Oracle诊断案例-SGA与Swap之一

link:

http://www.eygle.com/case/sga1.htm

案例描述:

用户报告，服务器启动一段时间以后，无法建立数据库连接
重新启动几分钟以后，再次无法连接

系统无法正常使用.

1.登陆系统

SunOS 5.8

login: root
Password:
Last login: Tue Mar 23 13:56:59 from 172.16.31.41
Sun Microsystems Inc. SunOS 5.8 Generic Patch October 2001
You have new mail.

2.su 为Oracle用户
检查启动的Oracle进程

发现后台进程正常，有一定量的用户连接

> > wapplatform:/>su - oracle
> > Sun Microsystems Inc. SunOS 5.8 Generic Patch October 2001
> > You have new mail.
> > /export/home1/oracle>ls
> > admin codesyndealt31 exp.sh local.cshrc local.profile oraclebak oui v6_database
> > app exp.log jre local.login nsmail oradata swan
> > export/home1/oracle>cd admin
> > /export/home1/oracle/admin>ps -ef|grep ora
> > oracle 25269 25258 0 13:58:36 pts/3 0:00 grep ora
> > oracle 25257 24906 0 13:58:31 pts/4 0:00 vi alert_HSWAPDB.log
> > oracle 25267 1 1 13:58:34 ? 0:00 oracleHSWAPDB (LOCAL=NO)
> > oracle 25184 1 0 13:56:57 ? 0:00 ora_p007_HSWAPDB
> > oracle 25182 1 0 13:56:57 ? 0:00 ora_p006_HSWAPDB
> > oracle 25193 1 0 13:57:03 ? 0:01 oracleHSWAPDB (LOCAL=NO)
> > oracle 25209 1 0 13:57:09 ? 0:00 oracleHSWAPDB (LOCAL=NO)
> > oracle 25176 1 0 13:56:57 ? 0:00 ora_p003_HSWAPDB
> > oracle 25180 1 0 13:56:57 ? 0:00 ora_p005_HSWAPDB
> > oracle 25172 1 0 13:56:56 ? 0:00 ora_p001_HSWAPDB
> > oracle 25178 1 0 13:56:57 ? 0:00 ora_p004_HSWAPDB
> > oracle 25170 1 0 13:56:56 ? 0:00 ora_p000_HSWAPDB
> > oracle 24254 24240 0 12:08:25 pts/2 0:00 -ksh
> > oracle 25174 1 0 13:56:56 ? 0:00 ora_p002_HSWAPDB
> > oracle 25244 1 1 13:58:23 ? 0:00 oracleHSWAPDB (LOCAL=NO)
> > oracle 25218 1 0 13:57:23 ? 0:00 oracleHSWAPDB (LOCAL=NO)
> > oracle 25159 1 0 13:56:42 ? 0:02 ora_qmn0_HSWAPDB
> > oracle 25230 1 0 13:57:40 ? 0:01 oracleHSWAPDB (LOCAL=NO)
> > oracle 25161 1 0 13:56:42 ? 0:00 ora_s000_HSWAPDB
> > oracle 25149 1 0 13:56:41 ? 0:01 ora_lgwr_HSWAPDB
> > oracle 25157 1 0 13:56:42 ? 0:00 ora_cjq0_HSWAPDB
> > oracle 24906 3698 0 13:47:47 pts/4 0:00 -ksh
> > oracle 25153 1 0 13:56:42 ? 0:01 ora_smon_HSWAPDB
> > oracle 25058 7464 0 13:55:14 pts/1 0:00 -ksh
> > oracle 25163 1 0 13:56:42 ? 0:00 ora_d000_HSWAPDB
> > oracle 25155 1 0 13:56:42 ? 0:00 ora_reco_HSWAPDB
> > oracle 25151 1 0 13:56:41 ? 0:00 ora_ckpt_HSWAPDB
> > oracle 25145 1 0 13:56:41 ? 0:00 ora_dbw0_HSWAPDB
> > oracle 25199 1 15 13:57:04 ? 0:49 ora_j000_HSWAPDB
> > oracle 4149 4146 0 12:05:11 pts/5 0:00 -ksh
> > oracle 25232 1 0 13:57:41 ? 0:00 oracleHSWAPDB (LOCAL=NO)
> > oracle 25119 1 0 13:56:29 ? 0:00 oraclehswapdb (LOCAL=NO)
> > oracle 25075 1 0 13:55:34 ? 0:00 /export/home1/oracle/app/bin/tnslsnr LISTENER -inherit
> > oracle 24374 4149 0 12:21:56 pts/5 0:00 sqlplus /nolog
> > oracle 25143 1 0 13:56:41 ? 0:00 ora_pmon_HSWAPDB
> > oracle 25258 25242 0 13:58:31 pts/3 0:00 -ksh
> > /export/home1/oracle/admin>ps -ef|grep ora_
> > oracle 25275 25258 0 13:58:42 pts/3 0:00 grep ora_
> > oracle 25184 1 0 13:56:57 ? 0:00 ora_p007_HSWAPDB
> > oracle 25182 1 0 13:56:57 ? 0:00 ora_p006_HSWAPDB
> > oracle 25176 1 0 13:56:57 ? 0:00 ora_p003_HSWAPDB
> > oracle 25180 1 0 13:56:57 ? 0:00 ora_p005_HSWAPDB
> > oracle 25172 1 0 13:56:56 ? 0:00 ora_p001_HSWAPDB
> > oracle 25178 1 0 13:56:57 ? 0:00 ora_p004_HSWAPDB
> > oracle 25170 1 0 13:56:56 ? 0:00 ora_p000_HSWAPDB
> > oracle 25174 1 0 13:56:56 ? 0:00 ora_p002_HSWAPDB
> > oracle 25159 1 0 13:56:42 ? 0:02 ora_qmn0_HSWAPDB
> > oracle 25161 1 0 13:56:42 ? 0:00 ora_s000_HSWAPDB
> > oracle 25149 1 0 13:56:41 ? 0:01 ora_lgwr_HSWAPDB
> > oracle 25157 1 0 13:56:42 ? 0:00 ora_cjq0_HSWAPDB
> > oracle 25153 1 0 13:56:42 ? 0:01 ora_smon_HSWAPDB
> > oracle 25163 1 0 13:56:42 ? 0:00 ora_d000_HSWAPDB
> > oracle 25155 1 0 13:56:42 ? 0:00 ora_reco_HSWAPDB
> > oracle 25151 1 0 13:56:41 ? 0:00 ora_ckpt_HSWAPDB
> > oracle 25145 1 0 13:56:41 ? 0:00 ora_dbw0_HSWAPDB
> > oracle 25199 1 13 13:57:04 ? 0:51 ora_j000_HSWAPDB
> > oracle 25143 1 0 13:56:41 ? 0:00 ora_pmon_HSWAPDB

