Managed Standby Database

Has anyone been able to get a managed standby database working using Oracle 8i? Here's what I've done so far:

1. Created an instance on an identical server to the production server.
2. Created an entry in tnsnames.ora on production server to reference standby server.
3. Backed up all datafiles on production server (I've tried both cold and hot).
4. Created a standby control file on the production server and copied it to the standby server.
5. Configured init_standby.ora on the standby server as follows:

    control_files = ("/..pathtocontrolfile../standby_control_01.ctl")     log_archive_start = true
    log_archive_dest = /..pathtoarchive../arch1     standby_archive_dest = /..pathtoarchive../arch1     log_archive_format = osid_%t_%s.dbf

6. Copied all backup files to standby machine 7. Switched log files on production server. Copied all log files to

    location defined in init_standby.ora 8. Started up the standby machine nomount with pfile=init_standby.ora 9. Mounted the standby database as follows:

    ALTER DATABASE MOUNT STANDBY DATABASE;

1. Attempted to recover the standby database in managed mode:

    RECOVER MANAGED STANDBY DATABASE     I've tried this step with and without the timeout option.

At this point the process just hangs and I get the following in the alert log:

    ...
    Media Recovery Log /..pathtoarchive../osid_1_2930.dbf     Media Recovery Log /..pathtoarchive../osid_1_2931.dbf     Media Recovery Waiting for thread 1 seq# 2932

The archive log file it's looking for exists on the standby machine in the archive log destination; so I don't understand why it's waiting for it.

On the other side (production server), I have the init.ora set up like this (as far as archive log stuff goes):

   log_archive_start = true
   log_archive_dest_1 = 'location=/..pathtoarchive../arch1'
   log_archive_dest_2 = 'location=/..pathtoarchive../arch2'
   log_archive_dest_3 = 'service=standby1 reopen=60'
   log_archive_dest_state_3 = enable
   log_archive_format = osid_%t_%s.dbf

It complains that it can't write to log_archive_dest_3:

   ARCH: Failed to archive log# 1 seq# 2941    Mon May 17 12:04:21 1999

   ARC0: Standby archivelog instance not available at host 'standby1'
   ARC0: Error creating archivelog file 'standby1'
   ARC0: Completed archiving log# 1 seq# 2941

I can understand this, since standby1 probably isn't ready to receive files, as it's waiting to recover itself. So does anyone have any idea why the standby database is pending recovery based on an existing archived redo log file?

By the way; I can cancel the above recovery and manually apply the redo log. This does not get me into the desired managed state, however.

Thanks much,

Tom Barry
TwoCoast Consulting

--== Sent via Deja.com http://www.deja.com/ ==-- ---Share what you know. Learn what you don't.--- Received on Mon May 17 1999 - 14:56:12 CDT