URGENT: Hot Standby Graceful failover and failback

2001-04-17 Thread Leng Kaing

Good morning everyone,

HELP! HELP!!!

It's very early in the morning and I'm still struggling with the 
Oracle Hot Standby - 8.0.5 and 8I. ARGH! 

Ok, I got the failover going. I can activate the standby database and
use it as a normal database. But when I try to turn it into a standby
database again and then reactivate it again, it errors out on me.
Oracle Support just tells me that I'm not supposed to use it this way.
HUH? I'm not trying to put it back in standby mode so I can recover
from the original primary database! I want to put it back in standby
mode to similulate a total primary site loss. And always have a standby
database to activate.

Ok, I may be rambling on a bit here. Very tired.

So here's what I've done so far:


1. create a hot standby from a primary database
2. activate the hot standby
3. shutdown the new primary and made a backup of the new primary
database
4. restart the new primary, created a table and create a standby
controlfile
5. shutdown the new primary
6. restore from backup created in step 3, and replaced the current
control files with the standby controlfiles created from stop 4
7. mounted the database in standby mode again
8. tried to activate the new standby database but it failed with
ORA-1152 and ORA-1110 complaining that the system.dbf is not restored
from a sufficiently old backup.

Argh! How do we have a continuous failover and failback scenario? Any
help would be greatly appreciated.

I've just discovered Lawrence To's article on Graceful Switch Over and
Switch back as well. The following from page 11 scares me:

"Graceful switchover and switchback are not possible when the
production database's online redo logs are not accessible.

A graceful switchover and switch back is NOT possible whenever a
production database or standby database executes one of the following:

- alter database open resetlogs or
- alter database activate standby database (which does an implicit
resetlogs operation)"


HUH??? There are certainly times when we have to do a resetlogs. One of
the problems our Unix boxes have at the moment is a redo log corruption
so there are not choices but to resetlogs. So what does this all mean?
That if we resetlogs we can't ever have continuous failover and
failback? 

ARGH!! What am I dealing with here!

I need a holiday from this crazy scenario!

Ok, please let me know if you've had any success with what I'm
describing here. Am I aiming for the impossible? Surely not! 


Thanks,

Leng.



=
Leng Kaing - [EMAIL PROTECTED]
AUSOUG-VIC : http://www.ausoug.org/vic/

_
http://movies.yahoo.com.au - Yahoo! Movies
- Now showing: Dude Where's My Car, The Wedding Planner, Traffic..
-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: =?iso-8859-1?q?Leng=20Kaing?=
  INET: [EMAIL PROTECTED]

Fat City Network Services-- (858) 538-5051  FAX: (858) 538-5051
San Diego, California-- Public Internet access / Mailing Lists

To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).



RE: URGENT: Hot Standby Graceful failover and failback

2001-04-17 Thread Lanteigne, Mike

I think you may be trying to do a little too much with the current standby
architecture. Oracle now has  product called Data Guard that will let you do
what you want. I've only read the docs on Data Guard, but supposedly you can
switch primary and standby database, and switch back without re-creating
anything.

Good luck

Mike

The views expressed here are mine and do not reflect the official position
of my employer or the organization through which the internet was accessed.


 -Original Message-
 From: Leng Kaing [SMTP:[EMAIL PROTECTED]]
 Sent: Tuesday, April 17, 2001 12:10 PM
 To:   Multiple recipients of list ORACLE-L
 Subject:  URGENT: Hot Standby Graceful failover and failback
 
 Good morning everyone,
 
 HELP! HELP!!!
 
 It's very early in the morning and I'm still struggling with the 
 Oracle Hot Standby - 8.0.5 and 8I. ARGH! 
 
 Ok, I got the failover going. I can activate the standby database and
 use it as a normal database. But when I try to turn it into a standby
 database again and then reactivate it again, it errors out on me.
 Oracle Support just tells me that I'm not supposed to use it this way.
 HUH? I'm not trying to put it back in standby mode so I can recover
 from the original primary database! I want to put it back in standby
 mode to similulate a total primary site loss. And always have a standby
 database to activate.
 
 Ok, I may be rambling on a bit here. Very tired.
 
 So here's what I've done so far:
 
 
 1. create a hot standby from a primary database
 2. activate the hot standby
 3. shutdown the new primary and made a backup of the new primary
 database
 4. restart the new primary, created a table and create a standby
 controlfile
 5. shutdown the new primary
 6. restore from backup created in step 3, and replaced the current
 control files with the standby controlfiles created from stop 4
 7. mounted the database in standby mode again
 8. tried to activate the new standby database but it failed with
 ORA-1152 and ORA-1110 complaining that the system.dbf is not restored
 from a sufficiently old backup.
 
