Log Management

The Kato platform displays component logs in the form of real-time push, which is convenient for users to quickly query logs to locate software faults. You can also query historical logs by querying historical log files. Component logs include two categories, operation process logs and component operation output logs.

Component Operation Log

The component overview page presents the component operation history and the log record of each operation, especially the log of the build operation. Please pay attention to the log output of the build operation. When the build fails, please check the log output reminder content to guide the user to the irregularity of the code Improve.

The operation log of the component is also a record of the operation of the component, which is convenient for collaboration and operation review among multiple users.

Component Operation Log Output

After the component runs, it will output to standard output (stdout) and standard error output (stderr) logs will be captured by Kato and aggregated and stored. The way Kato processes component operation logs is to aggregate the logs of multiple instances to the component level, push them to the browser for real-time display, and store them on disk at the same time.

Run log output interface

  • Pause the push: Pause the push of the log, it is convenient to pause the analysis when you see the key information
  • Historical log download: Kato will collect the logs in the last 7 days for storage. Users can download the logs in the last 7 days to the machine for analysis by clicking on the historical log download on the log interface
  • Last 1000 logs: output the last 1000 logs

In the log display box, the user can select the container ID and only query the running log of a certain instance; we try to push the log to the console in real time, but there will be a certain delay due to intermediate processing.

We recommend that users divide component operation logs into access logs and program debug logs. Access logs generally want to be statistically analyzed, so more processing is required. It is recommended to output it to a persistent file, and then connect to other log analysis components for log analysis. The Debug log of the program is directly output, and it is quickly presented to developers to find and locate problems in time.

In Kato, log collection can also be connected to log analysis systems such as ELASTICSEARCH, and logs can be directly transmitted to the analysis system for analysis; for details, please refer to Log docking ELK system.