昨天一套开发库在开发人员修改了db_cache_size后instance crash了, 10.2.0.1 single instance on RHEL linux,是因为触发了一个bug,这里简单记录一下.
# db alert
Mon Dec 22 22:00:58 2014
db_recovery_file_dest_size of 2048 MB is 0.38% used. This is a
user-specified limit on the amount of space that will be used by this
database for recovery-related files, and does not reflect the amount of
space available in the underlying filesystem or ASM diskgroup.
Mon Dec 22 22:01:07 2014
GATHER_STATS_JOB encountered errors. Check the trace file.
Mon Dec 22 22:01:07 2014
Errors in file /oracle/admin/anbobtest/bdump/anbobtest_j001_31269.trc:
ORA-20000: index “anbob_SHORT”.”SYS_MTABLE_0000260EA_IND_1″ or partition of such index is in unusable state
Mon Dec 22 22:40:42 2014
Errors in file /oracle/admin/anbobtest/bdump/anbobtest_mman_31205.trc:
ORA-00600: internal error code, arguments: [kmgs_pre_process_request_6], [6], [895], [0], [3], [0x459C1F3D8], [], []
Mon Dec 22 22:40:43 2014
Errors in file /oracle/admin/anbobtest/bdump/anbobtest_mman_31205.trc:
ORA-00600: internal error code, arguments: [kmgs_pre_process_request_6], [6], [895], [0], [3], [0x459C1F3D8], [], []
Mon Dec 22 22:40:43 2014
MMAN: terminating instance due to error 822
Instance terminated by MMAN, pid = 31205
$ vi /oracle/admin/anbobtest/bdump/anbobtest_mman_31205.trc
MOS note# 373802.1
MMON Terminates Instance With ORA-600 [kmgs_pre_process_request_6] or ORA-600 [kmgs_update_target_size_1] When Resizing Caches
ORA-00600: internal error code, arguments: [kmgs_update_target_size_1]
The call stack from the ORA-00600 trace file will be similar to:
… kmgs_pre_process_request <- kmgsdrv <- ksbabs <- ksbrdp <- opirip <- opidrv <- sou2o <- opimai_real <- main <- start or ... kmgs_update_target_size <- kmgs_update_parameter_ctx <- kmgs_update_param_manual_helper <- kmgs_update_param_manual
Cause
This is due to Bug:4433838 Similar issues have been filed as bugs which have also been determined to be duplicates of Bug:4433838 . The error occurs when the parameter SGA_TARGET is set to an exact multiple of 4Gb. For instance, if the SGA_TARGET value in HEXADECIMAL has 00000000 as the last 4 bytes then you can hit this problem. The bug indicates this happens on an exact multiple of 4G. Solution
1. Upgrade to 10.2.0.4 or higher.
OR
2 Apply the one-off Patch:4433838 depending on its availability for your patchset-level and platform
OR
3. Use the workaround: Set your SGA_TARGET parameter to any non-multiple values of 4 Gb. In other words, ensure that your SGA_TARGET value in HEXADECIMAL does not have 00000000 as the last 4 bytes.
我后来禁用了ASMM,直接把SGA_TARGET=0后,错误没有再出现
# db alert
Mon Dec 22 22:00:58 2014
db_recovery_file_dest_size of 2048 MB is 0.38% used. This is a
user-specified limit on the amount of space that will be used by this
database for recovery-related files, and does not reflect the amount of
space available in the underlying filesystem or ASM diskgroup.
Mon Dec 22 22:01:07 2014
GATHER_STATS_JOB encountered errors. Check the trace file.