3.检查Alert.log警报日志文件

> > /export/home1/oracle/admin>ls
> > hswapdb
> > /export/home1/oracle/admin>cd *
> > /export/home1/oracle/admin/hswapdb>ls
> > bdump cdump create pfile udump
> > /export/home1/oracle/admin/hswapdb>cd bdump
> > /export/home1/oracle/admin/hswapdb/bdump> > > /export/home1/oracle/admin/hswapdb/bdump>ls -l *.log > > -rw-r--r-- 1 oracle dba 813396 Mar 23 13:57 alert_HSWAPDB.log
> /export/home1/oracle/admin/hswapdb/bdump>vi .log
> "alert_HSWAPDB.log" 18888 lines, 813396 characters (115 null)
> Tue Jun 24 21:17:14 2003
> Starting ORACLE instance (normal)
> LICENSE_MAX_SESSION = 0
> LICENSE_SESSIONS_WARNING = 0
> SCN scheme 3
> Using log_archive_dest parameter default value
> LICENSE_MAX_USERS = 0
> SYS auditing is disabled
> Starting up ORACLE RDBMS Version: 9.2.0.3.0.
> System parameters with non-default values:
> processes = 400
> timed_statistics = TRUE
> shared_pool_size = 117440512
> large_pool_size = 83886080
> java_pool_size = 33554432
> control_files = /export/home1/oracle/oradata/hswapdb/control01.ctl, > > /export/home1/oracle/oradata/hswapdb/control02.ctl,
> /export/home1/oracle/oradata/hswapdb/control03.ctl
> db_block_size = 8192
> db_cache_size = 352321536
> compatible = 9.2.0.0.0
> db_file_multiblock_read_count= 16
> fast_start_mttr_target = 300
> undo_management = AUTO
> undo_tablespace = UNDOTBS1
> undo_retention = 10800
> remote_login_passwordfile= EXCLUSIVE
> db_domain = eygle.com
> instance_name = hswapdb
> dispatchers = (PROTOCOL=TCP) (SERVICE=hswapdbXDB)
> job_queue_processes = 10
> hash_join_enabled = TRUE
> background_dump_dest = /export/home1/oracle/admin/hswapdb/bdump
> user_dump_dest = /export/home1/oracle/admin/hswapdb/udump
> core_dump_dest = /export/home1/oracle/admin/hswapdb/cdump
> sort_area_size = 524288
> db_name = hswapdb
> open_cursors = 300
> star_transformation_enabled= FALSE
> query_rewrite_enabled = FALSE
> pga_aggregate_target = 154140672
> aq_tm_processes = 1 > > ................. > > Tue Mar 23 13:40:45 2004
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 11, op = fork, loc = skgpspawn5
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> Tue Mar 23 13:42:02 2004
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> Tue Mar 23 13:55:38 2004
> Starting ORACLE instance (normal)
> Shutting down instance: further logons disabled
> Tue Mar 23 13:56:20 2004
> Shutting down instance (abort)
> License high water mark = 26
> Instance terminated by USER, pid = 25112
> Tue Mar 23 13:56:37 2004
> Starting ORACLE instance (normal)
> LICENSE_MAX_SESSION = 0
> LICENSE_SESSIONS_WARNING = 0
> SCN scheme 3
> Using log_archive_dest parameter default value
> LICENSE_MAX_USERS = 0
> SYS auditing is disabled
> Starting up ORACLE RDBMS Version: 9.2.0.3.0.
> System parameters with non-default values:
> processes = 400
> timed_statistics = TRUE
> shared_pool_size = 117440512
> large_pool_size = 83886080
> java_pool_size = 33554432
> control_files = /export/home1/oracle/oradata/hswapdb/control01.ctl, > > /export/home1/oracle/oradata/hswapdb/control02.ctl,
> /export/home1/oracle/oradata/hswapdb/control03.ctl
> db_block_size = 8192
> db_cache_size = 352321536
> compatible = 9.2.0.0.0
> db_file_multiblock_read_count= 16
> fast_start_mttr_target = 300
> undo_management = AUTO
> undo_tablespace = UNDOTBS1
> undo_retention = 10800
> remote_login_passwordfile= EXCLUSIVE
> db_domain = eygle.com
> instance_name = hswapdb
> dispatchers = (PROTOCOL=TCP) (SERVICE=hswapdbXDB)
> remote_dependencies_mode = SIGNATURE
> job_queue_processes = 10
> hash_join_enabled = TRUE
> background_dump_dest = /export/home1/oracle/admin/hswapdb/bdump
> user_dump_dest = /export/home1/oracle/admin/hswapdb/udump
> core_dump_dest = /export/home1/oracle/admin/hswapdb/cdump
> sort_area_size = 524288
> db_name = hswapdb
> open_cursors = 300
> star_transformation_enabled= FALSE
> parallel_automatic_tuning= TRUE
> query_rewrite_enabled = FALSE
> pga_aggregate_target = 154140672
> aq_tm_processes = 1
> PMON started with pid=2
> DBW0 started with pid=3
> LGWR started with pid=4
> CKPT started with pid=5
> SMON started with pid=6
> RECO started with pid=7
> CJQ0 started with pid=8
> QMN0 started with pid=9
> Tue Mar 23 13:56:42 2004
> starting up 1 shared server(s) ...
> Tue Mar 23 13:56:42 2004
> starting up 1 dispatcher(s) for network address '(ADDRESS=(PARTIAL=YES)(PROTOCOL=TCP))'...
> Tue Mar 23 13:56:43 2004
> ALTER DATABASE MOUNT
> Tue Mar 23 13:56:47 2004
> Successful mount of redo thread 1, with mount id 3253076635.
> Tue Mar 23 13:56:47 2004
> Database mounted in Exclusive Mode.
> Completed: ALTER DATABASE MOUNT
> Tue Mar 23 13:56:47 2004
> Current log# 2 seq# 2136 mem# 0: /export/home1/oracle/oradata/hswapdb/redo02.log
> Successful open of redo thread 1.
> Tue Mar 23 12:24:54 2004
> SMON: enabling cache recovery
> Tue Mar 23 12:24:56 2004
> Undo Segment 1 Onlined
> Undo Segment 2 Onlined
> Undo Segment 3 Onlined
> Undo Segment 4 Onlined
> Undo Segment 5 Onlined
> Undo Segment 6 Onlined
> Undo Segment 7 Onlined
> Undo Segment 8 Onlined
> Undo Segment 9 Onlined
> Undo Segment 10 Onlined
> Successfully onlined Undo Tablespace 1.
> Tue Mar 23 12:24:56 2004
> SMON: enabling tx recovery
> Tue Mar 23 12:24:56 2004
> Database Characterset is ZHS16GBK
> Tue Mar 23 12:25:01 2004
> SMON: Parallel transaction recovery tried
> Tue Mar 23 12:25:01 2004
> replication_dependency_tracking turned off (no async multimaster replication found)
> Completed: ALTER DATABASE OPEN
> Tue Mar 23 12:28:26 2004
> / OracleOEM / ALTER DATABASE DATAFILE '/export/home1/oracle/oradata/hswapdb/users01.dbf' RESIZE 2501760K
> Tue Mar 23 12:28:26 2004
> ORA-3297 signalled during: / OracleOEM / ALTER DATABASE DATAFILE '/export/h...
> Tue Mar 23 12:28:32 2004
> / OracleOEM / ALTER DATABASE DATAFILE '/export/home1/oracle/oradata/hswapdb/users01.dbf' RESIZE 2501760K
> ORA-3297 signalled during: / OracleOEM / ALTER DATABASE DATAFILE '/export/h...
> Tue Mar 23 12:28:53 2004
> / OracleOEM / ALTER DATABASE DATAFILE '/export/home1/oracle/oradata/hswapdb/users01.dbf' RESIZE 3501760K
> Tue Mar 23 12:28:53 2004
> ORA-3297 signalled during: / OracleOEM */ ALTER DATABASE DATAFILE '/export/h...
> Tue Mar 23 13:40:45 2004
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 11, op = fork, loc = skgpspawn5
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> Tue Mar 23 13:42:02 2004
> skgpspawn failed:category = 27142, depinfo = 12, op = fork, loc = skgpspawn3
> :q

