昨天有客户打电话说，由于机房突然断电，导致数据库服务器宕机，无法启动，经过主机工程师的不懈努力重要启动起来。

可数据库启动一直报

Fri Jun 21 16:00:34 2024

GEN0 started with pid=5, OS id=3027 

Fri Jun 21 16:00:34 2024

DIAG started with pid=6, OS id=3029 

Fri Jun 21 16:00:34 2024

DBRM started with pid=7, OS id=3031 

Fri Jun 21 16:00:34 2024

DIA0 started with pid=8, OS id=3033 

Fri Jun 21 16:00:34 2024

MMAN started with pid=9, OS id=3035 

Fri Jun 21 16:00:34 2024

DBW0 started with pid=10, OS id=3037 

Fri Jun 21 16:00:34 2024

DBW1 started with pid=11, OS id=3039 

Fri Jun 21 16:00:34 2024

DBW2 started with pid=12, OS id=3041 

Fri Jun 21 16:00:34 2024

LGWR started with pid=13, OS id=3043 

Fri Jun 21 16:00:34 2024

CKP

T started with pid=14, OS id=3045 

Fri Jun 21 16:00:34 2024

SMON started with pid=15, OS id=3047 

Fri Jun 21 16:00:34 2024

RECO started with pid=16, OS id=3049 

Fri Jun 21 16:00:34 2024

MMON started with pid=17, OS id=3051 

Fri Jun 21 16:00:34 2024

MMNL started with pid=18, OS id=3053 

starting up 1 dispatcher(s) for network address '(ADDRESS=(PARTIAL=YES)(PROTOCOL=TCP))'...

starting up 1 shared server(s) ...

ORACLE_BASE from environment = /home/oracle/app/oracle

Fri Jun 21 16:00:34 2024

Exception [type: SIGSEGV, Address not mapped to object] [ADDR:0x0] [PC:0x418BDAE, nttaddr2bnd()+2158] [flags: 0x0, count: 1]

Fri Jun 21 16:00:35 2024

ALTER DATABASE   MOUNT

Errors in file /home/oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_d000_3055.trc  (incident=290566):

ORA-07445: exception encountered: core dump [nttaddr2bnd()+2158] [SIGSEGV] [ADDR:0x0] [PC:0x418BDAE] [Address not mapped to object] []

Incident details in: /home/oracle/app/oracle/diag/rdbms/orcl/orcl/incident/incdir_290566/orcl_d000_3055_i290566.trc

Use ADRCI or Support Workbench to package the incident.

See Note 411.1 at My Oracle Support for error and packaging details.

Dumping diagnostic data in directory=[cdmp_20240621160036], requested by (instance=1, osid=3055 (D000)), summary=[incident=290566].

Fri Jun 21 16:00:38 2024

Sweep [inc][290566]: completed

Sweep [inc][289380]: completed

Sweep [inc][289379]: completed

Sweep [inc2][290566]: completed

Sweep [inc2][289380]: completed

Sweep [inc2][289379]: completed

Successful mount of redo thread 1, with mount id 1700046371

Database mounted in Exclusive Mode

Lost write protection disabled

Completed: ALTER DATABASE   MOUNT

Fri Jun 21 16:00:39 2024

ALTER DATABASE OPEN-----------数据库也能正常启动。

ORA-07445: exception encountered: core dump [nttaddr2bnd()+2158] [SIGSEGV] [ADDR:0x0] [PC:0x418BDAE] [[]---报了此错误，感觉是和地址有关，  
问题非常奇怪，数据库正常open后，登录到库里后检查文件也都正常。查看了trc里的内容也都是报address的问题。
查了下MOS，发现这样一篇文章比较符合，说的是主机的hosts文件不符合要求会导致这个问题。
Error in the Alert Log: ORA-7445[nttaddr2bnd] (Doc ID 2393755.1)
SYMPTOMS
The following error occurred in the alert log:
ORA-07445: exception encountered: core dump [nttaddr2bnd()+2246] [SIGSEGV] [ADDR:0x0] [PC:0x55869B6] [Address not mapped to object] []
 
The Call Stack Trace in the associated incident trace file shows:
nttaddr2bnd <- ntacaddr2bnd <- nsgetaddr <- kmdcac <- kmdmai <- kmmrdp <-opirip <- opidrv <- sou2o <- opimai_real <- ssthrdmain <- main
 
CAUSE
This issue is caused by a misconfiguration of the file /etc/hosts.  For example:
$ cat /etc/hosts
<IP1> localhost.localdomain localhost
<IP2> localhost localhost.localdomain
  而本次故障hosts 设置为
  127.0.0.1 localhost.localdomain localhost
   ip1  localhost
   
   1.修改主机名
hostnamectl set-hostname oradb
   2.修改hosts
   ip1  oradb
   3.启动数据库。
   报错消失。
 说到最后，没有想到停电会导致这个问题，所以安装数据库要找专业人员。（猜测安装问到出问题主机应该没有重启过）。