溫馨提示×
您好，登錄后才能下訂單哦！
點擊 登錄注冊 即表示同意《億速云用戶服務條款》
您好，登錄后才能下訂單哦！
Oracle Study之案例--Oracle Sqlplus錯誤
系統環境：
操作系統：AIX5.3
數據庫： Oracle 10gR2
案例分析：
1、查看空間信息 [oracle@aix220 ~]$df -m Filesystem MB blocks Free %Used Iused %Iused Mounted on /dev/hd4 17408.00 1238.15 93% 37699 12% / /dev/hd2 8192.00 6310.39 23% 46534 4% /usr /dev/hd9var 2048.00 1725.55 16% 1704 1% /var /dev/hd3 2048.00 1902.58 8% 420 1% /tmp /dev/fwdump 3072.00 3071.21 1% 4 1% /var/adm/ras/platform /dev/hd1 2048.00 1821.98 12% 74 1% /home /proc - - - - - /proc /dev/hd10opt 2048.00 1374.77 33% 8934 3% /opt /dev/lv00 1024.00 991.80 4% 18 1% /var/adm/csd /dev/arch 10240.00 8347.63 19% /dev/lv02 15360.00 10097.80 35% 123693 4% /u01 2、查看文件權限 [oracle@aix220 ~]$ls -ld /u01/app/oracle/admin/master/ drwxr-x--- 8 oracle dba 512 Feb 17 16:40 /u01/app/oracle/admin/master/ [oracle@aix220 ~]$ls -ld /u01/app/oracle/admin/master/adump/ drwxr-x--- 2 oracle dba 1536 Feb 17 19:48 /u01/app/oracle/admin/master/adump/ 3、查看文件系統信息 [oracle@aix220 dbs]$cat /etc/filesystems /u01: dev = /dev/lv02 vfs = jfs log = /dev/loglv00 mount = true options = rw account = false [root@aix220 /]#lsvg -l oraclevg oraclevg: LV NAME TYPE LPs PPs PVs LV STATE MOUNT POINT loglv00 jfslog 1 1 1 closed/syncd N/A lv02 jfs 120 120 1 closed/syncd /u01 [root@aix220 /]#df -m Filesystem MB blocks Free %Used Iused %Iused Mounted on /dev/hd4 17408.00 1238.18 93% 37700 12% / /dev/hd2 8192.00 6310.39 23% 46534 4% /usr /dev/hd9var 2048.00 1725.56 16% 1704 1% /var /dev/hd3 2048.00 1902.58 8% 420 1% /tmp /dev/fwdump 3072.00 3071.21 1% 4 1% /var/adm/ras/platform /dev/hd1 2048.00 1821.98 12% 74 1% /home /proc - - - - - /proc /dev/hd10opt 2048.00 1374.77 33% 8934 3% /opt /dev/lv00 1024.00 991.80 4% 18 1% /var/adm/csd /dev/arch 10240.00 8347.63 19% 52 1% /arch /dev/lv01 5120.00 4959.25 4% 16 1% /flash /dev/lv03 25600.00 24796.43 4% 17 1% /home/oracle/arch_master rhel152:/backup/cuug15/storage30 14111.05 13052.23 8% 31 1% /backup mount文件系統失敗： [root@aix220 /]#mount /u01 Replaying log for /dev/lv02. mount: /dev/lv02 on /u01: Unformatted or incompatible media The superblock on /dev/lv02 is dirty. Run a full fsck to fix. 修復文件系統： [root@aix220 /]#fsck -y /dev/lv02 ...... ** Phase 5 - Check Inode Map ** Phase 6 - Check Block Map Bad Block Map (SALVAGED) ** Phase 6b - Salvage Block Map Superblock is marked dirty (FIXED) 123665 files 10749336 blocks 20707944 free ***** Filesystem was modified ***** mount文件系統成功： [root@aix220 /]#mount /u01 [root@aix220 /]#df -m Filesystem MB blocks Free %Used Iused %Iused Mounted on /dev/hd4 17408.00 1238.16 93% 37699 12% / /dev/hd2 8192.00 6310.39 23% 46534 4% /usr /dev/hd9var 2048.00 1725.55 16% 1704 1% /var /dev/hd3 2048.00 1902.58 8% 420 1% /tmp /dev/fwdump 3072.00 3071.21 1% 4 1% /var/adm/ras/platform /dev/hd1 2048.00 1821.98 12% 74 1% /home /proc - - - - - /proc /dev/hd10opt 2048.00 1374.77 33% 8934 3% /opt /dev/lv00 1024.00 991.80 4% 18 1% /var/adm/csd /dev/lv02 15360.00 10111.30 35% 123674 4% /u01 執行sqlplus命令成功： [root@aix220 /]#su - oracle [oracle@aix220 ~]$sqlplus '/as sysdba' SQL*Plus: Release 10.2.0.1.0 - Production on Tue Feb 17 19:46:03 2015 Copyright (c) 1982, 2005, Oracle. All rights reserved. Connected to an idle instance. SQL>
