#!/bin/bash
# send_wechat.sh
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn

WEBHOOK_URL=”https://qyapi.weixin.qq.com/cgi-bin/webhook/send?key=your_key_here”

MESSAGE=$1

curl -s -X POST “${WEBHOOK_URL}” \
-H “Content-Type: application/json” \
-d “{
\”msgtype\”: \”markdown\”,
\”markdown\”: {
\”content\”: \”${MESSAGE}\”
}
}”

# 步骤2：赋予执行权限
$ chmod +x /GoldenGate/app/scripts/send_wechat.sh

# 步骤3：测试企业微信告警
$ /GoldenGate/app/scripts/send_wechat.sh “## OGG告警通知\n> 进程状态：异常\n> 延迟：5分钟\n> 时间：$(date)”

{“errcode”:0,”errmsg”:”ok”}

# 步骤4：集成到监控脚本
# 在ogg_monitor.sh中添加企业微信通知函数
send_wechat() {
local message=”$1″
/GoldenGate/app/scripts/send_wechat.sh “$message”
log “企业微信已发送: $message”
}

# 监控告警架构
1. Manager进程事件监控
2. 自定义脚本定时监控
3. 企业微信实时通知
4. 短信紧急告警
5. 邮件详细报告

# 监控配置
# Manager配置
PORT 7809
PURGEOLDEXTRACTS /GoldenGate/app/dirdat/*, USECHECKPOINTS, MINKEEP 24 HOURS

— 邮件配置
MAILPROG /usr/sbin/sendmail
MAILALERTS ALL
MAILOPTIONS FROM “ogg-alert@bank.com” TO “dba-team@bank.com”

— 事件告警
EVENTEXTRACTSTOPPED SHELL “/GoldenGate/app/scripts/alert.sh EXTRACT_STOPPED E_FGEDU01”
EVENTEXTRACTABENDED SHELL “/GoldenGate/app/scripts/alert.sh EXTRACT_ABENDED E_FGEDU01”
EVENTREPLICATSTOPPED SHELL “/GoldenGate/app/scripts/alert.sh REPLICAT_STOPPED R_FGEDU01”
EVENTREPLICATABENDED SHELL “/GoldenGate/app/scripts/alert.sh REPLICAT_ABENDED R_FGEDU01″

# 告警脚本
#!/bin/bash
# alert.sh
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn

ALERT_TYPE=$1
PROCESS_NAME=$2
TIMESTAMP=$(date ‘+%Y-%m-%d %H:%M:%S’)

case $ALERT_TYPE in
EXTRACT_STOPPED)
MESSAGE=”OGG告警：Extract进程${PROCESS_NAME}已停止，时间：${TIMESTAMP}”
;;
EXTRACT_ABENDED)
MESSAGE=”OGG严重告警：Extract进程${PROCESS_NAME}异常终止，时间：${TIMESTAMP}”
;;
REPLICAT_STOPPED)
MESSAGE=”OGG告警：Replicat进程${PROCESS_NAME}已停止，时间：${TIMESTAMP}”
;;
REPLICAT_ABENDED)
MESSAGE=”OGG严重告警：Replicat进程${PROCESS_NAME}异常终止，时间：${TIMESTAMP}”
;;
esac

# 发送企业微信
/GoldenGate/app/scripts/send_wechat.sh “## OGG告警通知\n> ${MESSAGE}”

# 严重告警发送短信
if [[ $ALERT_TYPE == *”ABENDED”* ]]; then
/GoldenGate/app/scripts/send_sms.sh “138****1234” “$MESSAGE”
fi

# 发送邮件
echo “$MESSAGE” | mail -s “OGG告警” dba-team@bank.com

# 监控效果
– 告警响应时间：从30分钟缩短到1分钟
– 故障发现率：从80%提升到99%
– 误报率：控制在5%以下
– 运维效率：提升50%
from GoldenGate视频:www.itpux.com

# 原因分析
邮件服务器配置错误或网络不通

# 解决方案
# 1. 检查邮件服务器连接
$ telnet smtp.fgedu.net.cn 25
Trying 192.168.1.100…
Connected to smtp.fgedu.net.cn.

# 2. 检查sendmail配置
$ cat /etc/mail/sendmail.mc | grep SMART_HOST
define(`SMART_HOST’, `smtp.fgedu.net.cn’)

# 3. 重启sendmail服务
$ systemctl restart sendmail

# 4. 重新测试
GGSCI (fgedu.net.cn)> SEND MGR TESTMAIL
Test mail sent successfully.

# 问题2：告警风暴
# 错误现象
短时间内收到大量重复告警

# 原因分析
未配置告警静默时间

# 解决方案
# 在监控脚本中添加告警静默逻辑
ALERT_LOCK=”/tmp/ogg_alert.lock”
ALERT_INTERVAL=300 # 5分钟

if [ -f “$ALERT_LOCK” ]; then
LOCK_TIME=$(cat $ALERT_LOCK)
CURRENT_TIME=$(date +%s)
DIFF=$((CURRENT_TIME – LOCK_TIME))

