| cold backup of controlfile raw device [message #218162] |
Wed, 07 February 2007 01:01  |
lokeshonline
Messages: 29
Registered: April 2006
|
Junior Member |
|
|
We have Oracle 9i on Aix 5 on node1 and node3.We have same /nfs filesystem mounted on node1 and node3. Datafiles and controlfiles are implemented as raw device on node1. DB_BLOCK_SIZE is 8k. I copied the raw device controlfile from node1 to /nfs. On node3 in the init.ora I changed the parameter controlfile to '/nfs/f_ctl1'. But when I try to mount it gives error "controfile version incompatible with oracle version". In alert log it shows following error
Corrupt block relative dba: 0x00000001 (file 0, block 1)
Fractured block found during control file header read
Data in bad block -
type: 0 format: 0 rdba: 0x00000000
last change scn: 0x0000.00000000 seq: 0x0 flg: 0x00
consistency value in tail: 0x00000000
check value in block header: 0x0, block checksum disabled
spare1: 0x0, spare2: 0x0, spare3: 0x0
What is the problem in copying the controlfile? If I take the hot backup of controlfile in /nfs using alter database command and give its path in node3, it mounts successfully. What can be the exact problem.??
[Updated on: Wed, 07 February 2007 01:07] Report message to a moderator |
|
|
|
|
|
| Re: cold backup of controlfile raw device [message #218374 is a reply to message #218256] |
Wed, 07 February 2007 22:45  |
lokeshonline
Messages: 29
Registered: April 2006
|
Junior Member |
|
|
I tired these steps.
A) Shutdown the database on node1
B) copied contorfile to /ora/backup on node1
using dd if=/dev/rctl1 of=/ora/backup/f_rctl1
C) mput the file f_rctl1 in binary mode on node3 in /ora/backup
D) In initMYDB.ora on node3 gave the path of controlfile '/ora/backup'
On mounting it gave the same ora-0201 "controlfile version incompatible with Oracle version"
In alert log
Corrupt block relative dba: 0x00000001 (file 0, block 1)
Fractured block found during control file header read
Data in bad block -
type: 0 format: 0 rdba: 0x00000000
last change scn: 0x0000.00000000 seq: 0x0 flg: 0x00
consistency value in tail: 0x00000000
check value in block header: 0x0, block checksum disabled
spare1: 0x0, spare2: 0x0, spare3: 0x0
|
|
|
|
|
|
| Re: cold backup of controlfile raw device [message #218416 is a reply to message #218381] |
Thu, 08 February 2007 02:26 |
lokeshonline
Messages: 29
Registered: April 2006
|
Junior Member |
|
|
Yes we have RAC on node1 and node2. I have shutdown properly both nodes. I tried one more step which helped me to mount.
1) I shutdown the DB on node1 and node2
2) copied the control file using dd
dd if=/dev/rctl1 of=/nfs/f_rctl1 bs=4k skip=1
(raw to filesystem)
3) As this /nfs was accessible on node 3 as well, on node3
I gave command
dd if=/nfs/f_rctl1 of=/dev/rctl1
(filesystem to raw)
4) On node3 I gave the path of controlfile to /dev/rctl1. It mounted successfully but if I gave the path to filesystem /nfs or /ora it gave same error.
What is the problem if I use the controlfile copied on filesystem?
|
|
|
|
] 
Blog
Search
Help
Members
Register
Login
Home
