ORA-00600: internal error code, arguments: [kfmrgReg01]

while creating a diskgroup on a 10.2.0.4 ASM we encountered

ORA-00600: internal error code, arguments: [kfmrgReg01], [24287664],

[84759367], [], [], [], [], []

tried dd,rebuilding volume group but eventually shutting down the asm removed this issue. asm gone crazy

ASM -renamedg KFNDG-00305: file not found

make sure you are issuing the renamedg from dba owned directory . as the renamedg create a file called rename_dg

and also make sure the diskgroup is dismounted

ORA-17628: Oracle error 19505 returned by remote Oracle server

precreate the contolfile directory in asm diskgroup then retry the duplicate from active database

rman duplicate from active database -PARAMETER_VALUE_CONVERT

incorrect/incomplete PARAMETER_VALUE_CONVERT can crash your target database . my target database crashed as i missed

it from rman duplicate while just trying to create the database with init.ora having only db_name entry

my controlfile/datafile/online log file in asm so i had to do as below

DUPLICATE TARGET DATABASE

TO t11gdact

FROM ACTIVE DATABASE

DB_FILE_NAME_CONVERT '+DATA/t11gdb/','+DATA/t11gdact/'

SPFILE

PARAMETER_VALUE_CONVERT '/u03/arch/t11gdb/','/u03/arch/t11gdact/','+DATA/t11gdb/','+DATA/t11gdact/'

SET SGA_MAX_SIZE '300M'

SET SGA_TARGET '250M'

SET LOG_FILE_NAME_CONVERT '+DATA/t11gdb/',

'+DATA/t11gdact/';

HPUX_SCHED_NOAGE

a must for oracle in hp-ux

ORA-29701: unable to connect to Cluster Manager on

checking the grep cssd I saw it was running as root user

+ASM:/home/oracle $ ps -ef | grep css

root 60799 20 0 10:14:39 ? 0:00 /bin/sh /sbin/init.d/init.cssd run

then asking the SA to do

localconfig delete

localconfig add

under OH fixed the issue by starting css the right way

oracle 35487 20 0 14:10:33 ? 0:00 /apps/oracle/product/10.2.0.3/bin/ocssd.bin

on 11gR2

mp012:+ASM:/apps/oracle/product/11.2.0/grid/dbs: crsctl start resource ora.cssd

CRS-2672: Attempting to start 'ora.cssd' on 'mp012'

CRS-2672: Attempting to start 'ora.diskmon' on 'mp012'

CRS-2676: Start of 'ora.diskmon' on 'mp012' succeeded

CRS-2676: Start of 'ora.cssd' on 'mp012' succeeded

mp012:+ASM:/apps/oracle/product/11.2.0/grid/dbs: crsctl modify resource "ora.cssd" -attr "AUTO_START=1"

mp012:+ASM:/apps/oracle/product/11.2.0/grid/dbs: crsctl modify resource "ora.diskmon" -attr "AUTO_START=1"

ORA-15096: lost disk write detected

one of the asm diskgroup after dropping it was still showing in v$asm_disk with error state. bringing down the asm didnt help either. so had to do dd to clean up the disk properly and then was able to recreate it.

ORA-15037:disk '/dev/asmvg02/rasmdisk02' is smaller than mimimum of 4 MBs

while creating a asm diskgroup I encountered this issue .

SQL> create diskgroup DFDG external redundancy disk

'/dev/asmvg02/rasmdisk02'; 2

create diskgroup DFDG external redundancy disk

*

ERROR at line 1:

ORA-15018: diskgroup cannot be created

ORA-15031: disk specification '/dev/asmvg02/rasmdisk02' matches no disks

ORA-15037: disk '/dev/asmvg02/rasmdisk02' is smaller than mimimum of 4 MBs

asking the SA he verified

vgdisplay -v /dev/asmvg02

diskinfo /dev/rdisk/disk11

and compared with the other working diskgroup disks.

but it didnt give any indication why i was hitting this error for this new diskgroup i am trying to create . So i gave a shot at dd and ifter that i was able to create diskgroup

+ASM:/dev/asmvg02: dd if=/dev/zero of=/dev/asmvg02/rasmdisk02 bs=4096 count=12800

12800+0 records in

12800+0 records out

+ASM:/dev/asmvg02: sqlplus '/as sysasm'

SQL> create diskgroup DFDG external redundancy disk

'/dev/asmvg02/rasmdisk02'; 2

Diskgroup created.

ORA-15063: ASM discovered an insufficient number of disks for diskgroup

We encountered this issue after the san copy is done.

ORA-15032: not all alterations performed

ORA-15063: ASM discovered an insufficient number of disks for diskgroup "DFDATA_2"

ORA-15063: ASM discovered an insufficient number of disks for diskgroup "DFDATA_1"

Looking into asm_diskstring dont seem to be any issue . however seems like the device mapping file was changed so the /dev/asmvg01 directory was not showing the proper asm disk group name .

once the SA fixed the device mapping file it recognized the asm disks and the system back online

kill -9 doesn't work always

well it didnt for me on a RHEL 5.3 for a process accessing NFS filesystem.

check - fuser -k

ORA-12853 / ORA-4031

ORA-12801: error signaled in parallel query server P002, instance xt0p-2:stpa2 (

2) ORA-12853: insufficient memory for PX buffers: current 68544K, max needed 15360K

ORA-04031: unable to allocate 131120 bytes of shared memory (""shared pool"",""unknown object"",""sga heap(2,0)"",""PX msg pool"")

3 node rac . parallel query process was spawning to node -2 . increasing the shared pool size fixed the query . and also changed the parallel instance groups.

ORA-14266: data type or length of an index subpartitioning column may

I was getting this error while trying to alter the data type for a column. as per the error . but as per the cause /action for this error there were no indexes/partitions on this column .

was not sure what is going on so i just dropped all the indexes( few normal index and partioned bitmap indexes) for this table and then when i try to do the alter it went fine .

move tablespace

select 'alter table mtsuser.'||table_name || ' move tablespace mtsuser_tab_data ;'

from dba_tables

where owner='MTSUSER'

select 'alter index mtsuser.'||index_name || ' rebuild tablespace mtsuser_IND_DATA ;'

from dba_indexes

where table_owner='MTSUSER'

set linesize 500

SELECT 'ALTER TABLE mtsuser.'||table_name ||' move lob('||column_name||') store as (tablespace mtsuser_tab_data );' FROM DBA_LOBS

WHERE owner='MTSUSER'

select 'alter table mtsuser.'||TABLE_NAME ||' move partition '||PARTITION_NAME||' tablespace mtsuser_data;' from dba_tab_partitions

where table_owner='MTSUSER'

select 'ALTER INDEX mtsuser.'||index_name ||' rebuild partition ' || partition_name ||' tablespace tnarchive_index ;'

from dba_ind_partitions

where index_owner='MTSUSER'

select 'alter table mtsuser.'||table_name ||' move partition '||partition_name||' tablespace mtsuser_data lob('||column_name ||') store as

(tablespace mtsuser_lob_data);'

from dba_lob_partitions

where table_owner='MTSUSER'

remote_listener -

It is recommended to set it up in RAC enviornment . but setting it up in regular standby databse gives rise to grid control alert ( observe the service update on both the network log files ) at times the grid control alert are misleading .

a good reference

http://oracledocaccess.blogspot.com/2009/04/remote-listener.html