數據庫庫報ORA-09925: Unable to create audit trail file，當時查 df -h有可用空間，目錄的權限也正確，未df -i查看Inodes使用情況，審計目錄下有將近24W個文件，初步猜測是因為審計生成文件過多導致目錄所在分區的Inodes用光了，當時刪除部分審計TRACE文件后正常未查看 Inodes使用情況。
原因一：目錄權限問題--目錄權限被改，無權限向目錄寫審計數據
Oracle bin目錄（實際是整個oracle目錄）的屬主被更改: --ls -al bin
解決方法
chown -R oracle.oinstall /opt/orace
原因二：確實磁盤沒空間了 --df -h
原因三：磁盤Inodes用光--df -i
參考：模擬Linux磁盤分區有可用空間無可用Inodes時報錯：No space left on device
ORA-09925: Unable to create audit trail file
Linux Error: 13: Permission denied
Additional information: 9925
ORA-09925: Unable to create audit trail file
Linux Error: 13: Permission denied
Additional information: 9925
ORA-09925: Unable to create audit trail file
Linux-x86_64 Error: 30: Read-only file system
Additional information: 9925
ORA-01075: you are currently logged on
##############################
MOS上的一段描述及解決：
Problem Description -------------------
Problem occurs when: $ORACLE_HOME/rdbms/audit directory is full or $ORACLE_HOME/rdbms/audit directory doesn't exist as a side effect of this problem oracle asks for password after connect internal
Solution Description --------------------
Make space available in $ORACLE_HOME/rdbms/audit by removing files that are not needed or Make sure the directory exists and is readable by oracle or Change init.ora audit_file_dest to an existing directory. Change init.ora parameter means restart instance as workaround kill pmon.
Explanation -----------
The behavior on Unix is to write a file named ora_<process_id>.aud into the directory specified by audit_file_dest. The default directory is /rdbms/audit. The file will contain an entry for the audited action, which would be internal logins or startup. On ports which support an OS audit trail, these records will be written there. This feature is not affected by the settings of the audit_trail parameter.
ORA-09925: "Unable to create audit trail file"
Cause: ORACLE was not able to create the file being used to hold audit trail records.
Action: Check the UNIX error number for a possible operating system error.
If there is no error, contact ORACLE customer support.
References ----------
[NOTE:1018924.102] ORA-09925 ON DATABASE STARTUP [NOTE:1056988.6] ORA-09925 DURING HOT BACKUPS [NOTE:21073.1]
OERR: ORA-9925 "Unable to create audit trail [BUG:723955] SQLPLUS ALLOWS DB STARTUP (BUT NOT SHUTDOWN)
IF AUDIT_FILE_DEST DOESN'T EXIST
Search Words ------------ ORA-09925 audit trail audit_file_dest
免責聲明：本站發布的內容（圖片、視頻和文字）以原創、轉載和分享為主，文章觀點不代表本網站立場，如果涉及侵權請聯系站長郵箱：is@yisu.com進行舉報，并提供相關證據，一經查實，將立刻刪除涉嫌侵權內容。