发现数据库多次重起，并记录了部分错误信息

该提示说明数据库无法 spawn a new session.

quote Yong Huang 's comment:

The number in "skgpspawn failed:category = 27142" is probably ORA error:

$ oerr ora 27142
27142, 0000, "could not create new process"
// *Cause: OS system call
// *Action: check errno and if possible increase the number of processes

OSD (OS-dependent) errors are almost always shown as an skg... error (probably means "system, kernel generic").

I don't know what "depinfo = 12" means.

4.尝试连接数据库

收到错误信息，无法连接数据库

>
> $ sqlplus "/ as sysdba" > > SQLPlus: Release 9.2.0.3.0 - Production on 星期二 3月 23 14:14:06 2004 > > Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved. > > ERROR:
> ORA-12540: TNS: 超出内部限制
> > > 请输入用户名:
> ERROR:
> ORA-12540: TNS: 超出内部限制
> > > 请输入用户名:
> ERROR:
> ORA-12540: TNS: 超出内部限制
> > > SP2-0157: 在3次尝试之后无法 CONNECT 到 ORACLE, 退出 SQLPlus >
>
>

内部限制超过，通常说明某些系统资源不足.

5.检查监听器

发现部分连接被拒绝

> /export/home1/oracle>lsnrctl services > > LSNRCTL for Solaris: Version 9.2.0.3.0 - Production on 23-3月 -2004 14:37:23 > > Copyright (c) 1991, 2002, Oracle Corporation. All rights reserved. > > 正在连接到 (DESCRIPTION=(ADDRESS=(PROTOCOL=IPC)(KEY=EXTPROC)))
> 服务摘要..
> 服务 "PLSExtProc" 包含 1 个例程。
> 例程 "PLSExtProc", 状态 UNKNOWN, 包含此服务的 1 个处理程序...
> 处理程序:
> "DEDICATED" 已建立:0 已被拒绝:0
> LOCAL SERVER
> 服务 "hswapdb.eygle.com" 包含 2 个例程。
> 例程 "hswapdb", 状态 UNKNOWN, 包含此服务的 1 个处理程序...
> 处理程序:
> "DEDICATED" 已建立:6 已被拒绝:0
> LOCAL SERVER
> 例程 "hswapdb", 状态 READY, 包含此服务的 1 个处理程序...
> 处理程序:
> "DEDICATED" 已建立:21 已拒绝:6 状态:ready
> LOCAL SERVER
> 服务 "hswapdbXDB.eygle.com" 包含 1 个例程。
> 例程 "hswapdb", 状态 READY, 包含此服务的 1 个处理程序...
> 处理程序:
> "D000" 已建立:0 已被拒绝:0 当前: 0 最大: 972 状态: ready
> DISPATCHER

