Help Center/ Well-Architected Framework/ Well-Architected Framework and Practices/ Operational Excellence Pillar/ Examples/ Gathering Various Device Logs Using LTS for Full-Link Issue Tracking and Service Operations Analysis
Updated on 2025-05-22 GMT+08:00

Gathering Various Device Logs Using LTS for Full-Link Issue Tracking and Service Operations Analysis

A company specializes in IT information transmission, technical exchange, education and training, and professional technical talent services. The company has over 1,000 customers and partners with more than 32 million registered users.

Customer pain points:

  • Unified collection tools on the device side, unavailable custom domain names, and complex fault locating.

Logs on multiple device sides, such as web, iOS, Android, Baidu mini program, and WeChat mini program, cannot be collected using tools of the same vendor. This forces users to switch between multiple tools when locating and analyzing issues, making troubleshooting more complex. Additionally, users cannot customize the server domain name for log reporting, raising compliance concerns among some users.

  • The device logs report slowly and often get lost: Their reporting speed is tracked hourly. This loss impacts both end-to-end issue locating and service integrity analysis.
  • Difficult service mining and analysis: Logs cannot be written to data lake insight (DLI). The data needs to be delivered to Kafka and then consumed by DLI, causing high links and high costs.

Solutions:

Benefits:

  • Comprehensive collection and access of device logs and reporting of custom domain names: The multi-device software development kits (SDK) provided by LTS are integrated to comprehensively collect device logs and access LTS. The reporting of domain name of the server helps ensure service consistency and compliance on the user plane, reducing issue locating complexity and improving O&M efficiency.
  • Millisecond-level reporting of device log data and zero data loss: This feature ensures millisecond-level reporting of device logs with no data loss. It enables full-link tracing and analysis from the frontend to the backend. In addition, service integrity analysis is supported.
  • Easy and low-cost log obtaining, facilitating service mining and analysis: DLI-Flink integrates seamlessly with Connector to fetch logs from LTS in real time, enabling efficient service computing and analysis. Logs can also be stored on OBS for quick access by DLI, supporting offline service computing and analysis.