数据库报ora-00603 ora-ora27504 (转)

环境是: OS:AIX 5304 HA:5.2 ORACLE 9206 RAC 2 NODES ALERT日志报错: ORA-00603: ORACLE server session terminated by fatal error ORA-27504: IPC error creating OSD context ORA-27300: OS system dependent operation:socket failed with status: 11 ORA-27301: OS failure message: Resource temporarily unavailable ORA-27302: failure occurred at: sskgxpcre1 Wed Jun 27 11:44:33 2007 Errors in file /oracle/app/oracle/admin/settledb/udump/settle1_ora_528530.trc: ORA-00603: ORACLE server session terminated by fatal error ORA-27504: IPC error creating OSD context ORA-27300: OS system dependent operation:socket failed with status: 11 ORA-27301: OS failure message: Resource temporarily unavailable ORA-27302: failure occurred at: sskgxpcre1 Wed Jun 27 11:44:33 2007 Errors in file /oracle/app/oracle/admin/settledb/udump/settle1_ora_3871064.trc: ORA-00603: ORACLE server session terminated by fatal error ORA-27504: IPC error creating OSD context ORA-27300: OS system dependent operation:socket failed with status: 11 ORA-27301: OS failure message: Resource temporarily unavailable Redo thread mounted by this instance: 1 Oracle process number: 0 造成用户无法使用,我用SQLPLUS连接时报ORACLE目录没有空间,df -k查看使用率100% rm 掉先前的TRACE文件,但是数据库实时的在产生TRACE文件,速度很快,过了一会,终于连上了,试着关闭数据库,成功关闭,重新启动还是报同样的错误,有些不解,近期没有对程序进行修改,为什么会出现这种问题,大约过了一个小时,数据库自动的就好了,我没有对数据库进行任何修改.更有些不解 接下来,到METALINK上去查找相关的资料,得出可能是由于操做系统的一个参数设置问题.但不能确定,后来公司申请了原厂,第二天跟原厂工程师一起查找原因,最终确定是由于操做系统的参数rfc1323设置问题,具体如下: 相关文档 bug#4365217 bug#4069090 bug#4060597 bug#3400991 1，ORA-00603 ORACLE server session terminated by fatal error Cause: An Oracle Server session is in an unrecoverable state. Action: Log in to Oracle again so a new server session will be created automatically. Examine the session trace file for more information. 2.ORA-27300 OS system dependent operation:string failed with status: string Cause: Operating system call error. Action: Contact Oracle Support Services ORA-27301 OS failure message: string Cause: Operating system call error. Action: Contact Oracle Support Services. metalink 相关文档: Bug No. 4060597 TEST ENVIRONMENT: OS: AIX5.1 PRODUCT AND SHIPHOME: DB9206-RC3 Locale : ja_JP Cluster Enviroment : 2 nodes Install files Path : /net/psaixb01/project/aix5l64/920/install/RC3/ ship.patch/9206_aix5l_release.zip PROBLEM: After applying 9206 onto 9205, the following error occurs when try to startup database: ORA-00603: ORACLE server session terminated by fatal error. WORKAROUND: After changing 5 network params as bug3400991 mentioned, db can be started. TEST CASE: 1. Install db9201 without error. 2. Apply db9205 onto db9201. 3. Create new database on db9205. 4. Perform catpatch.sql to update database conponent to db9205. (due to bug3113355) Confirmed that database can be started and shutdown without error. 5. Apply db9206 onto db9205. 6. Issue either of the following commands, error ORA-00603 will occur: SQL> startup SQL> startup nomount After changing the following network configuration, database can be started: /usr/sbin/no -o tcp_sendspace=65536 /usr/sbin/no -o tcp_recvspace=65536 /usr/sbin/no -o udp_sendspace=65536 /usr/sbin/no -o udp_recvspace=655360 /usr/sbin/no -o rfc1323=1 ========================================================================= Bug No. 4365217 Filed 11-MAY-2005 Updated 27-NOV-2006 Product PLE-TOOLS Product Version 9.2.0.6 Platform AIX5L Based Systems (64-bit) Platform Version 9.2.0.6 Database Version N/A Affects Platforms Port-Specific Severity Complete Loss of Service Status To Filer for Review Base Bug N/A Fixed in Product Version 9.2.0.6 *** SWHITMAN 05/11/05 02:46 pm *** @ I was not able to bring up an AIX 9206 RAC database on ddribm9 today. @ This is is the first time I have tried this. The alert is showing: @ . @ Wed May 11 13:20:16 2005 @ Errors in file /ddr/aix/plenv/srchome117/rdbms/log/st920117_ora_40414.trc: @ ORA-00603: ORACLE server session terminated by fatal error @ ORA-27504: IPC error creating OSD context @ ORA-27300: OS system dependent operation:sendmsg failed with status: 59 @ ORA-27301: OS failure message: Message too long @ ORA-27302: failure occurred at: sskgxpsnd1 @ . @ And the trace file: @ . @ *** 2005-05-11 13:20:16.122 @ SKGXPVFYMMTU: Could not create context which satisfies the minimum @ MTU requirement of 32768 bytes @ kgefec: fatal error 0 @ *** 2005-05-11 13:20:16.123 @ ksedmp: internal or fatal error @ ORA-00603: ORACLE server session terminated by fatal error @ ORA-27504: IPC error creating OSD context @ ORA-27300: OS system dependent operation:sendmsg failed with status: 59 @ ORA-27301: OS failure message: Message too long @ ORA-27302: failure occurred at: sskgxpsnd1 @ Current SQL information unavailable - no session @ . @ This looks similar to bug 3400991 which looks not to apply to @ 9206. Web-iv is showing several notes on this behavior in 9206 @ though. The solution seems to involve changing kernel parameters. @ However I am not sure what the changes should be for aix and @ if the changes for 9206 will influence the pre-9206 environments @ we have. Notes, 3400991.8, 304171.1, 295832.1 discuss similar @ issues on 9206. The suggestions basically involve setting: @ . @ tcp_sendspace=65536 @ tcp_recvspace=65536 @ udp_sendspace=65536 @ udp_recvspace=655360[@more@]

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/9650775/viewspace-923260/，如需转载，请注明出处，否则将追究法律责任。

转载于:http://blog.itpub.net/9650775/viewspace-923260/