2007-5-30 15:42
argenw
oracle 9208 for hp-ux ia 内存问题?
每个空连接就占用15M-20M系统内存,8G内存就能连接400用户内存就满了。
是不是什么参数没有设置明白,谁给点建议。
ibm , sun 上一个连接就占3-5M 左右,hp-ux ia 上怎么这么多?
2007-5-30 16:11
netbbs
用共享模式吧
2007-5-30 16:23
argenw
晕。。。。。。:L
2007-5-30 16:57
argenw
长事务,大事务,
共享模式没有太大意义吧?
2007-5-30 17:19
hexiong
swap 设置为多大? 再cat /proc/PID/status 结果 操作系统异常 和 alert.log的日志 贴出来看看
(其中 PID为ps -ef | grep NO的进程号)
2007-5-30 17:51
hexiong
不知道 hp-unix 的cat /proc/PID/status 结果 是否可以看得到
redhat as4 可以看到
solaris 的pmap命令 也可以进行更进一步具体观察和诊断
2007-5-31 01:14
oraix
Please paste the init parameters in here
maybe some bad setting in PGA or sort area
2007-6-1 10:29
argenw
*.background_dump_dest='/orasoft/admin/pjmis/bdump'
*.compatible='9.2.0.0.0'
*.control_files='/oracle/pjmis/control01.ctl','/oracle/pjmis/control02.ctl','/oracle/pjmis/control03.ctl'
*.core_dump_dest='/orasoft/admin/pjmis/cdump'
*.db_block_size=8192
*.db_cache_size=3145728000
*.db_domain=''
*.db_file_multiblock_read_count=16
*.db_name='pjmis'
*.fast_start_mttr_target=300
*.hash_join_enabled=TRUE
*.instance_name='pjmis'
*.java_pool_size=0
*.large_pool_size=8388608
*.open_cursors=600
*.pga_aggregate_target=25165824
*.processes=1000
*.query_rewrite_enabled='FALSE'
*.remote_login_passwordfile='EXCLUSIVE'
*.sga_max_size=6291456000
*.shared_pool_size=838860800
*.sort_area_size=524288
*.star_transformation_enabled='FALSE'
*.timed_statistics=TRUE
*.undo_management='AUTO'
*.undo_retention=10800
*.undo_tablespace='UNDOTBS1'
*.user_dump_dest='/orasoft/admin/pjmis/udump'
帮着看看吧。
2007-6-1 10:30
argenw
PGA 改为512m le,写错了,呵呵。SGA 也改小了,2G.
2007-6-1 19:48
rwq_
我测试了下,我们的系统当达到800个连接的时候,就开始报内存资源不足了,一直不解为什么HP-UX下这么吃内存(HP-UX8G物理内存,SGA4G,PGA1G)
2007-6-1 19:51
rwq_
[quote]原帖由 [i]argenw[/i] 于 2007-6-1 10:29 发表 [url=http://www.loveunix.com/redirect.php?goto=findpost&pid=668914&ptid=72834][img]http://www.loveunix.com/images/common/back.gif[/img][/url]
*.background_dump_dest='/orasoft/admin/pjmis/bdump'
*.compatible='9.2.0.0.0'
*.control_files='/oracle/pjmis/control01.ctl','/oracle/pjmis/control02.ctl','/oracle/pjmis/control03.ctl'
*.core_du ... [/quote]
靠,我还没有看过这么配置的,SGA过大了吧,PGA过小了吧。:L :L :L :L :L :L
2007-6-4 12:44
argenw
[quote]原帖由 [i]argenw[/i] 于 2007-6-1 10:30 发表 [url=http://bbs.loveunix.net/redirect.php?goto=findpost&pid=668918&ptid=72834][img]http://bbs.loveunix.net/images/common/back.gif[/img][/url]
PGA 改为512m le,写错了,呵呵。SGA 也改小了,2G. [/quote]
配置一点也不大啊,
hp就是吃内存啊,ibm,sun 都不吃这么多。
2007-6-4 12:44
mp3www
换个论坛吧:lol :lol :lol
2007-6-4 15:11
mp3www
Oracle recommends the following steps to address tuning for the increased memory allocation required for Persistent Private SQL Areas and Large Virtual Memory Page Sizes:
[list]Keep the value for CURSOR_SPACE_FOR_TIME to TRUE, unless the system indicates library cache misses while running the application. In that case, the shared pool may be small enough to hold the SQL areas for all concurrent open cursors.
[*]Decrease the virtual memory data page size for the Oracle9i release 2 (9.2.0.1.0) executable as necessary. Alter the page size setting with the following command:
/usr/bin/chatr +pd newsize $ORACLE_HOME/bin/oracle[*]newsize 越小内存用的越小,但是对cpu消耗越大[*]最小值为4k 多20%cpu[*]最大值1G,比4M要多用50%的内存[/list]
2007-6-4 21:56
hexiong
楼上好文!
这个连接 也有好的说明
[url]http://www.linuxmine.com/51672.html[/url]
2007-6-7 10:41
argenw
改成多小呢?我早就改成64M了。没有效果啊!
页:
[1]
Powered by Discuz! Archiver 5.5.0
© 2001-2006 Comsenz Inc.