  1<machine: 25839="" pid:="" wapplatform,="">   
  2&gt;  (ADDRESS=(PROTOCOL=tcp)(HOST=wapplatform)(PORT=32869))   
  3&gt;  命令执行成功 
  4&gt;     
  5&gt;     
  6&gt;        
  7  
  8---  
  9  
 10在listener.log中找到了相关错误信息 
 11
 12&gt; 23-3\324\302 -2004 12:19:40 * (CONNECT_DATA=(SID=hswapdb)(CID=(PROGRAM=C:\WINNT\Microsoft.NET\Framework\v1.1.4322\aspnet_wp.e   
 13&gt;  xe)(HOST=SWAN)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=172.16.30.125)(PORT=1291)) * establish * hswapdb * 12500   
 14&gt;  TNS-12500: TNS\243\272\274\340\314\375\306\367\316\264\304\334\306\364\266\257\327\250\323\303\265\304\267\376\316\361\306\36   
 15&gt;  7\275\370\263\314   
 16&gt;  TNS-12540: TNS\243\272\263\254\263\366\304\332\262\277\274\253\317\336\317\336\326\306   
 17&gt;  TNS-12560: TNS: \320\255\322\351\312\312\305\344\306\367\264\355\316\363   
 18&gt;  TNS-00510: \263\254\263\366\304\332\262\277\274\253\317\336\317\336\326\306   
 19&gt;  Solaris Error: 12: Not enough space    
 20&gt;  23-3\324\302 -2004 12:19:50 * (CONNECT_DATA=(SID=hswapdb)(CID=(PROGRAM=C:\Program Files\PLSQL Developer\PLSQLDev.exe)(HOST=SW   
 21&gt;  AN)(USER=Administrator))) * (ADDRESS=(PROTOCOL=tcp)(HOST=172.16.30.125)(PORT=1292)) * establish * hswapdb * 12500   
 22&gt;  TNS-12500: TNS\243\272\274\340\314\375\306\367\316\264\304\334\306\364\266\257\327\250\323\303\265\304\267\376\316\361\306\36   
 23&gt;  7\275\370\263\314   
 24&gt;  TNS-12540: TNS\243\272\263\254\263\366\304\332\262\277\274\253\317\336\317\336\326\306   
 25&gt;  TNS-12560: TNS: \320\255\322\351\312\312\305\344\306\367\264\355\316\363   
 26&gt;  TNS-00510: \263\254\263\366\304\332\262\277\274\253\317\336\317\336\326\306   
 27&gt;  Solaris Error: 12: Not enough space    
 28&gt;    
 29&gt;  /export/home1/oracle/app/network/log&gt;grep -w 12 /usr/include/sys/errno.h   
 30&gt;  #define ENOMEM 12 /* Not enough core 
 31&gt;     
 32&gt;     
 33&gt;        
 34  
 35---  
 36  
 37quote  Yong Huang  's comment: 
 38
 39$ grep -w 12 /usr/include/sys/errno.h   
 40#define ENOMEM 12 /* Not enough core */   
 41  
 42Here "core" means memory, including real RAM memory and swap space.   
 43
 44
 456.退出Oracle用户检查   
 46  
 47检查系统日志信息，发现大量失败的su操作   
 48有swap区不足的报告 
 49
 50&gt; 
 51&gt;     /export/home1/oracle/admin/hswapdb/bdump&gt;exit  
 52&gt;     &gt; 
 53&gt;                 wapplatform:/&gt;dmesg
 54&gt; 
 55&gt; 2004年03月23日 星期二 14时00分32秒 CST   
 56&gt;  Mar 22 22:52:36 wapplatform elfexec: [ID 700856 kern.notice] ps: Cannot find ^?ELF^A^B^A   
 57&gt;  Mar 22 22:53:00 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 58&gt;  Mar 22 22:53:09 wapplatform elfexec: [ID 700856 kern.notice] w: Cannot find ^?ELF^A^B^A   
 59&gt;  Mar 22 22:53:53 wapplatform last message repeated 4 times   
 60&gt;  Mar 22 22:56:28 wapplatform elfexec: [ID 700856 kern.notice] ipnat: Cannot find ^?ELF^B^B^A   
 61&gt;  Mar 22 22:58:00 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 62&gt;  Mar 22 22:59:54 wapplatform elfexec: [ID 700856 kern.notice] ipnat: Cannot find ^?ELF^B^B^A   
 63&gt;  Mar 22 23:02:26 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 64&gt;  Mar 22 23:03:00 wapplatform last message repeated 1 time   
 65&gt;  Mar 22 23:08:00 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 66&gt;  Mar 22 23:08:34 wapplatform elfexec: [ID 700856 kern.notice] ipnat: Cannot find ^?ELF^B^B^A   
 67&gt;  Mar 22 23:10:27 wapplatform last message repeated 3 times   
 68&gt;  Mar 22 23:11:49 wapplatform elfexec: [ID 700856 kern.notice] ipnat: Cannot find ^?ELF^B^B^A   
 69&gt;  Mar 22 23:11:52 wapplatform last message repeated 1 time   
 70&gt;  Mar 22 23:13:01 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 71&gt;  Mar 22 23:18:01 wapplatform last message repeated 1 time   
 72&gt;  Mar 22 23:23:01 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 73&gt;  Mar 22 23:28:01 wapplatform last message repeated 1 time   
 74&gt;  Mar 22 23:33:01 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 75&gt;  Mar 22 23:38:01 wapplatform last message repeated 1 time   
 76&gt;  Mar 22 23:43:01 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 77&gt;  Mar 22 23:48:01 wapplatform last message repeated 1 time   
