zioinfo-mail/workspace/guardia-itsm/scripts/sm/search/search_elasticsearch_sm.sh
DESKTOP-TKLFCPR\ython cfe2901a55 refactor(structure): consolidate all projects under workspace/
- itsm/    -> workspace/guardia-itsm/
- manager/ -> workspace/guardia-manager/
- app/     -> workspace/guardia-messenger/
- manual/  -> workspace/guardia-docs/

workspace/zioinfo-web/ unchanged.
git mv preserves full commit history.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-31 23:50:56 +09:00

152 lines
6.9 KiB
Bash

#!/bin/bash
# ============================================================
# GUARDiA SM | search_elasticsearch_sm.sh
# 대상: Elasticsearch (7.x / 8.x)
# 파라미터: ES_HOST=localhost ES_PORT=9200 ES_SCHEME=http
# ES_USER=elastic ES_PASS=elastic
# INDEX_HEALTH_WARN=yellow
# ============================================================
set -euo pipefail
ES_HOST=${ES_HOST:-localhost}
ES_PORT=${ES_PORT:-9200}
ES_SCHEME=${ES_SCHEME:-http}
ES_USER=${ES_USER:-elastic}
ES_PASS=${ES_PASS:-""}
ES_BASE="${ES_SCHEME}://${ES_HOST}:${ES_PORT}"
INDEX_HEALTH_WARN=${INDEX_HEALTH_WARN:-yellow}
OK="[OK]"; WARN="[WARN]"; CRIT="[CRIT]"
SEP="─────────────────────────────────────────"
RESULT=0
# curl 헬퍼
_es() {
local URL="$1"
local AUTH=""
[ -n "$ES_PASS" ] && AUTH="-u ${ES_USER}:${ES_PASS}"
curl -sk --max-time 10 $AUTH "${ES_BASE}${URL}" 2>/dev/null
}
echo "======================================================"
echo " GUARDiA SM 점검 | Elasticsearch | $(hostname -s)"
echo " 점검 시각: $(date '+%Y-%m-%d %H:%M:%S %Z')"
echo "======================================================"
# ── 1. 프로세스 ───────────────────────────────────────────
echo; echo "[$SEP] 1. Elasticsearch 프로세스"
ES_PROC=$(pgrep -c "elasticsearch" 2>/dev/null || echo 0)
if [ "$ES_PROC" -gt 0 ]; then
echo " ${OK} Elasticsearch 실행 중 (${ES_PROC}개)"
ES_PID=$(pgrep -f "elasticsearch" | head -1)
RSS_MB=$(awk '/VmRSS/{print $2}' /proc/${ES_PID}/status 2>/dev/null | \
awk '{printf "%d", $1/1024}' || echo "N/A")
echo " RSS 메모리: ${RSS_MB} MB"
else
echo " ${CRIT} Elasticsearch 프로세스 없음"
RESULT=2
fi
# ── 2. 포트 리스닝 ────────────────────────────────────────
echo; echo "[$SEP] 2. 포트 리스닝"
for PORT in $ES_PORT 9300; do
ss -tlnp 2>/dev/null | grep -q ":${PORT} " && \
echo " ${OK} 포트 ${PORT} LISTEN" || \
echo " ${WARN} 포트 ${PORT} LISTEN 없음"
done
# ── 3. 클러스터 상태 ──────────────────────────────────────
echo; echo "[$SEP] 3. 클러스터 상태"
if command -v curl &>/dev/null; then
HEALTH=$(_es "/_cluster/health?pretty")
if [ -n "$HEALTH" ]; then
STATUS=$(echo "$HEALTH" | python3 -c \
"import sys,json; d=json.load(sys.stdin); print(d.get('status','unknown'))" 2>/dev/null)
NODES=$(echo "$HEALTH" | python3 -c \
"import sys,json; d=json.load(sys.stdin); print(d.get('number_of_nodes',0))" 2>/dev/null)
DATA_NODES=$(echo "$HEALTH" | python3 -c \
"import sys,json; d=json.load(sys.stdin); print(d.get('number_of_data_nodes',0))" 2>/dev/null)
