Logging¶

kraken-connector uses Python's stdlib logging module. All loggers are under the kraken_connector namespace.

REST request logging¶

Enable request/response logging via ResilienceConfig:

import logging
from kraken_connector import HTTPClient, ResilienceConfig

logging.basicConfig(level=logging.DEBUG)

config = ResilienceConfig(enable_logging=True, log_level=logging.DEBUG)
client = HTTPClient("https://api.kraken.com", resilience=config)

This logs:

DEBUG kraken_connector: HTTP GET https://api.kraken.com/0/public/Time
DEBUG kraken_connector: HTTP GET https://api.kraken.com/0/public/Time -> 200 (0.12s)

Retry logging¶

When retry is enabled, transient failures are logged at WARNING:

config = ResilienceConfig(max_retries=3, enable_logging=True)
client = HTTPClient("https://api.kraken.com", resilience=config)

WARNING kraken_connector: Retry 1/3 after 0.50s: [ConnectTimeout]

WebSocket logging¶

The WebSocket client logs under kraken_connector.ws:

Event	Level	Example
Connect/disconnect	INFO	`Connecting to wss://ws.kraken.com/v2`
System status	INFO	`System status: online (connection 12345)`
Reconnect attempts	WARNING	`Reconnect 1/10 after 0.50s`
Pong timeout	WARNING	`Pong timeout after 10.0s`
Heartbeat timeout	WARNING	`Heartbeat timeout: 31.2s since last message`
Sequence gap	WARNING	`Sequence gap on executions: expected 5, got 7`
Checksum mismatch	WARNING	`Book checksum mismatch for BTC/USD`
Re-subscribe failures	ERROR	`Re-subscribe failed for ticker: ...`
Max retries exhausted	ERROR	`Max reconnect attempts (10) reached`

Logger names¶

Logger	Source
`kraken_connector`	REST client request/response hooks, retry transport
`kraken_connector.ws`	WebSocket client lifecycle and events

Integration with structlog¶

Since kraken-connector uses stdlib logging, any framework with stdlib integration works:

import structlog

structlog.configure(
    wrapper_class=structlog.stdlib.BoundLogger,
    logger_factory=structlog.stdlib.LoggerFactory(),
)