Quick Start

Through this quick start guide, you can complete the first end-to-end workflow in BlueKing Lite CMDB within 30 minutes: define an asset model, enter or import instances, establish relationships, verify results through views and search, and then use auto discovery to continuously supplement data. The entire process requires no secondary development and is suitable for teams to first establish an asset foundation that is "usable, searchable, and continuously updated."

---

1. Step-by-Step Guide

Step 1: Create Your First Asset Model

BlueKing Lite CMDB uses a "model-driven" design. The first step is to define asset standards.

1.1 Create a Model Category

1. Navigate to the Asset Model Management page
2. Click the "+Category" button to create your first category
3. Enter the category name (e.g., "Host Resources," "Container Platform")

1.2 Create an Asset Model

1. Click "+Model" under the target category
2. Configure the model basic information:
   • Model ID: English characters, e.g., linux_host
   • Model Name: Display name, e.g., "Linux Host"
   • Icon: Select an icon matching the asset type

1.3 Configure Model Attributes

Enter the model detail page and configure fields on the "Attributes" tab:

Recommended Field	Field Type	Constraint Suggestion	Description
ip_addr	String	Required, Unique	Primary identifier for association and search
status	Enumeration	Required	Running status, referencing the public option library
environment	Enumeration	Required	Environment type (Production/Testing/Development)
owner	User	Editable	Asset owner
tags	Tag	-	Flexible labeling, supports multi-dimensional search

1.4 Set Up Field Groups (Optional)

1. Switch to the "Field Groups" tab
2. Create groups such as "Basic Information," "Network Configuration," and "Operations Information"
3. Drag and drop fields into the corresponding groups and adjust the display order

1.5 Define Model Relationships

1. Switch to the "Relationships" tab
2. Click "+Add Relationship" and configure the relationship type:
   • Source Model: Current model
   • Target Model: Related model (e.g., "Business System," "Network Device")
   • Relationship Type: Select "Runs on," "Depends on," "Contains," etc.
   • Mapping Constraint: Set 1:1, 1:n, n:n, etc.

UI Guide:

• Diagram explanation: The Model Management page uses a "Category - Model - Detail" three-level navigation. The left category tree supports drag-and-drop sorting, center model cards display instance statistics, and the right detail page organizes attributes, groups, relationships, and other configurations through tabs.

Efficiency Tip

To create similar models, use the Model Duplication feature:

1. Click the "Duplicate" button on an existing model
2. Select to duplicate "Attributes," "Field Groups," and "Relationships"
3. The new model inherits all configurations and only needs minor adjustments

---

Step 2: Enter or Import Your First Batch of Instances

Once the model is ready, start entering actual asset data.

Method A: Single Entry (Suitable for Validation Phase)

1. Navigate to the Asset Data page
2. Select the target model from the left panel
3. Click the "+New" button
4. Fill in attribute values in the form and submit

Method B: Batch Import (Suitable for Legacy Migration)

1. Click the "Import" button on the asset list page
2. Download the import template (Excel format)
3. Organize existing data according to the template format:
   • First row contains field names (corresponding to model attribute IDs)
   • One instance record per row
   • Enumeration fields should contain option IDs
4. Upload the file; after system validation, confirm the import

Note

For the first import, it is recommended to validate with 5-10 sample records:

• Verify field mappings are correct
• Check for uniqueness constraint conflicts
• Confirm enumeration values match

Proceed with large batch imports only after successful validation.

Efficiency Tip

When performing batch instance data import, you can simultaneously plan and enter relationships between instances. For assets that already have clear upstream/downstream, deployment, or ownership relationships, it is recommended to supplement relationship data immediately after importing instances, reducing the repetitive work of building relationships one by one later and forming a usable topology structure faster.

UI Guide:

• Diagram explanation: The asset list page has a filter bar at the top supporting attribute-based condition filtering; the left panel shows a model navigation tree; the center table displays instance data; the top-right corner provides import, export, and batch operation entry points.

---

Step 3: Establish Instance Relationships and View Topology

The core value of CMDB lies in expressing dependency structures between assets.

3.1 Create Instance Associations

1. Click a target instance in the asset list to enter the detail page
2. Switch to the "Relationships" tab
3. Click "+Add Association"
4. Select the relationship type and target instance
5. Confirm and save

Typical association scenario examples:

Source Instance	Relationship Type	Target Instance	Business Meaning
MySQL Instance	Runs on	Linux Host	Database deployment location
Business System	Depends on	MySQL Instance	Business dependency
Switch	Connected to	Router	Network topology

3.2 View Topology

1. Click "Topology View" on the instance detail page
2. Expand associated nodes upstream and downstream from the current instance
3. Click "+" to expand more levels; click a node to view details

UI Guide:

• Diagram explanation: The topology view uses a force-directed graph layout, with node size reflecting the number of associations and line colors distinguishing relationship types. Supports box selection, drag-and-drop, zoom, and double-click on nodes to expand the next level of associations.

---

Step 4: Configure Auto Discovery Tasks

Upgrade the CMDB from "one-time entry" to "continuous synchronization."

