Re: physical standby WAIT_FOR_GAP

Yeah, I'm on ASM myself. It's a bit of fiddle, have the standby at the nomount state and once you've copied the new standby controlfile you've created on the primary, it's a case of doing the restore controlfile to 'ASM location' from 'path to standby controlfile' type command. Obviously, you need to restore it to all your controlfile locations on the standby.

jason.

--
http://jarneil.wordpress.com



2009/12/22 Li Li <litanli_at_gmail.com>:


> Hi, Jason,

>

> We are using ASM on both primary and standby. so what I need to do is:

>

> 1. backup current controlfile for standby on primary side

> 2. copy the backup controlfile to standby side

> 3. restore the controlfile on standby side

>

> Is that what you did to resolve this issue?

>

> Thank you very much for your time!

> -Li

>

>

>

> On Tue, Dec 22, 2009 at 1:46 PM, jason arneil <jason.arneil_at_gmail.com> wrote:

>> Hello,

>>

>> I've seen this scenario occur quite a bit as well. I've seen creating

>> a new standby controlfile from the primary can straighten things out.

>>

>>

>> jason.

>>

>> --

>> http://jarneil.wordpress.com

>>

>>

>>

>> 2009/12/22 Li Li <litanli_at_gmail.com>:

>>> Greetings List,

>>>

>>> We have a physical standby running on a 4-node RAC. last night the

>>> apply instance (instance 1) had ora-04031 error and data guard broker

>>> changed the apply instance to be instance 2. Now apply instance

>>> (instance 2) shows "WAIT_FOR_GAP" for a log sequence 5xxx, which is a

>>> very old log, and has been applied a long time ago. Has any one seen

>>> this before? How do I trick the apply instance to skip the logs that

>>> have been applied already and start applying the logs that haven't

>>> been received but haven't applied.

>>>

>>> Oracle version 10.2.0.3 on RHEL AS 4.7

>>>

>>> Thanks and hope you all a merry christmas and happy new year!

>>> -Li

>>> --

>>> http://www.freelists.org/webpage/oracle-l

>>>

>>>

>>>

>>

>

--
http://www.freelists.org/webpage/oracle-l