SHARDS=$(echo "$HEALTH" | python3 -c \
"import sys,json; d=json.load(sys.stdin); print(d.get('active_shards',0))" 2>/dev/null)
UNASSIGNED=$(echo "$HEALTH" | python3 -c \
"import sys,json; d=json.load(sys.stdin); print(d.get('unassigned_shards',0))" 2>/dev/null)
case "${STATUS,,}" in
green)
echo " ${OK} 클러스터 상태: GREEN"
;;
yellow)
echo " ${WARN} 클러스터 상태: YELLOW (미배정 샤드 있음)"
[ $RESULT -lt 1 ] && RESULT=1
;;
red)
echo " ${CRIT} 클러스터 상태: RED (샤드 손실)"
RESULT=2
;;
esac
echo " 노드 수: ${NODES} (데이터: ${DATA_NODES})"
echo " 활성 샤드: ${SHARDS}, 미배정 샤드: ${UNASSIGNED}"
[ "${UNASSIGNED:-0}" -gt 0 ] && \
echo " ${WARN} 미배정 샤드 ${UNASSIGNED}" && \
[ $RESULT -lt 1 ] && RESULT=1
else
echo " ${CRIT} Elasticsearch 응답 없음 (${ES_BASE})"
RESULT=2
fi
fi
# ── 4. 노드 정보 ──────────────────────────────────────────
echo; echo "[$SEP] 4. 노드 정보"
if command -v curl &>/dev/null; then
NODES_INFO=$(_es "/_cat/nodes?v&h=name,ip,role,heapPercent,ramPercent,cpu,load_1m,node.role")
if [ -n "$NODES_INFO" ]; then
echo "$NODES_INFO" | head -10 | sed 's/^/ /'
# 힙 90% 초과 체크
HEAP_CRIT=$(echo "$NODES_INFO" | awk 'NR>1{if($4+0 >= 90) print $1": "$4"%"}' | head -5)
[ -n "$HEAP_CRIT" ] && echo " ${WARN} 힙 90% 초과 노드:" && echo "$HEAP_CRIT" | sed 's/^/ /' && \
[ $RESULT -lt 1 ] && RESULT=1
fi
fi
# ── 5. 인덱스 상태 (상위 10) ─────────────────────────────
echo; echo "[$SEP] 5. 인덱스 상태 (상위 10)"
if command -v curl &>/dev/null; then
INDICES=$(_es "/_cat/indices?v&s=store.size:desc&h=health,status,index,docs.count,store.size" \
2>/dev/null | head -12)
echo "$INDICES" | sed 's/^/ /'
RED_INDICES=$(echo "$INDICES" | grep "^red" | wc -l || echo 0)
[ "$RED_INDICES" -gt 0 ] && echo " ${CRIT} RED 인덱스: ${RED_INDICES}" && RESULT=2
fi
# ── 6. 응답 시간 ─────────────────────────────────────────
echo; echo "[$SEP] 6. 쿼리 응답 시간"
if command -v curl &>/dev/null; then
RESP_TIME=$(curl -sk -o /dev/null -w "%{time_total}" \
--max-time 5 "${ES_BASE}/_search" 2>/dev/null || echo "N/A")
echo " 기본 검색 응답: ${RESP_TIME}s"
if awk "BEGIN{exit !(\"$RESP_TIME\"+0 > 3.0)}" 2>/dev/null; then
echo " ${WARN} 응답 시간 과다 (${RESP_TIME}s > 3s)"
[ $RESULT -lt 1 ] && RESULT=1
fi
fi
# ── 7. 디스크 사용 (Data path) ────────────────────────────
echo; echo "[$SEP] 7. 디스크 사용"
if command -v curl &>/dev/null; then
DISK=$(_es "/_cat/allocation?v&h=node,disk.used,disk.avail,disk.percent")
echo "$DISK" | sed 's/^/ /'
HIGH=$(echo "$DISK" | awk 'NR>1{if($4+0 >= 85) print $1": "$4"%"}')
[ -n "$HIGH" ] && echo " ${WARN} 디스크 사용률 85% 초과:" && echo "$HIGH" | sed 's/^/ /' && \
[ $RESULT -lt 1 ] && RESULT=1
fi
# ── 요약 ─────────────────────────────────────────────────
echo
echo "======================================================"
case $RESULT in
0) echo " 최종 결과: ${OK} Elasticsearch 정상" ;;
1) echo " 최종 결과: ${WARN} 주의 항목 있음" ;;
2) echo " 최종 결과: ${CRIT} 즉시 조치 필요" ;;
esac
echo " 점검 완료: $(date '+%Y-%m-%d %H:%M:%S')"
echo "======================================================"
exit $RESULT