How to remove old oracle home ? - Get tar file of old oracle home. /<$BACKUP_LOCATION>/db_home.tar.gz $ORACLE_HOME - Go to $ORACLE_HOME/oui/bin location - Check oratab file before remove file. - Execute below in silent mode to detach home. -bash-4.4$ ./runInstaller -silent -waitforcompletion -detachHome ORACLE_HOME='/u01/app/oracle/product/12.2.0/dbhome_1' -local - After executed above check the oratab file. Output of above command: Starting Oracle Universal Installer... Checking swap space: must be greater than 500 MB. Actual 38817 MB Passed The inventory pointer is located at /var/opt/oracle/oraInst.loc 'DetachHome' was ...
Posts
ASM DISK MIGRATION [OCR/DATA/RECO DISK GROUPS] in RAC
Current Disk group as below. You must prepare candidate disks with proper permissions by storage level. Here we are going to add below new disks for each disk group with rebalance power 6 and we can do afd label below disks and assign to ASM disk group using asmca or sqlplus commands. After added new disks to relevant disk groups you may drop the old disks in DATA and RECO disk groups with rebalance power using asmca or sqlplus. But note here, we do not drop old disk of OCR disk group due to existing voting disks. Therefore, you must need to transfer voting disks in to another ASM disk group which has sufficient space. Before moving voting disk check OCR devices and existing voting disk location. Note you need to have OCR mirroring device before move voting disk to another ASM disk group and get a manual backup of OCR. To check OCR device = > ocrcheck To check voting disks => crsctl query css votedisk To check OCR backup=> ocrconfig -showbackup <AUTO|MAN...
ORA-00609: could not attach to incoming connection in database alert log
Environment Details ---------------------------- Oracle Real Application Cluster 02 Nodes Oracle Solaris 11.3 SPARC Oracle version 19.6.1.0.0 Problem Description --------------------------- Above error continuously reported in the database alert. In that case we have to check certain logs instead to database alert. Before that, we must find for any firewall rule changes or any IP filter changes in between application server IP and database IPs with the internal network team. So the error reported in the db alert as follows. 2021-11-23T15:20:09.766518+05:30 opiodr aborting process unknown ospid (9758) as a result of ORA-609 2021-11-23T15:20:09.770202+05:30 Errors in file /u01/app/oracle/diag/rdbms/db/db1/trace/db1_ora_9758.trc : ORA-00609: could not attach to incoming connection ORA-12547: TNS:lost contact Considering the alert log error we must check listener logs as well. This can be done finding and comparing the similar timestamp in the listener logs. 23-NOV-2021 15:...
ORA-01555 caused by SQL statement below (SQL ID: 6pfwn5k98dywk, Query Duration=0 sec, SCN: 0x000000070377b986)
ORA-01555 caused by SQL statement below (SQL ID: 6pfwn5k98dywk, Query Duration=0 sec, SCN: 0x000000070377b986) in the DB alert log Environment Details ------------------------- Oracle Real Application Cluster Multitenant Oracle Solaris 11.3 SPARC Oracle version 19.12.0.0.0 Problem Description -------------------------- Above error was continuously occurring daily in a specific time period. Basically when there is an error like above in the db alert we must check if any task has been performed during the incident time period. ORA-01555 is an error which occurs due to insufficient rollback segments or UNDO_RETENTION parameter value [Default Value is 900]. Therefore increasing this value as you want will not solve the issue. You must find why does this error occur?. In this scenario, expdp was executing during the incident time using the parameter value FLASHBACK_SCN which means CONSISTENT=Y. So the flashback method depends on the UNDO data. Therefore the UNDO must retain...