What the data for this category looks like
Core data sources for this category include China’s Disease Prevention and Control Information System, hospital HIS systems, infectious disease reporting systems, and laboratory testing systems. Data updates occur daily. Documents processed daily include infectious disease reporting cards, epidemic analysis reports, case correction records, clustered outbreak reports, and epidemiological investigation reports. Each document type includes specific fields such as statutory infectious disease categories, case diagnosis time, reporting institution code, and test result values. All content must strictly match format requirements specified by statutory reporting rules.
What constraints these characteristics impose on workflow orchestration
Multi-source heterogeneous data sources require workflows to configure multiple nodes to access data from different systems. This prevents validation failures caused by inconsistent data formats. Daily data update frequency requires workflows to balance both batch daily processing and real-time response to abnormal data. Diverse document types and specific fields require workflows to use flexible variable mapping to adapt to different document structures. Unlike daily hospital clinical management, processing must strictly follow statutory infectious disease reporting rules. Any field validation or process deviation may lead to underreporting, delayed reporting, delayed epidemic alerts, and similar consequences. Complete exception capture and retry mechanisms must be configured.
How to set configurations
Reference only; verify against the target FastGPT release and deployment documentation.
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
| Trigger Mode | Scheduled daily trigger + abnormal data trigger | Matches daily data update frequency, and covers abnormal cases reported temporarily |
| Variable Mapping Rules | Bind each data source field via variables | Adapts to different field structures of multi-source heterogeneous documents, and uniformly matches policy validation requirements |
| Exception Capture Node | Bind exception node nodeId, retry interval 60 seconds, retry count 3 times | Covers temporary system fluctuations, and avoids frequent retries consuming system resources |
| Policy Knowledge Base Association | Bind statutory infectious disease policy library, recall top 5 entries per query | Accurately matches policy clauses, and avoids redundant information interfering with judgment |
| Code Execution Dependencies | Set dependency installation paths based on actual testing | Adapts to container directory structures of different deployment environments, and ensures library files load correctly |
Three common configuration mistakes
- An upstream boolean node in a workflow shows true, but the downstream judge always returns false. The cause is failure to correctly bind the upstream node’s validation result to the judge node’s input parameters via
variables. This causes the judge node to read empty or incorrectly formatted parameters. - After connecting a workflow tool call module to a knowledge base, knowledge base content cannot be extracted. The cause is failure to correctly associate the knowledge base node’s output parameters with the context input items of the tool call node, and failure to specify
nodeIdpointing to the configured knowledge base node. - Third-party libraries imported by a workflow code execution module fail to load properly. The cause is failure to declare dependency libraries in the workflow’s environment configuration, or failure to install dependencies to the public dependency directory of the FastGPT container. This causes the code execution environment to fail to recognize the imported libraries.
How to confirm the configuration is complete
- Manually trigger the workflow once. Check the output logs of each node to confirm that field mapping values in
variablesmatch source documents. - Simulate abnormal data such as timeout messages or duplicate report data. Verify whether the exception capture node triggers the preset retry or alert process.
- Check the dependency declaration of the code execution module. Confirm that imported libraries have been installed in the container environment.
- Retrieve the workflow’s running records. Confirm that the trigger mode meets the configuration requirements for daily scheduling and abnormal triggering.
The values provided are common starting points and should be evaluated against the reader's own samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-09-28.