 78&gt;  Mar 22 23:53:01 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 79&gt;  Mar 22 23:58:01 wapplatform last message repeated 1 time   
 80&gt;  Mar 23 00:00:00 wapplatform ufs: [ID 213553 kern.notice] NOTICE: realloccg /export/home1: file system full   
 81&gt;  Mar 23 00:00:00 wapplatform sendmail[3075]: [ID 702911 mail.crit] My unqualified host name (wapplatform) unknown; sleeping 
 82&gt; 
 83&gt; for retry   
 84&gt;  Mar 23 00:01:00 wapplatform sendmail[3075]: [ID 702911 mail.alert] unable to qualify my own domain name (wapplatform) -- 
 85&gt; 
 86&gt; using short name   
 87&gt;  Mar 23 00:02:36 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 88&gt;  Mar 23 00:03:02 wapplatform last message repeated 1 time   
 89&gt;  Mar 23 00:08:02 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 90&gt;  .... 
 91&gt; 
 92&gt; Mar 23 10:18:15 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 93&gt;  Mar 23 10:20:41 wapplatform ufs: [ID 213553 kern.notice] NOTICE: realloccg /export/home1: file system full   
 94&gt;  Mar 23 10:20:47 wapplatform last message repeated 1 time   
 95&gt;  Mar 23 10:23:15 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
 96&gt;  Mar 23 10:24:38 wapplatform ufs: [ID 213553 kern.notice] NOTICE: realloccg /export/home1: file system full   
 97&gt;  Mar 23 10:24:43 wapplatform last message repeated 1 time   
 98&gt;  Mar 23 10:24:55 wapplatform ufs: [ID 213553 kern.notice] NOTICE: realloccg /export/home1: file system full   
 99&gt;  Mar 23 10:25:06 wapplatform last message repeated 2 times   
100&gt;  Mar 23 11:09:31 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3118 (su)   
101&gt;  Mar 23 11:09:39 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3121 (su)   
102&gt;  Mar 23 11:10:48 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3137 (su)   
103&gt;  Mar 23 11:18:02 wapplatform sshd[3620]: [ID 800047 auth.error] error: grantpt: Not enough space   
104&gt;  Mar 23 11:18:02 wapplatform sshd[3620]: [ID 800047 auth.error] error: session_pty_req: session 0 alloc failed   
105&gt;  Mar 23 11:18:43 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3636 (su)   
106&gt;  Mar 23 11:19:47 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3672 (su)   
107&gt;  Mar 23 11:20:20 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3694 (su)   
108&gt;  Mar 23 11:22:23 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3736 (sshd)   
109&gt;  Mar 23 11:23:17 wapplatform tmpfs: [ID 518458 kern.warning] WARNING: /tmp: File system full, swap space limit exceeded   
110&gt;  Mar 23 11:23:40 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3804 (su)   
111&gt;  Mar 23 11:23:40 wapplatform last message repeated 8 times   
112&gt;  Mar 23 11:23:56 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3806 (ps)   
113&gt;  Mar 23 11:23:56 wapplatform last message repeated 12 times   
114&gt;  Mar 23 11:24:01 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3808 (w)   
115&gt;  Mar 23 11:24:01 wapplatform last message repeated 8 times   
116&gt;  Mar 23 13:40:56 wapplatform su: [ID 810491 auth.crit] 'su root' failed for root on /dev/pts/2   
117&gt;  Mar 23 13:46:26 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 24888 
118&gt; 
119&gt; (sqlplus)   
120&gt;  Mar 23 13:49:18 wapplatform su: [ID 810491 auth.crit] 'su oracle' failed for root on /dev/pts/6   
121&gt;  Mar 23 13:54:03 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 25035 (su)   
122&gt;  Mar 23 13:54:08 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 25036 (su)   
123  
124---  
125  
126现在基本可以判断是交换区的问题，当然和Oracle SGA设置有关. 
127
1287.检查系统内存及交换区使用   
129
130
131&gt; 
132&gt;     /export/home1/oracle/admin/hswapdb/bdump&gt;exit  
133&gt;     &gt; 
134&gt;                 wapplatform:/&gt;dmesg
135&gt; 
136&gt; 2004年03月23日 星期二 14时00分32秒 CST   
137&gt;  Mar 22 22:52:36 wapplatform elfexec: [ID 700856 kern.notice] ps: Cannot find ^?ELF^A^B^A   
138&gt;  Mar 22 22:53:00 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
139&gt;  Mar 22 22:53:09 wapplatform elfexec: [ID 700856 kern.notice] w: Cannot find ^?ELF^A^B^A   
