Cold backups

I am running Oracle8 on Sun Solaris 2.51. My cold backup is giving me some problems. The process itself runs fine, but when the database is brought back up, I start getting core dumps for any kind of sql statement that is ran against the database. You can login to the database but you can not run any sql statements. If I shutdown the database and bring it back up...it's fine. I can not figure what causes this problem. I have checked the listener to make sure it is correct. The whole process is run through a cron job. Could this have anything to do with it? Roots owns the crontab entry and is the owner who runs it.

How is it shutdown? Make sure you shutdown immediate or shutdown normal before you make a cold backup.

It is being shutdown immediate.

Well I dont think it will create problem. Right now I am using two instances in the same server, started running simultaneously.

Only issue is, u must have sufficient RAM size, processor capacity, proper disk I/O load shed and proper Init parameter tuning. Otherwise the server will suffer performance issues.
---
What is the server capacity and its details, oracle version and the database sizes? Max. user sessions expected etc., etc.,

I am sorry, How come my posting goes to wrong thread ?(after reading the thread, by selecting the post reply option, i replied my solution, but it seems to be appearing here.

sorry guys, it is definitely some technical problem. dbasupport.com - please consider this matter.

------------

Any how for this thread, (cold backup issue) - first of all try restoring the cold backup in your test server. I doubt that whether the backup itself is working fine or not.

Then we will probe further.

I tested restoring with the cold backup files twice and it works fine. I am wonder it the problem is because root runs the cron job and the owner of the datafiles is oracledb, but that I don't think that should pose a problem.

If the backup seems to be okay means,

It is not the problem because of taking Cold backup or the cron jobs, it might be because of shutdown issues.

Please try a look at your alert log files and other logfile entries in listener, sqlnet etc., etc.,

I cant able to probe this peculiar problem. Never and ever heard such a thing.

Can you paste the log file entries.

Below is the sample of the shutdown information from the alert log...

Shutting down instance (immediate)
License high water mark = 7
Sun Apr 1 06:00:07 2001
ALTER DATABASE CLOSE NORMAL
Sun Apr 1 06:00:07 2001
SMON: disabling tx recovery
SMON: disabling cache recovery
Sun Apr 1 06:00:08 2001
Thread 1 closed at log sequence 2551
Sun Apr 1 06:00:08 2001
Completed: ALTER DATABASE CLOSE NORMAL
Sun Apr 1 06:00:08 2001
ALTER DATABASE DISMOUNT
Completed: ALTER DATABASE DISMOUNT
Sun Apr 1 08:15:39 2001
Starting ORACLE instance (normal)
LICENSE_MAX_SESSION = 0
LICENSE_SESSIONS_WARNING = 0
LICENSE_MAX_USERS = 0
Starting up ORACLE RDBMS Version: 8.0.5.0.0.
System parameters with non-default values:
processes = 100
sessions = 50
shared_pool_size = 6000000
control_files = /oracle/oradata/afegdb/control01.ctl, /database/oracle2/oradata/afegdb/control02.ctl, /database/oracle3/oradata/afegdb/control03.ctl
db_block_buffers = 200
log_archive_start = TRUE
log_archive_dest = /oracle/app/oracle/product/8.0.5/dbs/arch
log_buffer = 8192
log_checkpoint_interval = 10000
db_files = 80
db_file_multiblock_read_count= 8
dml_locks = 100
rollback_segments = r01, r02, r03, r04
sequence_cache_entries = 10
sequence_cache_hash_buckets= 10
global_names = TRUE
db_name = afegdb
ifile = /oracle/app/oracle/admin/afegdb/pfile/configafegdb.ora
background_dump_dest = /oracle/app/oracle/admin/afegdb/bdump
user_dump_dest = /oracle/app/oracle/admin/afegdb/udump
max_dump_file_size = 10240
core_dump_dest = /oracle/app/oracle/admin/afegdb/cdump
PMON started with pid=2
DBW0 started with pid=3
ARCH started with pid=4
LGWR started with pid=5
CKPT started with pid=6
SMON started with pid=7
RECO started with pid=8
Sun Apr 1 08:15:41 2001
alter database mount
Sun Apr 1 08:15:45 2001
Successful mount of redo thread 1, with mount id 3338510385.
Sun Apr 1 08:15:45 2001
Database mounted in Exclusive Mode.
Completed: alter database mount
Sun Apr 1 08:15:45 2001
alter database open
Sun Apr 1 08:15:46 2001
Thread 1 opened at log sequence 2551
Current log# 3 seq# 2551 mem# 0: /database/oracle3/oradata/afegdb/redoafegdb03.log
Current log# 3 seq# 2551 mem# 1: /oracle/oradata/afegdb/redoafegdb06.log
Successful open of redo thread 1.
Sun Apr 1 08:15:46 2001
SMON: enabling cache recovery
SMON: enabling tx recovery
Sun Apr 1 08:15:49 2001
Completed: alter database open
Sun Apr 1 12:02:30 2001
Errors in file /oracle/app/oracle/admin/afegdb/udump/afegdb_ora_10080.trc:
ORA-07445: exception encountered: core dump [kghfrf()+4] [SIGSEGV] [Address not mapped to object] [4294967284] [] []
Sun Apr 1 12:15:52 2001
Errors in file /oracle/app/oracle/admin/afegdb/udump/afegdb_ora_10150.trc:
ORA-07445: exception encountered: core dump [kghfrf()+4] [SIGSEGV] [Address not mapped to object] [4294967284] [] []
Sun Apr 1 13:26:34 2001
Errors in file /oracle/app/oracle/admin/afegdb/udump/afegdb_ora_10403.trc:
ORA-07445: exception encountered: core dump [kghfrf()+4] [SIGSEGV] [Address not mapped to object] [4294967284] [] []
Sun Apr 1 13:27:29 2001
Errors in file /oracle/app/oracle/admin/afegdb/udump/afegdb_ora_10405.trc:
ORA-07445: exception encountered: core dump [kghfrf()+4] [SIGSEGV] [Address not mapped to object] [4294967284] [] []
Sun Apr 1 13:38:00 2001
Errors in file /oracle/app/oracle/admin/afegdb/udump/afegdb_ora_10473.trc:
ORA-07445: exception encountered: core dump [kghfrf()+4] [SIGSEGV] [Address not mapped to object] [4294967284] [] []
Sun Apr 1 13:43:57 2001
Errors in file /oracle/app/oracle/admin/afegdb/udump/afegdb_ora_10479.trc:
ORA-07445: exception encountered: core dump [kghfrf()+4] [SIGSEGV] [Address not mapped to object] [4294967284] [] []
Sun Apr 1 13:45:23 2001
Errors in file /oracle/app/oracle/admin/afegdb/udump/afegdb_ora_10481.trc:
ORA-07445: exception encountered: core dump [kghfrf()+4] [SIGSEGV] [Address not mapped to object] [4294967284] [] []
Sun Apr 1 13:46:27 2001

Hi,

I have tried my best from Oracle Meta link,

I think it is the high time to call Oracle Support team. Please proceed so, if you wish, and dont forget to post the solution, once the bug fixed.

I will definitely do that. Thanks for the help and advice.

:D