if [ $DIFF -lt $ALERT_INTERVAL ]; then
echo “告警静默中，跳过本次告警”
exit 0
fi
fi

# 发送告警后记录时间
date +%s > $ALERT_LOCK

# 问题3：脚本执行权限问题
# 错误现象
$ /GoldenGate/app/scripts/ogg_monitor.sh
Permission denied

# 解决方案
# 检查脚本权限
$ ls -l /GoldenGate/app/scripts/ogg_monitor.sh
-rw-r–r– 1 root root 1234 Apr 10 10:00 ogg_monitor.sh

# 赋予执行权限
$ chmod +x /GoldenGate/app/scripts/*.sh

# 验证权限
$ ls -l /GoldenGate/app/scripts/ogg_monitor.sh
-rwxr-xr-x 1 root root 1234 Apr 10 10:00 ogg_monitor.sh

LOG_DIR=/GoldenGate/app/logs
LOG_FILE=monitor.log
MAX_SIZE=10485760 # 10MB
MAX_FILES=7

cd $LOG_DIR

# 检查日志文件大小
if [ -f “$LOG_FILE” ]; then
SIZE=$(stat -f%z “$LOG_FILE” 2>/dev/null || stat -c%s “$LOG_FILE”)

if [ $SIZE -gt $MAX_SIZE ]; then
# 轮转日志
for i in $(seq $((MAX_FILES-1)) -1 1); do
if [ -f “${LOG_FILE}.$i” ]; then
mv “${LOG_FILE}.$i” “${LOG_FILE}.$((i+1))”
fi
done

mv “$LOG_FILE” “${LOG_FILE}.1”
touch “$LOG_FILE”
fi
fi

# 解决方案2：监控数据采集与展示
# 使用Prometheus采集OGG监控数据

# 创建监控数据导出脚本
$ vi /GoldenGate/app/scripts/prometheus_exporter.sh

#!/bin/bash
# prometheus_exporter.sh
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn

cd /GoldenGate/app

# 获取进程状态
STATUS=$(./ggsci </dev/null | grep -E “EXTRACT|REPLICAT”
INFO ALL
EOF
)

# 输出Prometheus格式指标
echo “# HELP ogg_process_status OGG process status (1=running, 0=stopped)”
echo “# TYPE ogg_process_status gauge”

echo “$STATUS” | while read line; do
PROCESS=$(echo $line | awk ‘{print $3}’)
STATE=$(echo $line | awk ‘{print $2}’)

if [ “$STATE” = “RUNNING” ]; then
echo “ogg_process_status{process=\”$PROCESS\”} 1″
else
echo “ogg_process_status{process=\”$PROCESS\”} 0″
fi
done

# 解决方案3：告警历史记录查询
# 创建告警历史查询脚本
$ vi /GoldenGate/app/scripts/query_alert.sh

#!/bin/bash
# query_alert.sh
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn

ALERT_LOG=/GoldenGate/app/logs/alert.log
DAYS=$1

if [ -z “$DAYS” ]; then
DAYS=7
fi

echo “=== 最近${DAYS}天告警记录 ===”
echo “”

find /GoldenGate/app/logs -name “alert*.log” -mtime -$DAYS -exec cat {} \; | \
sort | uniq -c | sort -rn | head -20

echo “”
echo “=== 告警统计 ===”
echo “总告警数: $(find /GoldenGate/app/logs -name “alert*.log” -mtime -$DAYS -exec cat {} \; | wc -l)”
echo “严重告警: $(find /GoldenGate/app/logs -name “alert*.log” -mtime -$DAYS -exec cat {} \; | grep -c “严重”)”
echo “警告告警: $(find /GoldenGate/app/logs -name “alert*.log” -mtime -$DAYS -exec cat {} \; | grep -c “警告”)”

# 执行查询
$ ./query_alert.sh 7

=== 最近7天告警记录 ===

15 2026-04-10 进程停止: E_FGEDU01
8 2026-04-09 延迟告警: Replicat延迟超过5分钟
5 2026-04-08 磁盘使用率: 85%

=== 告警统计 ===
总告警数: 28
严重告警: 15
警告告警: 13