140&gt;  Mar 22 22:53:53 wapplatform last message repeated 4 times   
141&gt;  Mar 22 22:56:28 wapplatform elfexec: [ID 700856 kern.notice] ipnat: Cannot find ^?ELF^B^B^A   
142&gt;  Mar 22 22:58:00 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
143&gt;  Mar 22 22:59:54 wapplatform elfexec: [ID 700856 kern.notice] ipnat: Cannot find ^?ELF^B^B^A   
144&gt;  Mar 22 23:02:26 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
145&gt;  Mar 22 23:03:00 wapplatform last message repeated 1 time   
146&gt;  Mar 22 23:08:00 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
147&gt;  Mar 22 23:08:34 wapplatform elfexec: [ID 700856 kern.notice] ipnat: Cannot find ^?ELF^B^B^A   
148&gt;  Mar 22 23:10:27 wapplatform last message repeated 3 times   
149&gt;  Mar 22 23:11:49 wapplatform elfexec: [ID 700856 kern.notice] ipnat: Cannot find ^?ELF^B^B^A   
150&gt;  Mar 22 23:11:52 wapplatform last message repeated 1 time   
151&gt;  Mar 22 23:13:01 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
152&gt;  Mar 22 23:18:01 wapplatform last message repeated 1 time   
153&gt;  Mar 22 23:23:01 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
154&gt;  Mar 22 23:28:01 wapplatform last message repeated 1 time   
155&gt;  Mar 22 23:33:01 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
156&gt;  Mar 22 23:38:01 wapplatform last message repeated 1 time   
157&gt;  Mar 22 23:43:01 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
158&gt;  Mar 22 23:48:01 wapplatform last message repeated 1 time   
159&gt;  Mar 22 23:53:01 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
160&gt;  Mar 22 23:58:01 wapplatform last message repeated 1 time   
161&gt;  Mar 23 00:00:00 wapplatform ufs: [ID 213553 kern.notice] NOTICE: realloccg /export/home1: file system full   
162&gt;  Mar 23 00:00:00 wapplatform sendmail[3075]: [ID 702911 mail.crit] My unqualified host name (wapplatform) unknown; sleeping 
163&gt; 
164&gt; for retry   
165&gt;  Mar 23 00:01:00 wapplatform sendmail[3075]: [ID 702911 mail.alert] unable to qualify my own domain name (wapplatform) -- 
166&gt; 
167&gt; using short name   
168&gt;  Mar 23 00:02:36 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
169&gt;  Mar 23 00:03:02 wapplatform last message repeated 1 time   
170&gt;  Mar 23 00:08:02 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
171&gt;  .... 
172&gt; 
173&gt; Mar 23 10:18:15 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
174&gt;  Mar 23 10:20:41 wapplatform ufs: [ID 213553 kern.notice] NOTICE: realloccg /export/home1: file system full   
175&gt;  Mar 23 10:20:47 wapplatform last message repeated 1 time   
176&gt;  Mar 23 10:23:15 wapplatform ufs: [ID 845546 kern.notice] NOTICE: alloc: /export/home1: file system full   
177&gt;  Mar 23 10:24:38 wapplatform ufs: [ID 213553 kern.notice] NOTICE: realloccg /export/home1: file system full   
178&gt;  Mar 23 10:24:43 wapplatform last message repeated 1 time   
179&gt;  Mar 23 10:24:55 wapplatform ufs: [ID 213553 kern.notice] NOTICE: realloccg /export/home1: file system full   
180&gt;  Mar 23 10:25:06 wapplatform last message repeated 2 times   
181&gt;  Mar 23 11:09:31 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3118 (su)   
182&gt;  Mar 23 11:09:39 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3121 (su)   
183&gt;  Mar 23 11:10:48 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3137 (su)   
184&gt;  Mar 23 11:18:02 wapplatform sshd[3620]: [ID 800047 auth.error] error: grantpt: Not enough space   
185&gt;  Mar 23 11:18:02 wapplatform sshd[3620]: [ID 800047 auth.error] error: session_pty_req: session 0 alloc failed   
186&gt;  Mar 23 11:18:43 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3636 (su)   
187&gt;  Mar 23 11:19:47 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3672 (su)   
188&gt;  Mar 23 11:20:20 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3694 (su)   
189&gt;  Mar 23 11:22:23 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3736 (sshd)   
190&gt;  Mar 23 11:23:17 wapplatform tmpfs: [ID 518458 kern.warning] WARNING: /tmp: File system full, swap space limit exceeded   
