求助：ORA-03113:通信通道的文件结尾

[badehuang]

我在数据库执行查询操作，SQL如下：
select a.col1, a.col2, a.col3 from tab a where stat = '0' and type = '1' and code = '2' order by col2 desc, col3 asc;
结果出现：ORA-03113:通信通道的文件结尾 进程ID:11892 会话ID:序列号：58109

然后查看oracle日志文件如下：
Trace file /u01/app/base/diag/rdbms/orcl12c/orcl/trace/orcl_ora_11892.trc
Oracle Database 12c Enterprise Edition Release 12.1.0.2.0 - 64bit Production
With the Partitioning, OLAP, Advanced Analytics and Real Application Testing options
ORACLE_HOME = /u01/app/base/db_home
System name: Linux
Node name: fss-02
Release: 2.6.32-431.el6.x86_64
Version: #1 SMP Fri Nov 22 03:15:09 UTC 2013
Machine: x86_64
Instance name: orcl
Redo thread mounted by this instance: 1
Oracle process number: 88
Unix process pid: 11892, image: oracle@fss-02


*** 2017-11-03 15:23:41.746
*** SESSION ID:(6.58109) 2017-11-03 15:23:41.746
*** CLIENT ID:() 2017-11-03 15:23:41.746
*** SERVICE NAME:(orcl12c.us.oracle.com) 2017-11-03 15:23:41.746
*** MODULE NAME:(PL/SQL Developer) 2017-11-03 15:23:41.746
*** CLIENT DRIVER:() 2017-11-03 15:23:41.746
*** ACTION NAME:(SQL <B4><B0><BF><DA> - select DISTINCT bill.ID, bill.BILL_TYPE_CODE, bill.RE) 2017-11-03 15:23:41.746
*** CONTAINER ID:(1) 2017-11-03 15:23:41.746

Exception [type: SIGSEGV, Address not mapped to object] [ADDR:0x4] [PC:0xCDFB0C6, qcdlgcd()+70] [flags: 0x0, count: 1]
DDE: Problem Key 'ORA 7445 [qcdlgcd]' was flood controlled (0x6) (incident: 24725)
ORA-07445: 鍑虹幇寮傚父閿欒: 鏍稿績杞偍 [qcdlgcd()+70] [SIGSEGV] [ADDR:0x4] [PC:0xCDFB0C6] [Address not mapped to object] []
Dumping swap information
Memory (Avail / Total) = 355.41M / 15950.90M
Swap (Avail / Total) = 2047.99M / 2047.99M
ssexhd: crashing the process...
Shadow_Core_Dump = PARTIAL
ksdbgcra: writing core file to directory '/u01/app/base/diag/rdbms/orcl12c/orcl/cdump'


各位大神，求教下这是什么问题？

[minsic78]

楼主你的tab表是不是其实是个复杂的视图？很可能是中奖踩到BUG了～

[badehuang]

我查找了半天的资料，也觉得是oracle的bug,但是不知道从哪里下手解决，大神给点思路

[碧水幽幽泉]

ORA-07445是Oracle的一个bug。
需要知道tab具体是什么。

[minsic78]

引用 2 楼 badehuang 的回复:

我查找了半天的资料，也觉得是oracle的bug,但是不知道从哪里下手解决，大神给点思路

既然是BUG，就没什么思路可言。 我刚MOS查了下，对应的BUG是与index join相关的，所以看下SQL的执行计划吧，是不是有index join，有的话就中了，修改SQL，或者调整表结构，让SQL不要走这个操作试试。

[EVISWANG]

alter session set "_fix_control"='14191778:0';
试试这个，然后再执行你的SQL