Connections
A connection is a saved, reusable record that tells Qinfinite how to reach an external system. It stores everything the platform needs — the address, the credentials, and the extra settings — so your automations can use it without you having to type those details again.
Think of a connection like a trusted contact card for a system. It knows where the system lives, how to sign in, and which security settings apply. Every automation that needs that system simply references the card by name.
What a Connection Contains
| Field | What It Means |
|---|---|
| Connection Name | A unique, friendly name for this connection. Must be 4 to 30 characters, using letters, numbers, underscores, and hyphens |
| Type | The specific system type, such as PostgreSQL, Kafka, Salesforce, or OpenAI. Drives which fields appear in the form |
| Category | The logical group the type belongs to, such as Relational Databases or Cloud Providers |
| Mandatory Fields | Required values for this connection type — for example, host, port, username, and password for a database |
| Advanced Fields | Optional settings such as SSL mode, connection timeouts, and pool size. Collapsed by default so they do not get in the way |
| Tags | Labels you add to help filter and search connections |
| Description | A free-text note explaining what this connection is for, who owns it, or what environment it belongs to |
| Testable | Whether this connection type supports a live connectivity test |
| Last Test Status | The result of the most recent test: Success, Failure, or Never Tested |
| Last Tested On | When the connection was last tested |
| Last Test Error | The error message returned if the last test failed |
| Created By / Modified By | Who created and last updated this connection, with timestamps |
Connection Categories
Qinfinite supports over 100 connection types organised into 15 categories. Every category is searchable and displayed in an accordion-style picker with a three-column grid of icons.
| Category | Example Types |
|---|---|
| Relational Databases | PostgreSQL, MySQL, Microsoft SQL Server, Aurora, Azure SQL, MariaDB, Google Cloud SQL, IBM Db2, SAP HANA |
| NoSQL Databases | MongoDB, Redis, DynamoDB, Cassandra, Neo4j, CouchDB, Couchbase, Amazon Neptune |
| Data Warehousing and Analytics | Snowflake, BigQuery, Redshift, Databricks |
| Object Storage | Amazon S3, Azure Blob Storage, Google Cloud Storage |
| Communication Platforms | Gmail, Outlook, Exchange, SMTP, SendGrid, Slack, Microsoft Teams, Discord, Zoom, WhatsApp, Telegram, Twilio SMS |
| Protocols | SFTP, FTP, FTPS, SCP, SSH, NFS, SMB, WebDAV, WebSocket |
| Enterprise Applications | Salesforce, ServiceNow, Microsoft 365, SAP, Jira, Zendesk, HubSpot, Confluence, Workday, NetSuite, Shopify, Notion |
| Messaging and Event Systems | Apache Kafka, RabbitMQ, Amazon SQS, Amazon SNS, Azure Service Bus, Google Pub/Sub, IBM MQ, ActiveMQ |
| Version Control Systems | GitHub, GitLab, Azure Repos, Bitbucket, SVN, Perforce |
| Cloud Providers | AWS, Microsoft Azure, Google Cloud, IBM Cloud, Oracle Cloud, Alibaba Cloud, DigitalOcean |
| Identity and Security | Okta, Auth0, Keycloak, HashiCorp Vault, CyberArk, LDAP |
| DevOps | Jenkins, CircleCI, GitHub Actions, GitLab CI, Azure DevOps |
| AI Platforms | OpenAI, Azure OpenAI, Anthropic Claude, Amazon Bedrock, Google Gemini, HuggingFace, IBM Watson, Vertex AI, Cohere |
| Analytics and BI | Tableau, Power BI, Looker, Datadog, Elastic Stack, Splunk |
| Monitoring | Prometheus, InfluxDB |
Use the search box in the connection type picker to find your system instantly without scrolling through categories.
How to Create a Connection
How Fields Change by Type
The form is fully dynamic. When you select a connection type, the platform loads the correct template for that system and displays only the fields that are relevant.
Mandatory fields are always visible and must be filled in before you can save. They represent the minimum information the platform needs to establish a connection — for example, host and port for a database, or tenant ID and client secret for Microsoft 365.
Advanced fields are hidden inside a collapsible panel. They cover optional settings such as SSL certificates, connection pool sizes, keep-alive intervals, and proxy configuration. Most users will never need to change these defaults.
Field types vary by system. Some fields are plain text boxes, some are number inputs, and some are dropdowns populated dynamically from the platform. Password-type fields always include a show/hide toggle, and their values are automatically encrypted when you save.