 Argh! How do we have a continuous failover and failback scenario? Any
 help would be greatly appreciated.
 
 I've just discovered Lawrence To's article on Graceful Switch Over and
 Switch back as well. The following from page 11 scares me:
 
 "Graceful switchover and switchback are not possible when the
 production database's online redo logs are not accessible.
 
 A graceful switchover and switch back is NOT possible whenever a
 production database or standby database executes one of the following:
 
 - alter database open resetlogs or
 - alter database activate standby database (which does an implicit
 resetlogs operation)"
 
 
 HUH??? There are certainly times when we have to do a resetlogs. One of
 the problems our Unix boxes have at the moment is a redo log corruption
 so there are not choices but to resetlogs. So what does this all mean?
 That if we resetlogs we can't ever have continuous failover and
 failback? 
 
 ARGH!! What am I dealing with here!
 
 I need a holiday from this crazy scenario!
 
 Ok, please let me know if you've had any success with what I'm
 describing here. Am I aiming for the impossible? Surely not! 
 
 
 Thanks,
 
 Leng.
 
 
 
 =
 Leng Kaing - [EMAIL PROTECTED]
 AUSOUG-VIC : http://www.ausoug.org/vic/
 
 __
 ___
 http://movies.yahoo.com.au - Yahoo! Movies
 - Now showing: Dude Where's My Car, The Wedding Planner, Traffic..
 -- 
 Please see the official ORACLE-L FAQ: http://www.orafaq.com
 -- 
 Author: =?iso-8859-1?q?Leng=20Kaing?=
   INET: [EMAIL PROTECTED]
 
 Fat City Network Services-- (858) 538-5051  FAX: (858) 538-5051
 San Diego, California-- Public Internet access / Mailing Lists
 
 To REMOVE yourself from this mailing list, send an E-Mail message
 to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
 the message BODY, include a line containing: UNSUB ORACLE-L
 (or the name of mailing list you want to be removed from).  You may
 also send the HELP command for other information (like subscribing).
-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Lanteigne, Mike
  INET: [EMAIL PROTECTED]

Fat City Network Services-- (858) 538-5051  FAX: (858) 538-5051
San Diego, California-- Public Internet access / Mailing Lists

To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).



Re: URGENT: Hot Standby Graceful failover and failback

2001-04-17 Thread Michael Netrusov

Hello Leng,

 1. create a hot standby from a primary database
 2. activate the hot standby
 3. shutdown the new primary and made a backup of the new primary
 database
 4. restart the new primary, created a table and create a standby
 controlfile
 5. shutdown the new primary
 6. restore from backup created in step 3, and replaced the current
 control files with the standby controlfiles created from stop 4
 7. mounted the database in standby mode again
 8. tried to activate the new standby database but it failed with
 ORA-1152 and ORA-1110 complaining that the system.dbf is not restored
 from a sufficiently old backup.

I thought you were supposed to do a hot backup of a new primary and restore it to the 
old primary. After that, mount the old primary
as a new standby.

 Argh! How do we have a continuous failover and failback scenario? Any
 help would be greatly appreciated.

 I've just discovered Lawrence To's article on Graceful Switch Over and
 Switch back as well. The following from page 11 scares me:

 "Graceful switchover and switchback are not possible when the
 production database's online redo logs are not accessible.

 A graceful switchover and switch back is NOT possible whenever a
 production database or standby database executes one of the following:

 - alter database open resetlogs or
 - alter database activate standby database (which does an implicit
 resetlogs operation)"

 HUH??? There are certainly times when we have to do a resetlogs. One of
 the problems our Unix boxes have at the moment is a redo log corruption
 so there are not choices but to resetlogs. So what does this all mean?
 That if we resetlogs we can't ever have continuous failover and
 failback?

Yes, it is correct. If you reset logs or loose the redo logs, forget about "graceful 
switchover".

 ARGH!! What am I dealing with here!

 I need a holiday from this crazy scenario!

 Ok, please let me know if you've had any success with what I'm
 describing here. Am I aiming for the impossible? Surely not!


I am sorry to admit, but you are :-)
So far at 8.1.7 we have only the "graceful switchover", awfully well described in 
Lawrence To's article.

HTH,
Michael

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Michael Netrusov
  INET: [EMAIL PROTECTED]

Fat City Network Services-- (858) 538-5051  FAX: (858) 538-5051
San Diego, California-- Public Internet access / Mailing Lists

To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).