记一次Oracle数据库归档日志暴增故障案例分析

  1 、故障描述

       某次,在对某用户进行数据库深度检查时,发现归档磁盘存储空间剩余不足 10GB 。如果此时,归档磁盘被打爆,则将导致数据库运行报错甚至宕机,因此需要即刻对该问题进行详细诊断并处理。


 2 、根因分析

对环境进行整体梳理后,情况为数据库归档磁盘存储剩余不足 10GB ,日均产生归档日志量达到 1 00G 左右,主库 RAC 集群对应的 ADG 备库 9 月份开始产生了 GAP 问题,问题表现如下:

a)  日均产生归档日志量达到 1 00G 左右

b)  主库 RAC 集群归档存储剩余空间不足 10GB

c)  主库 RAC 集群对应的 ADG 备库产生了 GAP 问题且时间较久

查看数据库产生归档日志量比较大数据库对象


查看备库有 GAP问题


  3 、解决方案

明确故障原因之后,接下来便是对 ADG进行同步修复操作,具体步骤参考如下:

a)  断开备库 ADG 的实时同步

b)  通过 rman 进入主库删除主库的归档日志释放存储空间

c)  重新部署备库 ADG 并实时应用


    Recovery Manager: Release 11.2.0.4.0 - Production on Thu Nov 16 15:53:17 2023

    

    Copyright (c) 1982, 2011, Oracle and/or its affiliates. All rights reserved.

 

    connected to target database: XXXXXX (DBID=2290076166)

    connected to auxiliary database: XXXXXX (not mounted)

 

    RMAN> 2> 3> 4> 5> 6> 7> 8> 9> 10> 11> 12> 13> 14> 15> 16>

    using target database control file instead of recovery catalog

    allocated channel: c11

    channel c11: SID=2710 instance=xxxxxx1 device type=DISK

 

    allocated channel: c12

    channel c12: SID=2933 instance=xxxxxx1 device type=DISK

 

    allocated channel: c13

    channel c13: SID=2081 instance=xxxxxx1 device type=DISK

 

    allocated channel: c14

    channel c14: SID=1707 instance=xxxxxx1 device type=DISK

 

    allocated channel: c1

    channel c1: SID=1153 device type=DISK

 

    allocated channel: c2

    channel c2: SID=1201 device type=DISK

 

    allocated channel: c3

    channel c3: SID=1249 device type=DISK

 

    allocated channel: c4

    channel c4: SID=1297 device type=DISK

 

    Starting Duplicate Db at 16-NOV-23

 

    contents of Memory Script:

    {

       backup as copy reuse

       targetfile '/u01/app/oracle/product/11.2.0/db_1/dbs/orapwrxxxxxx1' auxiliary format

       '/export/home/oracle11/app/product/11.2.0/Db_1/dbs/orapwrxxxxxxdg';

    }

    executing Memory Script

 

    Starting backup at 16-NOV-23

    Finished backup at 16-NOV-23

    input datafile copy RECID=43 STAMP=1153069415 file name=/Oradata/xxxxxxdg/datafile/ts_bas08.dbf

    datafile 35 switched to datafile copy

    input datafile copy RECID=44 STAMP=1153069415 file name=/Oradata/xxxxxxdg/datafile/ts_gis02.dbf

    datafile 36 switched to datafile copy

    input datafile copy RECID=45 STAMP=1153069415 file name=/Oradata/xxxxxxdg/datafile/ts_gis03.dbf

    datafile 37 switched to datafile copy

    input datafile copy RECID=46 STAMP=1153069415 file name=/Oradata/xxxxxxdg/datafile/ts_gis.309.1114946665

    datafile 38 switched to datafile copy

    input datafile copy RECID=47 STAMP=1153069415 file name=/Oradata/xxxxxxdg/datafile/ts_vect05.dbf

    Finished Duplicate Db at 16-NOV-23

 

    released channel: c1

    released channel: c2

    released channel: c3

    released channel: c4

    released channel: c11

    released channel: c12

    released channel: c13

    released channel: c14

 

    RMAN> Recovery Manager complete.

   

   在进行相关步骤操作后,主备同步正常:

        

    THREAD#  SEQUENCE# FIRST_CHANGE# NEXT_CHANGE# APPLIED

    -------- --------- ------------- ------------ --------

    1        16547     1.7785E+13    1.7785E+13   YES

    2        17425     1.7785E+13    1.7785E+13   YES

    1        16548     1.7785E+13    1.7785E+13   YES

    2        17426     1.7785E+13    1.7785E+13   YES

    ...






请使用浏览器的分享功能分享到微信等