You can also reference a Global Variable inside any field. This is useful when you want a field value to come from a centrally managed variable rather than being stored directly in the connection — for example, using a variable to hold an API base URL that differs between environments.
The mandatory and advanced field definitions come from the connection template for each type. If you do not see a field you expect, check whether it has been moved to the Advanced Fields section.
File-Reference Resolution
Some connection fields no longer hold a secret pasted inline — they reference a stored File instead. Fields such as an SSH private key or an SSL/TLS certificate are chosen from a dropdown that lists Files you have uploaded to Global Configurator, rather than typed into the form.
This keeps large or sensitive key material managed in one place (the Files section), where it is versioned and access-controlled, and lets several connections reference the same key without duplicating it.
How it resolves at runtime. The connection stores only a reference to the selected File. When an internal caller (a plugin or connector) needs the actual key or certificate content, it requests the connection with resolution enabled, and the platform expands each file reference into the real file content — decrypted and ready to use — just in time. The stored connection never contains the raw bytes; they are assembled only when needed.
Testing Connections
Most connection types support a live connectivity test. When you click Test Connection, Qinfinite attempts to reach the system using the details you have provided and returns one of three results:
| Result | What It Means |
|---|---|
| Success | Qinfinite connected to the system successfully. The connection is ready to use |
| Failure | The connection attempt failed. The error message is stored on the record so you can diagnose the problem |
| Never Tested | The connection has not been tested yet. It can still be saved and used, but its reliability is unknown |
The test result, timestamp, and any error message are saved with the connection so team members can see the health of each connection at a glance.
Always test a connection before using it in a production automation. A test that fails with a clear error message is much easier to fix than a workflow failure at midnight.
Duplicate Detection
Before saving, Qinfinite checks whether a connection with similar identifying details already exists. For most connection types, uniqueness is determined by fields like host and port for databases, or by the tenant ID and mailbox for email systems.
If a potential duplicate is found, the platform warns you and shows the existing connection. You can choose to continue creating a new one or go back and use the existing record.
This protects against accidentally creating two connections for the same system under different names — a common source of confusion in larger teams.
Security
Connection security is handled in two ways: automatic encryption and role-based access.
Automatic encryption The platform detects which fields in each connection template are passwords or secrets. When you save a connection, those fields are automatically encrypted using HashiCorp Vault's Transit secrets engine before they reach the database. A password is never stored in plain text — not in the database, not in logs, not in exports.
When you open a connection to edit it, password fields are decrypted on the way back to the form and displayed with a show/hide toggle. If you do not change a password field, the existing encrypted value is preserved.
Role-based access
Permissions follow the format GlobalConfigurator:Connections:Action. An administrator can grant view-only access so a developer can reference connections in workflows without being able to see credentials or delete records.
Soft delete Deleting a connection does not permanently remove it immediately. The record is soft-deleted, which means it is hidden from normal views but can be recovered if needed.
Never store a plain-text password in a description, tag, or connection name. These fields are not encrypted. Use the designated password or secret fields, which are encrypted automatically.
Good Naming Practices
A clear connection name makes it obvious which system, environment, and purpose the connection serves. This matters when a team has dozens of connections and someone needs to pick the right one quickly.
| Less Clear | Better |
|---|---|
db1 | Production_Finance_PostgreSQL |
api | Salesforce_Production_API |
mail | Finance_SMTP_Outbound |
sftp | Vendor_Invoice_SFTP |
aws | Production_S3_Reporting |
kafka | Customer_Events_Kafka |
teams | Operations_Microsoft_Teams |
openai | GPT4_Production_API |
Include the environment (Production, QA, Staging) in the connection name. It prevents automations from accidentally using the wrong system.
Tips and Reminders
You can use a Global Variable inside any connection field. This is useful for values that change between environments, such as API base URLs or port numbers, while keeping the credential fields locked to the connection record itself.
Connections support import and export. You can export all connections from your QA environment and import them into Production, with name conflicts resolved automatically by appending a timestamp. See Import and Export for details.
If you delete a connection that is actively referenced by running automations, those automations will fail at the step that tries to use it. Check whether a connection is in use before removing it.
Related Pages
- Overview — where connections fit in the Global Configurator picture
- Global Variables — reusable values that can be referenced inside connection fields
- Import and Export — moving connections between Dev, QA, Staging, and Production