MCP Prometheus MCP Server

MCP Prometheus 📈

Prometheus 기반 모니터링용 MCP 서버입니다. 엔트리포인트는 main.py입니다.

Quick Start 🚀

cd d:\MCPTools
uv sync
uv run python mcp_prometheus/main.py

프로젝트 구조 🧩

mcp_prometheus/
  main.py
  core/
    config.py
    runtime.py
    server.py
    time_utils.py
  domain/
    checks.py
  infra/
    prom_client.py
  tools/
    catalog.py
    alerts_runner.py
    checks_runner.py
    promql.py
  utils/
    query_utils.py
    summarize.py

Tools 요약 🛠️

`run_check` 입력 가이드 🧭

필수

• check_id

기간

• 상대: hours, minutes, days
• 절대: start_time_utc_iso, end_time_utc_iso
• 종료 오프셋: end_offset_minutes, end_offset_hours, end_offset_days

타겟 필터

• server_name
• instance (예: host-or-ip:9100)

필터 규칙:

• server_name와 instance를 함께 주면 AND 적용
• 하나만 주면 해당 라벨만 적용

`run_promql` 가드레일 🔒

• approved=False: 실행하지 않고 확인 메시지 반환
• approved=True: 실행

모드:

• instant=True -> /api/v1/query
• instant=False -> /api/v1/query_range

사용 예시 📌

1) 특정 서버 CPU 평균 (최근 24시간)

{
  "check_id": "cpu_avg_pct",
  "hours": 24,
  "instance": "10.23.12.11:9100",
  "environment": "prod"
}

2) 특정 서버 디스크 사용률 (mountpoint별)

{
  "check_id": "disk_used_pct_by_mount",
  "hours": 24,
  "server_name": "CMS AP #1",
  "environment": "prod"
}

3) 사용자 PromQL 실행 (instant)

{
  "promql": "up",
  "approved": true,
  "instant": true,
  "environment": "prod"
}

CHECKS Catalog ✅

Source: domain/checks.py (CHECKS)

System / Resource

• cpu_avg_pct: CPU average usage (%) by instance/server_name
• cpu_peak_pct: window peak CPU usage (%) over selected range
• mem_used_pct: memory used ratio (%)
• mem_swap_used_pct: swap used ratio (%)
• load15_avg: 15-minute load average
• cpu_iowait_pct: CPU iowait ratio (%)

Disk / Filesystem

• disk_used_pct_by_mount: filesystem used (%) by mountpoint/device (0-100 scale)
• disk_used_top5_pct: top 5 filesystem usage (%)
• disk_inodes_used_pct: inode usage (%)
• fs_readonly: readonly filesystem indicator (1=readonly)
• disk_io_busy_pct: disk I/O busy ratio (%)

Availability

• up: target liveness (1=up, 0=down)

Network / TCP

• net_in_bytes: inbound throughput (bytes/sec)
• net_out_bytes: outbound throughput (bytes/sec)
• net_errs_per_sec: RX+TX network errors per second
• tcp_retrans_per_sec: TCP retransmit segments per second
• tcp_established: established TCP connections
• tcp_time_wait: TIME_WAIT TCP sockets
• tcp_inuse: in-use TCP sockets
• tcp_orphan: orphan TCP sockets

Process Monitoring

• proc_cpu_pct: process group CPU usage (%)
• proc_mem_bytes: process group memory usage (bytes)
• proc_count: process group process count

PostgreSQL

• pg_up: PostgreSQL exporter up state (1=up, 0=down)
• pg_qps: PostgreSQL transactions/sec (commit + rollback)
• pg_cache_hit_pct: PostgreSQL buffer cache hit ratio (%)
• pg_active_conn: active PostgreSQL connections

환경 변수 요약 ⚙️

PROM_ENV_URLS={"prod":"http://...:9090","dev_test":"http://...:9090","dr":"http://...:9090"}
PROM_URL=http://...:9090
PROM_BEARER_TOKEN=
PROM_TIMEOUT_SEC=15

ALERT_WARN_PCT=85
ALERT_CRIT_PCT=95
ALERT_SUSTAIN_MINUTES=5

PROM_MAX_SAMPLES_PER_SERIES=5000
PROM_MAX_PARALLEL_CHECKS=6

환경 선택 우선순위:

1. environment
2. env_hint
3. PROM_URL fallback

운영 팁 💡

• 리포트 출력 시 % 단위를 명확히 표기하세요.
• 단일 서버 점검은 instance 또는 server_name 필터를 사용하세요.
• disk_used_pct_by_mount 값은 0~100 스케일입니다. (0.8 = 0.8%)