Mon Dec 22 22:01:07 2014
Errors in file /oracle/admin/anbobtest/bdump/anbobtest_j001_31269.trc:
ORA-20000: index “anbob_SHORT”.”SYS_MTABLE_0000260EA_IND_1″ or partition of such index is in unusable state
Mon Dec 22 22:40:42 2014
Errors in file /oracle/admin/anbobtest/bdump/anbobtest_mman_31205.trc:
ORA-00600: internal error code, arguments: [kmgs_pre_process_request_6], [6], [895], [0], [3], [0x459C1F3D8], [], []
Mon Dec 22 22:40:43 2014
Errors in file /oracle/admin/anbobtest/bdump/anbobtest_mman_31205.trc:
ORA-00600: internal error code, arguments: [kmgs_pre_process_request_6], [6], [895], [0], [3], [0x459C1F3D8], [], []
Mon Dec 22 22:40:43 2014
MMAN: terminating instance due to error 822
Instance terminated by MMAN, pid = 31205
$ vi /oracle/admin/anbobtest/bdump/anbobtest_mman_31205.trc
/oracle/admin/anbobtest/bdump/anbobtest_mman_31205.trc
Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - 64bit Production
With the Partitioning, OLAP and Data Mining options
ORACLE_HOME = /oracle/product/102
System name: Linux
Node name: OSS-TEST
Release: 2.6.32-358.el6.x86_64
Version: #1 SMP Tue Jan 29 11:47:41 EST 2013
Machine: x86_64
Instance name: anbobtest
Redo thread mounted by this instance: 1
Oracle process number: 4
Unix process pid: 31205, image: oracle@OSS-TEST (MMAN)
*** SERVICE NAME:(SYS$BACKGROUND) 2014-12-22 22:40:42.878
*** SESSION ID:(2203.1) 2014-12-22 22:40:42.878
*** 2014-12-22 22:40:42.878
ksedmp: internal or fatal error
ORA-00600: internal error code, arguments: [kmgs_pre_process_request_6], [6], [895], [0], [3], [0x459C1F3D8], [], []
----- Call Stack Trace -----
calling call entry argument values in hex
location type point (? means dubious value)
-------------------- -------- -------------------- ----------------------------
ksedst()+31 call ksedst1() 000000000 ? 000000001 ?
000000000 ? 000000000 ?
000000000 ? 000000001 ?
ksedmp()+610 call ksedst() 000000000 ? 000000001 ?
000000000 ? 000000000 ?
000000000 ? 000000001 ?
ksfdmp()+21 call ksedmp() 000000003 ? 000000001 ?
000000000 ? 000000000 ?
000000000 ? 000000001 ?
kgerinv()+161 call ksfdmp() 000000003 ? 000000001 ?
000000000 ? 000000000 ?
000000000 ? 000000001 ?
kgeasnmierr()+163 call kgerinv() 006579D60 ? 7F22AFC90040 ?
000000000 ? 000000000 ?
000000001 ? 000000001 ?
kmgs_pre_process_re call kgeasnmierr() 006579D60 ? 7F22AFC90040 ?
quest()+2982 000000000 ? 000000000 ?
000000000 ? 000000006 ?
kmgsdrv()+1921 call kmgs_pre_process_re 459C1F3D8 ? 7F22AFC90040 ?
quest() 000000000 ? 000000000 ?
000000000 ? 459C1F6A0 ?
ksbabs()+564 call kmgsdrv() 7FFF102EF770 ? 000000010 ?
000000000 ? 000000000 ?
000000000 ? 459C1F6A0 ?
ksbrdp()+727 call ksbabs() 7FFF102EF770 ? 000000010 ?
000000000 ? 000000000 ?
000000000 ? 459C1F6A0 ?
opirip()+616 call ksbrdp() 7FFF102EF770 ? 000000010 ?
000000001 ? 0600117C8 ?
000000000 ? 459C1F6A0 ?
opidrv()+582 call opirip() 000000032 ? 000000004 ?
7FFF102F0878 ? 0600117C8 ?
000000000 ? 459C1F6A0 ?
sou2o()+114 call opidrv() 000000032 ? 000000004 ?
7FFF102F0878 ? 0600117C8 ?
000000000 ? 459C1F6A0 ?
Cannot find symbol
Cannot find symbol
Cannot find symbol
opimai_real()+317 call sou2o() 7FFF102F0850 ? 000000032 ?
Cannot find symbol
Cannot find symbol
opimai_real()+317 call sou2o() 7FFF102F0850 ? 000000032 ?
000000004 ? 7FFF102F0878 ?
000000000 ? 459C1F6A0 ?
main()+116 call opimai_real() 000000003 ? 7FFF102F08E0 ?
