最近突然有一个oracle 12c R2环境的trace目录使用率暴涨，分析发现一天创建了几十万个trace，几乎是每个会话创建一个trace, 当前数据库也并未启用event。简单记录
# # sqlplus / as sysdba
SQL*Plus: Release 12.2.0.1.0 Production on Mon Nov 4 15:26:16 2019
Copyright (c) 1982, 2016, Oracle. All rights reserved.
Connected to:
Oracle Database 12c Enterprise Edition Release 12.2.0.1.0 - 64bit Production
USERNAME INST_NAME HOST_NAME SID SERIAL# VERSION STARTED SPID OPID CPID SADDR PADDR
-------------------- -------------------- ------------------------- ----- -------- ---------- -------- --------------- ----- --------------- ---------------- ----------------
SYS anbob-anbob1 weejar1 355 47536 12.2.0.1.0 20190910 47203 249 47201 00000005A22602C0 00000005E20E44A0
SQL> select * from v$diag_info;
INST_ID NAME VALUE CON_ID
---------- ----------------------------- -----------------------------------------------------------------------------
1 Diag Enabled TRUE 0
1 ADR Base /oracle/app/oracle 0
1 ADR Home /oracle/app/oracle/diag/rdbms/anbob/anbob1 0
1 Diag Trace /oracle/app/oracle/diag/rdbms/anbob/anbob1/trace 0
1 Diag Alert /oracle/app/oracle/diag/rdbms/anbob/anbob1/alert 0
1 Diag Incident /oracle/app/oracle/diag/rdbms/anbob/anbob1/incident 0
1 Diag Cdump /oracle/app/oracle/diag/rdbms/anbob/anbob1/cdump 0
1 Health Monitor /oracle/app/oracle/diag/rdbms/anbob/anbob1/hm 0
1 Default Trace File /oracle/app/oracle/diag/rdbms/anbob/anbob1/trace/anbob1_ora_47203.trc 0
1 Active Problem Count 0 0
1 Active Incident Count 0 0
11 rows selected.
SQL> host
# vi /oracle/app/oracle/diag/rdbms/anbob/anbob1/trace/anbob1_ora_47203.trc
Trace file /oracle/app/oracle/diag/rdbms/anbob/anbob1/trace/anbob1_ora_47203.trc
Oracle Database 12c Enterprise Edition Release 12.2.0.1.0 - 64bit Production
Build label: RDBMS_12.2.0.1.0_LINUX.X64_170125
ORACLE_HOME: /oracle/app/oracle/product/12.2.0/db_1
System name: Linux
Node name: weejar1
Release: 4.4.21-69-default
Version: #1 SMP Tue Oct 25 10:58:20 UTC 2016 (9464f67)
Machine: x86_64
Instance name: anbob1
Redo thread mounted by this instance: 1
Oracle process number: 249
Unix process pid: 47203, image: oracle@weejar1 (TNS V1-V3)
*** 2019-11-04T15:26:16.967435+08:00
*** CLIENT ID:() 2019-11-04T15:26:16.967480+08:00
*** SERVICE NAME:() 2019-11-04T15:26:16.967486+08:00
*** MODULE NAME:() 2019-11-04T15:26:16.967491+08:00
*** ACTION NAME:() 2019-11-04T15:26:16.967496+08:00
*** CLIENT DRIVER:() 2019-11-04T15:26:16.967501+08:00
KSIPC: KSIPC: Unable to attach to ipcor topo svc object
~
# cd /oracle/app/oracle/diag/rdbms/anbob/anbob1/trace
# ls -l|grep "^-" |awk '{print $6$7 " " $5 }'|sort|awk ' {
> arr[$1]+=$2
> freq[$1]++
> }
> END{
> printf "%20s\t%20s\t%20s\n" ,"date" , "size" , "num_file"
> printf "%20s\t%20s\t%20s\n" ,"----------------------" , "--------------------" , "-------------------"
> for (key in arr)
> printf "%20s\t%20s\t%20s\n", key, arr[key] ,freq[key]
> }'
date size num_file
---------------------- -------------------- -------------------
Oct14 10291 3
Oct15 15867899 9364
Oct16 44671620 17800
Oct17 1583739055 28698
Oct18 24193508779 160892
解决方法：
这是用户登录时遇到的IPC错误， 生成了不需要的trace文件，该错误不影响性能和使用， 可以直接删除，通常重启实例可以解决。