4.1 Select Collection Objects

1. Navigate to the Auto Discovery page
2. Select the target type from the "Collection Object Tree" on the left
3. Supported collection objects include:
   • Containers: K8s clusters, Docker
   • Virtualization: VMware vCenter
   • Network Devices: Switches, routers (SNMP)
   • Databases: MySQL, PostgreSQL, Redis
   • Cloud Platforms: Alibaba Cloud, Tencent Cloud
   • Hosts: Linux/Windows hosts
   • Middleware: Nginx, Kafka, Zookeeper, etc.

4.2 Configure Collection Task

Click "+New Task" and fill in the configuration:

Configuration Item	Description	Recommendation
Task Name	An easily identifiable name	Include object type and environment, e.g., "Production K8s Collection"
Driver Type	Protocol / Job (Script)	Select Protocol for cloud resources and containers, Job for hosts
Access Point	Collection proxy or gateway address	Select an access point connected to the target network
Collection Scope	IP range or instance list	Use a small scope for initial testing
Credentials	Access account and password	Use a minimal-permission read-only account
Timeout	Per-object collection timeout	Increase for poor network environments
Execution Cycle	Whether to execute periodically	Recommended for dynamic environments, e.g., every 6 hours
Write Mode	Direct write / Approval before write	Recommended: "Approval before write" for production

4.3 Execute Task and View Results

1. After saving the task, click "Execute" to trigger the first collection
2. Wait for the task to complete, then enter the task detail page
3. View the "Collection Summary":
   • Added: Newly discovered instances
   • Updated: Instances with changed attributes
   • Deleted: Instances that no longer exist (soft delete marker)
   • Associated: Automatically discovered relationships
   • Exceptions: Failed instances and reasons

UI Guide:

• Diagram explanation: The Auto Discovery page has a collection object tree on the left with different icons for different object types. The right panel shows the task list with execution status. Click a task name to enter details. The detail page displays the collection summary in card format, with expandable exception items for detailed error viewing.

---

Step 5: Verify Search and View Features

Confirm that asset data can be quickly retrieved and browsed.

5.1 Asset View Verification

1. Navigate to the Asset View homepage
2. Confirm that categories and models are displayed correctly
3. Verify that instance count statistics are accurate
4. Click a model card to verify navigation to the corresponding list

5.2 Full-Text Search Verification

1. Enter an instance name or IP in the top search box
2. View search results grouped by model with statistics
3. Click an instance to enter the detail page and verify information completeness
4. Try toggling the "Case Sensitive" option and observe result changes

UI Guide:

• Diagram explanation: Full-text search uses a "Model Aggregation + Instance List" dual-layer display. The left panel shows the number of matches per model, and the right panel shows specific instances. Supports paginated browsing; click an instance title to jump directly to the detail page.

---

Step 6: Configure Data Subscriptions

Enable proactive "data finds people" notification capabilities.

1. Navigate to the Data Subscription page
2. Click "+Subscription Rule"
3. Configure the rule:
   • Target Model: Select the asset type of interest
   • Filter Method: Filter by conditions or directly specify instances
   • Trigger Condition: Attribute change / Relationship change / Expiration reminder
   • Recipients: Select notification recipients
   • Notification Channel: In-app message / WeCom / DingTalk
4. Enable the rule

---

2. Result Verification and Closure

Verification Checklist

Verification Item	Verification Method	Expected Result
Model configuration is correct	Create one instance and check field display	Field groups, enumeration options, and required validation all match the design
Search function works	Search by IP, name, and tags	Accurate results returned in seconds
Topology relationships are accurate	Expand instance topology and check associations	Upstream/downstream dependencies match the actual architecture
Auto discovery is effective	Review collection summary	Added/updated counts match expectations, no error reports
Changes are traceable	Modify instance attributes and view change records	Can see operator, timestamp, and before/after value comparison

3. Common Issues and Troubleshooting

Symptom	Possible Cause	Solution
Import failed	Field mapping errors, enumeration value mismatch	Check that template fields match model attribute IDs, confirm enumeration option IDs are correct
Collection task exception	Network unreachable, incorrect credentials, insufficient permissions	Check access point connectivity, verify credential validity, confirm account permissions
Topology not displaying	No model relationships defined, no instance associations	First define relationship types at the model level, then create instance associations
Search returns empty	Index not updated, permission isolation	Wait for index refresh (~1 minute), check organization permission scope
Subscription not triggering	Rule not enabled, instances don't match filter conditions	Confirm rule status is "Enabled," check that filter conditions include target instances

---

4. Next Steps

After completing the quick start, consider exploring the following advanced capabilities:

1. Public Option Library: Unified maintenance of enumeration values for status, environment, etc., reusable across models
2. Model Duplication: Quickly duplicate standard models to multiple business domains
3. Advanced Search: Use combined condition filtering and save frequently used queries
4. API Integration: Connect CMDB data to external systems via OpenAPI
5. Dashboard Display: Build visual operations dashboards based on asset data

For more help, refer to the "Feature Guide" documentation or contact your platform administrator.