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

# 配置
GG_HOME=”/oracle/goldengate”
LOG_FILE=”$GG_HOME/troubleshoot/gg_troubleshoot_$(date +%Y%m%d).log”

# 创建故障排除目录
mkdir -p $GG_HOME/troubleshoot

# 记录日志
log_message() {
echo “$(date +”%Y-%m-%d %H:%M:%S”): $1″ >> $LOG_FILE
}

# 检查进程状态
check_processes() {
log_message “检查GoldenGate进程状态…”

PROCESS_STATUS=$($GG_HOME/ggsci << EOF info all EOF ) log_message "进程状态: $PROCESS_STATUS" # 检查是否有进程异常 if echo "$PROCESS_STATUS" | grep -E "ABENDED|STOPPED" > /dev/null; then
log_message “发现异常进程状态”
# 查看异常进程的详细信息
for PROCESS in $(echo “$PROCESS_STATUS” | grep -E “ABENDED|STOPPED” | awk ‘{print $3}’); do
log_message “查看进程 $PROCESS 的报告…”
REPORT=$($GG_HOME/ggsci << EOF view report $PROCESS EOF ) log_message "进程 $PROCESS 报告: $REPORT" done fi } # 检查复制延迟 check_lag() { log_message "检查复制延迟..." # 检查所有Extract进程 EXTRACTS=$($GG_HOME/ggsci << EOF info extract * EOF ) log_message "Extract进程: $EXTRACTS" # 检查所有Replicat进程 REPLICATS=$($GG_HOME/ggsci << EOF info replicat * EOF ) log_message "Replicat进程: $REPLICATS" } # 检查Trail文件 check_trail_files() { log_message "检查Trail文件..." TRAIL_FILES=$(ls -la $GG_HOME/dirdat/) log_message "Trail文件: $TRAIL_FILES" # 检查Trail文件大小 LARGE_TRAIL_FILES=$(find $GG_HOME/dirdat/ -name "*.r*" -size +1G) if [ -n "$LARGE_TRAIL_FILES" ]; then log_message "发现大于1GB的Trail文件: $LARGE_TRAIL_FILES" fi # 检查Trail文件数量 TRAIL_COUNT=$(ls -la $GG_HOME/dirdat/ | wc -l) if [ "$TRAIL_COUNT" -gt 100 ]; then log_message "Trail文件数量过多: $TRAIL_COUNT" fi } # 检查系统资源 check_system_resources() { log_message "检查系统资源..." # 检查CPU使用情况 CPU_USAGE=$(top -b -n 1 | grep "Cpu(s)") log_message "CPU使用情况: $CPU_USAGE" # 检查内存使用情况 MEM_USAGE=$(free -m) log_message "内存使用情况: $MEM_USAGE" # 检查存储空间 DISK_USAGE=$(df -h) log_message "存储空间使用情况: $DISK_USAGE" # 检查网络连接 NETWORK_STATUS=$(ping -c 3 fgedu-target.net.cn 2>&1)
log_message “网络连接状态: $NETWORK_STATUS”
}

# 主函数
main() {
log_message “开始GoldenGate故障排除”
check_processes
check_lag
check_trail_files
check_system_resources
log_message “GoldenGate故障排除完成”
}

# 执行主函数
main

Part04-生产案例与实战讲解

4.1 进程异常终止故障排除