191&gt;  Mar 23 11:23:40 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3804 (su)   
192&gt;  Mar 23 11:23:40 wapplatform last message repeated 8 times   
193&gt;  Mar 23 11:23:56 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3806 (ps)   
194&gt;  Mar 23 11:23:56 wapplatform last message repeated 12 times   
195&gt;  Mar 23 11:24:01 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 3808 (w)   
196&gt;  Mar 23 11:24:01 wapplatform last message repeated 8 times   
197&gt;  Mar 23 13:40:56 wapplatform su: [ID 810491 auth.crit] 'su root' failed for root on /dev/pts/2   
198&gt;  Mar 23 13:46:26 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 24888 
199&gt; 
200&gt; (sqlplus)   
201&gt;  Mar 23 13:49:18 wapplatform su: [ID 810491 auth.crit] 'su oracle' failed for root on /dev/pts/6   
202&gt;  Mar 23 13:54:03 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 25035 (su)   
203&gt;  Mar 23 13:54:08 wapplatform genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 25036 (su)   
204  
205---  
206  
207  
208现在基本可以判断是交换区的问题，当然和Oracle SGA设置有关.   
209  
2107.检查系统内存及交换区使用   
211  
212
213
214&gt; 
215&gt;     $ top
216&gt;     
217&gt;     last pid: 25456; load averages: 0.67, 0.70, 0.69 
218&gt;     
219&gt;     14:10:03
220&gt;     93 processes: 91 sleeping, 2 on cpu
221&gt;     CPU states: 72.7% idle, 14.9% user, 2.7% kernel, 9.7% iowait, 0.0% swap
222&gt;     Memory: 1024M real, 34M free, 752M swap in use, 10M swap free
223&gt;     
224&gt;     PID USERNAME THR PRI NICE SIZE RES STATE TIME CPU COMMAND
225&gt;     25199 oracle 1 40 0 674M 631M cpu/2 8:03 16.32% oracle
226&gt;     25209 oracle 1 30 0 675M 630M sleep 0:03 0.13% oracle
227&gt;     25159 oracle 1 48 0 674M 628M sleep 0:03 0.06% oracle
228&gt;     25384 oracle 1 58 0 2632K 1736K cpu/0 0:01 0.05% top
229&gt;     25145 oracle 143 58 0 682M 630M sleep 0:01 0.03% oracle
230&gt;     25446 oracle 1 58 0 674M 628M sleep 0:00 0.03% oracle
231&gt;     25149 oracle 15 58 0 682M 626M sleep 0:00 0.02% oracle
232&gt;     25075 oracle 1 48 0 17M 7208K sleep 0:00 0.01% tnslsnr
233&gt;     25151 oracle 11 58 0 676M 624M sleep 0:00 0.01% oracle
234&gt;     25366 oracle 1 10 0 674M 628M sleep 0:00 0.00% oracle
235&gt;     25356 oracle 1 18 0 674M 628M sleep 0:00 0.00% oracle
236&gt;     25360 oracle 1 20 0 674M 628M sleep 0:00 0.00% oracle
237&gt;     25364 oracle 1 20 0 674M 628M sleep 0:00 0.00% oracle
238&gt;     25362 oracle 1 20 0 674M 628M sleep 0:00 0.00% oracle
239&gt;     25330 oracle 1 28 0 674M 628M sleep 0:00 0.00% oracle
240&gt;     
241&gt;       
242  
243---  
244  
245发现物理内存仅为1G，free部分为34M，交换区使用了752M，仅10M free   
246系统内存严重不足，Swap区不足 
247
2488\. 检查数据库的SGA设置   
249  
250发现SGA设置为: 622299344 bytes   
251接近600M   
252
253
254&gt; 
255&gt;       
256&gt;     &gt; wapplatform:/&gt;su - oracle  
257&gt;     &gt; Sun Microsystems Inc. SunOS 5.8 Generic Patch October 2001  
258&gt;     &gt; You have new mail.  
259&gt;     &gt; /export/home1/oracle&gt;sqlplus "/ as sysdba"
260&gt; 
261&gt; SQL*Plus: Release 9.2.0.3.0 - Production on 星期二 3月 23 14:02:30 2004 
262&gt; 
263&gt; Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved.   
264&gt; 
265&gt; 
266&gt; 连接到:   
267&gt;  Oracle9i Enterprise Edition Release 9.2.0.3.0 - 64bit Production   
268&gt;  With the Partitioning, OLAP and Oracle Data Mining options   
269&gt;  JServer Release 9.2.0.3.0 - Production 
270&gt; 
271&gt; SQL&gt; show sga 
272&gt; 
273&gt; Total System Global Area 622299344 bytes   
274&gt;  Fixed Size 731344 bytes   
275&gt;  Variable Size 268435456 bytes   
276&gt;  Database Buffers 352321536 bytes   
277&gt;  Redo Buffers 811008 bytes   
278&gt;  SQL&gt;  
279  
280---  
281  
282  
283  
284对于RAM小于1G的系统，Dedicated模式下,Oracle的SGA一般不应超过1/2物理内存. 
285
2869.第一步调整   
287减小SGA，为系统保留足够的内存.   
288  
28910.增加swap区   
290
291
292&gt; 
293&gt;       
294&gt;     &gt;   
295&gt;     &gt; wapplatform:/&gt;df -k  
296&gt;     &gt; 文件系统 千字节 用了 可用 容量 挂接在  
297&gt;     &gt; /dev/dsk/c0t1d0s0 3099093 105421 2931691 4% /  
298&gt;     &gt; /dev/dsk/c0t2d0s0 10325760 8359637 1862866 82% /usr  
299&gt;     &gt; /proc 0 0 0 0% /proc  
