What this category’s data consists of
This category of data primarily comes from VPC planning documents, security group rule lists, load balancer configuration files, network topology diagrams, and traffic monitoring reports exported from cloud platform consoles. Update frequency aligns with each network change or support ticket. Common document structures include fields such as CIDR network segments, port ranges, health check thresholds, and access control policies. Typical units are IP ranges, seconds, and port numbers. Each document contains multiple nested rule configurations, and formats vary slightly across cloud platforms.
Constraints imposed on document parsing and chunking
Documents contain hierarchically nested network rules and frequently updated change content. Parsing must retain contextual connections between rules. Do not split content that breaks the binding relationship between network segments and ports. Individual documents have many entries and relatively high field standardization. However, configuration formats differ across cloud platforms. Parsing workflows must adapt to structural differences across source documents. Change tickets require parsing to match the latest configuration versions. Timeouts or incorrect chunking directly reduce the accuracy of subsequent compliance reviews.
How to configure settings
This guide is written based on the configuration logic of FastGPT v4.17.0.
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
chunkSize | 800–1200 characters | Matches the length of single rules for security groups and load balancer configurations. Prevents splitting that breaks the binding logic between network segments and ports |
chunkSplitter | Split by paragraph + semantic meaning | Adapts to the hierarchically nested structure of network documents. Retains contextual connections between rules |
PARSE_FILE_TIMEOUT_SECONDS | 300–600 seconds | Adapts to parsing durations for large individual change tickets or topology diagrams. Prevents parsing failures due to timeout |
customPdfParse | Enabled | Adapts to PDF configuration documents exported from cloud consoles. Retains original layout and field structure |
UPLOAD_FILE_MAX_SIZE | 1000 MB | Supports importing large network topology diagrams or batch change ticket documents |
chunkSettingMode | Manual mode | For hierarchical complex network configurations, manually adjust chunk boundaries to adapt to non-standard document structures |
Three common configuration errors
- Uploading a configuration document results in a request failure, and parsing takes longer than 2 minutes. The cause is failure to adjust the
PARSE_FILE_TIMEOUT_SECONDSparameter. The default timeout duration is insufficient to handle parsing for large change tickets or topology diagrams. - Importing a table-formatted configuration list results in empty fields or missing data. The cause is failure to correctly enable document parsing adaptation features. The system fails to identify core fields such as network segments and ports in the table.
- Parsed chunked content has broken binding relationships between network segments and ports. The cause is an overly small
chunkSizevalue, which splits a complete single security group rule or load balancer configuration.
How to verify successful configuration
- Upload a single 100 MB load balancer configuration PDF. Check if parsing completes within the duration set by
PARSE_FILE_TIMEOUT_SECONDS. - Import a table document containing multiple sets of security group rules. Verify that parsed chunks retain complete network segment and port configurations for each rule.
- Confirm that the total size of uploaded files does not exceed the
UPLOAD_FILE_MAX_SIZElimit. Avoid system interception due to oversized files. - Switch
chunkSettingModeto manual mode. Adjust chunk boundaries, then verify that chunked content matches the logical structure of network rules.
The values listed above are common starting points. Always test against your own specific samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-10-19.