Monitoring

Logging

ctc can log outgoing RPC requests and SQL queries. This functionality can be enabled or disabled using ctc setup.

Logs are stored in logs subdirectory of the ctc data dir (default = ~/ctc_data).

Running ctc log in the terminal will begin watching for changes to the log files. This provides a detailed view of external queries as they happen, which can be useful for debugging and ensuring that external calls are happening as expected.

Logs are written to disk using a non-blocking queue, making it suitable for async applications and imparting minimal impact on performance. These logs are also rotated once they reach a certain size (default = 10MB). However, being non-blocking also means that the timestamps in the logs lose a bit of temporal precision, and so they do not provide a precise picture of event timing.

Logs are managed by the Loguru package. Loguru must be installed for logging to be enabled (pip install loguru).

Other monitoring

Beyond the built-in logging, the best way to monitor ctc is through standard 3rd party tools.

Recommended utilities for profiling resource usage:

• CPU Usage: htop, btop

• Storage IO: iotop, btop

• Storage Capacity: du, dust, btop

• Network Usage: nethogs, btop

If your situation calls for a more programmatic monitoring approach, then you probably already know what tools you need.