords are stored in the UI. After sign-in the app opens on Schedules.
Layout

The sidebar has three groups:
| Group | Entries |
|---|---|
| Execution | Schedules, Jobs, Datasets, Actions |
| Configuration | Source Systems, Export Systems, Data access, Variables, Templates |
| Setup | Environments, Access, Runtime, About |
Two entries appear only when the feature is available: Environments on plans that allow more than one environment, and Data access once the Data API is switched on. The bottom of the sidebar shows the signed-in user and a Logout icon. An Enterprise install can replace the logo and primary colour under Setup → About → Branding.
The environment band is a sticky bar across the top of the page that names the environment you are working in, in its colour. It appears as soon as the install has a second environment, and is hidden on the Setup pages, which always show the default environment. See Environments.
On a phone the sidebar becomes a drawer with Schedules, Jobs, Datasets and About only; configuration pages redirect to Schedules.
Conventions used throughout
- Edit dialogs stay open on Save. The primary button reads Save while there are unsaved changes and Close once everything is saved, so you can keep working in the same dialog. Clicking outside a dialog, or pressing Escape, does nothing while it has unsaved changes. Leaving a form page through the sidebar asks before discarding edits.
- Status switches. On the Source Systems page the Status column of a source system, an instance or a dataset is its enabled switch. A new source system and a new instance start disabled; switch them on before the first refresh.
- Danger zones. Destructive actions (delete a source system, an environment) sit in a red Danger zone at the bottom of the edit dialog, behind a tick box that spells out what is destroyed, and then a confirmation dialog. There is no undo.
- Info accordions. Most list pages open with a collapsed "About …" panel that explains the page.
- Grids. List pages are data grids with sorting and filtering; the sort and filter you set are remembered in the browser.
- Variables. Typing
{{in any field that accepts variables opens a picker. Known references are drawn as blue chips, unknown ones as orange chips you can click to create the variable. See Variables. - Copy API trigger URL. Source systems, instances, datasets, jobs, schedules and actions carry a link icon that opens a popover with a ready-made API call for that object. See Triggering Runs from Outside.