环境：

os:linux redhat 7.9
oracel:11.2.0.4
服务器:192.168.10.50 
客户端:192.168.10.53 

1. on 192.168.10.50

----建表和存储过程

sqlplus test/test
create table as_test_exception (error_message varchar2(4000), tmstamp timestamp);
create or replace procedure log_error (sErrorMessage in varchar2)
is
pragma autonomous_transaction;
begin
insert into as_test_exception (error_message, tmstamp) values (sErrorMessage, systimestamp);
commit;
end;
/

select * from as_test_exception;

2. on 192.168.10.53

  建立连接：sqlplus test/test@192.168.10.50/orcl

3. on 192.168.10.50，查到10.53 连接过来的端口是 42072

[root@host123 yum.repos.d]# netstat -nap|grep "10.53"|grep "oracle"
tcp        0      0 192.168.10.50:1521      192.168.10.53:42072     ESTABLISHED 16713/oracleorcl    
[root@host123 yum.repos.d]# 

[root@host123 yum.repos.d]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
2: ens192: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 00:50:56:a8:a0:0a brd ff:ff:ff:ff:ff:ff
    inet 192.168.10.50/24 brd 192.168.10.255 scope global ens192
       valid_lft forever preferred_lft forever

4. on 192.168.10.53，执行存储过程

begin
dbms_lock.sleep(60);
exception
when others then
log_error(sqlerrm);
raise;
end;
/

5. on 192.168.10.50 ，使用tcpkill关闭查到的端口(42072)

 "-i" 后面的bond0是网卡，可以通过ifconfig找到
 "-9" 表示关闭连接的迫切程度，越大表示越强制，默认是3
 "port 42072" 是指定连接的表达式，这个和iptables差不多


[root@host123 yum.repos.d]# tcpkill -i ens192 -9 port 42072 
tcpkill: listening on ens192 [port 42072]

6. on 192.168.10.53，报错ORA-03135断开

SQL> begin
dbms_lock.sleep(60);
exception
  2    3    4  when others then
  5  log_error(sqlerrm);
  6  raise;
  7  end;
  8  /

begin
*
ERROR at line 1:
ORA-03135: connection lost contact
Process ID: 21147
Session ID: 2833 Serial number: 55307


SQL> 

7. on 192.168.10.50 ，netstat已经观察不到这个tcp连接

[oracle@host123 ~]$ netstat -nap|grep "10.53"|grep "oracle"
(Not all processes could be identified, non-owned process info
 will not be shown, you would have to be root to see it all.)
[oracle@host123 ~]$ 

----tcpkill捕获到此tcp交换，并关闭tcp连接
[root@host123 yum.repos.d]# tcpkill -i ens192 -9 port 42072 
tcpkill: listening on ens192 [port 42072]
192.168.10.50:1521 > 192.168.10.53:42072: R 22759912:22759912(0) win 0
192.168.10.50:1521 > 192.168.10.53:42072: R 22759978:22759978(0) win 0
192.168.10.50:1521 > 192.168.10.53:42072: R 22760110:22760110(0) win 0
192.168.10.50:1521 > 192.168.10.53:42072: R 22760308:22760308(0) win 0
192.168.10.50:1521 > 192.168.10.53:42072: R 22760572:22760572(0) win 0
192.168.10.50:1521 > 192.168.10.53:42072: R 22760902:22760902(0) win 0
192.168.10.50:1521 > 192.168.10.53:42072: R 22761298:22761298(0) win 0
192.168.10.50:1521 > 192.168.10.53:42072: R 22761760:22761760(0) win 0
192.168.10.50:1521 > 192.168.10.53:42072: R 22762288:22762288(0) win 0
192.168.10.53:42072 > 192.168.10.50:1521: R 4230768170:4230768170(0) win 0
192.168.10.53:42072 > 192.168.10.50:1521: R 4230768229:4230768229(0) win 0
192.168.10.53:42072 > 192.168.10.50:1521: R 4230768347:4230768347(0) win 0
192.168.10.53:42072 > 192.168.10.50:1521: R 4230768524:4230768524(0) win 0
192.168.10.53:42072 > 192.168.10.50:1521: R 4230768760:4230768760(0) win 0
192.168.10.53:42072 > 192.168.10.50:1521: R 4230769055:4230769055(0) win 0
192.168.10.53:42072 > 192.168.10.50:1521: R 4230769409:4230769409(0) win 0
192.168.10.53:42072 > 192.168.10.50:1521: R 4230769822:4230769822(0) win 0
192.168.10.53:42072 > 192.168.10.50:1521: R 4230770294:4230770294(0) win 0



----查看as_test_exception无记录
select * from as_test_exception;