000000004 ? 7FFF102F0878 ?
000000000 ? 459C1F6A0 ?
<0x36fb61ecdd> call main() 000000003 ? 7FFF102F08E0 ?
000000004 ? 7FFF102F0878 ?
000000000 ? 459C1F6A0 ?复制
MOS note# 373802.1
MMON Terminates Instance With ORA-600 [kmgs_pre_process_request_6] or ORA-600 [kmgs_update_target_size_1] When Resizing Caches
ORA-00600: internal error code, arguments: [kmgs_update_target_size_1]
The call stack from the ORA-00600 trace file will be similar to:
… kmgs_pre_process_request <- kmgsdrv <- ksbabs <- ksbrdp <- opirip <- opidrv <- sou2o <- opimai_real <- main <- start or ... kmgs_update_target_size <- kmgs_update_parameter_ctx <- kmgs_update_param_manual_helper <- kmgs_update_param_manual
Cause
This is due to Bug:4433838 Similar issues have been filed as bugs which have also been determined to be duplicates of Bug:4433838 . The error occurs when the parameter SGA_TARGET is set to an exact multiple of 4Gb. For instance, if the SGA_TARGET value in HEXADECIMAL has 00000000 as the last 4 bytes then you can hit this problem. The bug indicates this happens on an exact multiple of 4G. Solution
1. Upgrade to 10.2.0.4 or higher.
OR
2 Apply the one-off Patch:4433838 depending on its availability for your patchset-level and platform
OR
3. Use the workaround: Set your SGA_TARGET parameter to any non-multiple values of 4 Gb. In other words, ensure that your SGA_TARGET value in HEXADECIMAL does not have 00000000 as the last 4 bytes.
我后来禁用了ASMM,直接把SGA_TARGET=0后,错误没有再出现
「喜欢这篇文章,您的关注和赞赏是给作者最好的鼓励」
关注作者
【版权声明】本文为墨天轮用户原创内容,转载时必须标注文章的来源(墨天轮),文章链接,文章作者等基本信息,否则作者和墨天轮有权追究责任。如果您发现墨天轮中有涉嫌抄袭或者侵权的内容,欢迎发送邮件至:contact@modb.pro进行举报,并提供相关证据,一经查实,墨天轮将立刻删除相关内容。
评论
相关阅读
Oracle DataGuard高可用性解决方案详解
孙莹
552次阅读
2025-03-26 23:27:33
Oracle RAC 一键安装翻车?手把手教你如何排错!
Lucifer三思而后行
515次阅读
2025-04-15 17:24:06
XTTS跨版本迁移升级方案(11g to 19c RAC for Linux)
zwtian
419次阅读
2025-04-08 09:12:48
墨天轮个人数说知识点合集
JiekeXu
417次阅读
2025-04-01 15:56:03
【纯干货】Oracle 19C RU 19.27 发布,如何快速升级和安装?
Lucifer三思而后行
415次阅读
2025-04-18 14:18:38
Oracle SQL 执行计划分析与优化指南
Digital Observer
412次阅读
2025-04-01 11:08:44
Oracle数据库一键巡检并生成HTML结果,免费脚本速来下载!
陈举超
376次阅读
2025-04-20 10:07:02
Oracle 19c RAC更换IP实战,运维必看!
szrsu
357次阅读
2025-04-08 23:57:08
【活动】分享你的压箱底干货文档,三篇解锁进阶奖励!
墨天轮编辑部
335次阅读
2025-04-17 17:02:24
oracle定时任务常用攻略
virvle
324次阅读
2025-03-25 16:05:19
热门文章
移除DataGuard Standby配置导致Primary启动失败
2023-08-17 21290浏览
使用dblink产生的”SELECT /*+ FULL(P) +*/ * FROM XXXXX P ” 解析
2023-06-20 20891浏览
Troubleshooting 'ORA-28041: Authentication protocol internal error' change password 12c R2 DB
2020-04-08 13637浏览
浅谈ORACLE免费数据库Oracle Database XE (Express Edition) 版
2018-10-31 7585浏览
High wait event ‘row cache mutex’ in 12cR2、19c
2020-08-14 5565浏览