AgentSkillsCN

server-log-health-check

快速检查 IQRight 服务器管道的健康状况。显示数据包接收情况、查询次数、LoRa 响应，以及 MQTT 发布状态。当您需要检查服务器是否正常运行，或在部署完成后使用此功能。

SKILL.md
--- frontmatter
name: server-log-health-check
description: Quick health check of IQRight server pipeline. Shows packet reception, lookups, LoRa responses, and MQTT publishing status. Use when checking if the server is working or after deploying.

Server Log Health Check

Perform a 4-stage pipeline health check on the CaptureLora.py server logs.

Step 1: Locate and Assess Active Log

The server logs are at logs/IQRight_Daemon.debug with rotated backups at logs/IQRight_Daemon.debug.MMDDYY.

  • •Check file sizes with du -h logs/IQRight_Daemon.debug*
  • •Read the last 50 lines of the active log to see current state
  • •Extract first/last timestamps to show the date range covered

Step 2: Run Pipeline Health Check

Count packets at each of the 4 pipeline stages. Use Grep for each:

  1. •LoRa Reception: pattern Received data from scanner in logs/IQRight_Daemon.debug
  2. •Data Lookups: pattern Using local results|Using API results in logs/IQRight_Daemon.debug
  3. •LoRa Responses: pattern Sent DATA to scanner|Sent CMD in logs/IQRight_Daemon.debug
  4. •MQTT Successes: pattern MQTT-TX.*SUCCESS in logs/IQRight_Daemon.debug
  5. •MQTT Failures: pattern MQTT-TX.*FAILED in logs/IQRight_Daemon.debug

Present these 5 numbers in a clear table.

Step 3: Check Adaptive Logging State

Look for state transitions to understand operational timeline:

  • •SWITCHING TO ACTIVE MODE = operation started (HELLO received)
  • •SWITCHING TO IDLE MODE = operation ended (15 min no packets)
  • •Server started in IDLE mode = server restart

Step 4: Recent Errors

Find the last 15 ERROR lines to identify current issues. Categorize by type:

  • •LoRa errors: "Error in sendDataScanner", "create_packet"
  • •API errors: "Connection reset", "timed out", "Client error"
  • •MQTT errors: "MQTT-TX.*FAILED"
  • •Data errors: "Couldn't find Code", "Invalid DATA packet"

Step 5: Summary Report

Report:

  • •Time range: First and last log entry timestamps
  • •Operational state: IDLE or ACTIVE, last state change
  • •Pipeline status: All 4 stages working? Where are gaps?
  • •Error count: Total errors, top 3 types
  • •Recommendation: Is the server healthy? What needs attention?

Reference: See docs/LOG_ANALYSIS_SKILLS.md for detailed patterns and baselines.