有关rman备份ora-27192和ora-19511错误

环境描述：

1、数据库环境AIX5.32、数据库版本9.2.0.8.03、备份媒介：磁带库

解决过程

1.⽇常巡检过程中发现数据库备份异常，错误信息栈如下：

RMAN-00571: ===========================================================RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============RMAN-00571: ===========================================================RMAN-03002: failure of backup command at 10/24/2017 21:17:37ORA-27192: skgfcls: sbtclose2 returned error - failed to close fileORA-19511: Error received from media manager layer, error text: ANS1301E (RC1) Server detected system error

2.经过查阅资料得知ora-19511属于媒介管理错误，需要有媒介管理⽀持团队或供应商解决。

3.既然发现这个问题之后，那么就需要判断是否真的是因为媒介的问题，运⾏sbttest以测试Oracle是否可以与磁带库通信时，您可能会遇到以下错误：

[wsrz1]@[BJS5951_L2][/oracle/product/9.2.0/bin]./sbttest testThe sbt function pointers are loaded from libobk.a(shr.o) library.Return code -1 from sbtinit, bsercoer = 7011, bsercerrno = 2534sbtopen: System error - eg. malloc, fork errors

4.那么是否是因为脚本的问题导致的呢？（1）备份数据⽂件到本地磁盘，发现没有问题

RMAN> run {

allocate channel d1 type disk format='/tmp/%U';backup tablespace sysaux;}2> 3> 4>

allocated channel: d1

channel d1: SID=20 device type=DISK

Starting backup at 22 Oct 2017 10:34:37channel d1: starting full datafile backup setchannel d1: specifying datafile(s) in backup set

input datafile file number=00002 name=/opt/app/oracle/oradata/ORA112/datafile/o1_mf_sysaux_62ojt49t_.dbfchannel d1: starting piece 1 at 25 Oct 2017 10:34:38channel d1: finished piece 1 at 25 Oct 2017 10:35:23

piece handle=/tmp/p7mn6kju_1_1 tag=TAG20110922T143437 comment=NONEchannel d1: backup set complete, elapsed time: 00:00:45Finished backup at 22 SEP 2011 14:35:23

Starting Control File and SPFILE Autobackup at 25 Oct 2017 10:35:23

piece handle=/opt/app/oracle/flash_recovery_area/ORA112/autobackup/2017_10_25/o1_mf_s_762532523_77ogrf5v_.bkp comment=NONEFinished Control File and SPFILE Autobackup at 25 Oct 2017 10:35:30released channel: d1

（2）备份到⼀个假的磁带库上，判断也没有问题

$ rman target /

Recovery Manager: Release 9.2.0.8.0 - Production on Thu Oct 25 10:24:56 2017Copyright (c) 1982, 2009, Oracle and/or its affiliates. All rights reserved.connected to target database: ORA112 (DBID=232055638)

RMAN> run {

2> allocate channel t1 type 'SBT' PARMS 'SBT_LIBRARY=oracle.disksbt,ENV=(BACKUP_DIR=/tmp)';3> backup tablespace sysaux;4> }

using target database control file instead of recovery catalogallocated channel: t1

channel t1: SID=20 device type=SBT_TAPEchannel t1: WARNING: Oracle Test Disk API

Starting backup at 22 Oct 2017 10:25:46channel t1: starting full datafile backup setchannel t1: specifying datafile(s) in backup set

input datafile file number=00002 name=/opt/app/oracle/oradata/ORA112/datafile/o1_mf_sysaux_62ojt49t_.dbfchannel t1: starting piece 1 at 25 Oct 2017 14:25:48channel t1: finished piece 1 at 25 Oct 2017 14:26:43

piece handle=p5mn6k3b_1_1 tag=TAG20110922T1427 comment=API Version 2.0,MMS Version 8.1.3.0channel t1: backup set complete, elapsed time: 00:00:55Finished backup at 25 Oct 2017 10:26:45

Starting Control File and SPFILE Autobackup at 25 Oct 2017 10:26:45

piece handle=c-232055638-20110922-00 comment=API Version 2.0,MMS Version 8.1.3.0Finished Control File and SPFILE Autobackup at 25 Oct 2017 10:26:52released channel: t1

（5）经过上⾯的分析，排除了脚本的问题，我们基本上确认就是因为磁带库的通讯问题导致的，这个时候询问同事，确实是因为磁带库的通讯问题导致的，⼀直没有解决的原因就是因为有的时候可以备份成功，不影响⽣产。找供应商解决问题，⼀直都不给解决。所以就⼀直拖着。