300&gt;     &gt; fd 0 0 0 0% /dev/fd  
301&gt;     &gt; mnttab 0 0 0 0% /etc/mnttab  
302&gt;     &gt; /dev/dsk/c0t1d0s3 1018382 285914 671366 30% /var  
303&gt;     &gt; swap 3904 24 3880 1% /var/run  
304&gt;     &gt; swap 3936 56 3880 2% /tmp  
305&gt;     &gt; /dev/dsk/c0t1d0s5 1671823 459202 1162467 29% /opt  
306&gt;     &gt; /dev/dsk/c0t2d0s7 7087473 6068462 948137 87% /export/home  
307&gt;     &gt; /dev/dsk/c2t1d0s7 17413250 15900222 1338896 93% /export/home2  
308&gt;     &gt; /dev/dsk/c0t3d0s7 17413250 13749782 3489336 80% /export/home1  
309&gt;     &gt; /dev/dsk/c0t1d0s1 771110 382410 334723 54% /usr/openwin  
310&gt;     &gt; /export/home/wapgw/luke  
311&gt;     &gt; 7087473 6068462 948137 87% /home/wap
312&gt; 
313&gt; wapplatform:/var/swap&gt;cd /export/home1   
314&gt;  wapplatform:/export/home1&gt;ls   
315&gt;  TT_DB lost+found oracle oracli9   
316&gt;  wapplatform:/export/home1&gt;mkdir swap   
317&gt;  wapplatform:/export/home1&gt;cd swap   
318&gt;  wapplatform:/export/home1/swap&gt;mkfile -v 1g swapfile1   
319&gt;  swapfile1 1073741824 bytes   
320&gt;  wapplatform:/export/home1/swap&gt;id   
321&gt;  uid=0(root) gid=1(other)   
322&gt;  wapplatform:/export/home1/swap&gt;swap -a /export/home1/swap/swapfile1   
323&gt;  wapplatform:/export/home1/swap&gt;swap -s   
324&gt;  总数：分配了 623160k 字节 + 保留 162704k = 已使用 785864k，1010936k 可用   
325  
326---  
327  
328  
32911.连接测试   
330  
331系统恢复正常，问题解决 
332
333&gt; 
334&gt;       
335&gt;     &gt;   
336&gt;     &gt; wapplatform:/export/home1/swap&gt;su - oracle  
337&gt;     &gt; Sun Microsystems Inc. SunOS 5.8 Generic Patch October 2001  
338&gt;     &gt; You have new mail.  
339&gt;     &gt; /export/home1/oracle&gt;sqlplus "/ as sysdba"
340&gt; 
341&gt; SQL*Plus: Release 9.2.0.3.0 - Production on 星期四 3月 25 11:56:28 2004 
342&gt; 
343&gt; Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved.   
344&gt; 
345&gt; 
346&gt; 连接到:   
347&gt;  Oracle9i Enterprise Edition Release 9.2.0.3.0 - 64bit Production   
348&gt;  With the Partitioning, OLAP and Oracle Data Mining options   
349&gt;  JServer Release 9.2.0.3.0 - Production 
350&gt; 
351&gt; SQL&gt; exit   
352&gt;  从Oracle9i Enterprise Edition Release 9.2.0.3.0 - 64bit Production   
353&gt;  With the Partitioning, OLAP and Oracle Data Mining options   
354&gt;  JServer Release 9.2.0.3.0 - Production中断开   
355&gt;  /export/home1/oracle&gt;top 
356&gt; 
357&gt; last pid: 5372; load averages: 0.25, 0.22, 0.29 
358&gt; 
359&gt; 11:57:58   
360&gt;  148 processes: 137 sleeping, 9 zombie, 2 on cpu   
361&gt;  CPU states: 98.8% idle, 0.2% user, 0.7% kernel, 0.2% iowait, 0.0% swap   
362&gt;  Memory: 1024M real, 17M free, 824M swap in use, 934M swap free 
363&gt; 
364&gt; PID USERNAME THR PRI NICE SIZE RES STATE TIME CPU COMMAND   
365&gt;  5363 root 1 58 0 2680K 1736K sleep 0:00 0.24% top   
366&gt;  5370 oracle 1 58 0 514M 469M sleep 0:00 0.18% oracle   
367&gt;  5366 oracle 1 28 0 514M 469M sleep 0:00 0.11% oracle   
368&gt;  5341 oracle 1 58 0 2680K 1736K cpu/2 0:00 0.10% top   
369&gt;  5372 oracle 1 48 0 61M 3288K cpu/3 0:00 0.06% oracle   
370&gt;  1288 oracle 1 48 0 514M 468M sleep 5:33 0.05% oracle   
371&gt;  607 root 12 48 0 2768K 2312K sleep 1:48 0.03% mibiisa   
372&gt;  25075 oracle 1 48 0 17M 7208K sleep 0:16 0.02% tnslsnr   
373&gt;  1278 oracle 15 58 0 522M 466M sleep 0:49 0.02% oracle   
374&gt;  374 root 11 53 0 3504K 2888K sleep 0:16 0.01% nscd   
375&gt;  1280 oracle 19 58 0 518M 466M sleep 0:28 0.00% oracle   
376&gt;  5361 root 1 46 0 1024K 680K sleep 0:00 0.00% sleep   
377&gt;  5362 root 1 46 0 1024K 680K sleep 0:00 0.00% sleep   
378&gt;  5469 root 1 36 0 1952K 1176K sleep 30:09 0.00% monithttp   
379&gt;  4167 oracle 1 40 0 515M 471M sleep 29:38 0.00% oracle   
380  
381---  
382  
383  
384问题总结:   
385  
386Oracle数据库问题的解决从来就离不开操作系统   
387  
388很多时候我们必须通过操作系统一级的手段来诊断并解决问题.   
389  
390关于操作系统   
391  
392一般Swap区的推荐值为2XRAM   
393如果Ram很大，不一定非要把Swap设置为2xSwap   
394但是通常至少设置Swap = Ram   
395  
396如果Swap区过小，在系统繁忙期间   
397产生大量交换无法换到磁盘，就会出现问题.   
398如本案例就是这样。   
399  
400另外，如果系统Ram较小   
401通常设置SGA &lt; 1/2 Ram   
402  
403要为Server process及OS保留足够的内存空间.</machine:>