OraFAQ Forum: Data Guard » RFS error after failover

Home » Server Options » Data Guard » RFS error after failover

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

RFS error after failover [message #246890]

Fri, 22 June 2007 06:49

mlgoins
Messages: 115
Registered: March 2007
Location: Denver, CO

Senior Member

I get the following RFS errors from the database after a FAILOVER. We're on DG 9.2, and the current database was a Physical Standby database prior to the failover. Here's the error set from the ALERT LOG (repeats every 6 minutes):

Fri Jun 22 06:31:28 2007
RFS: Possible network disconnect with primary database
RFS: Destination database mount ID mismatch [0x81245c3d:0x8124e956]
RFS: Not using real application clusters
Fri Jun 22 06:37:29 2007
RFS: Possible network disconnect with primary database
RFS: Destination database mount ID mismatch [0x81245c3d:0x8124e956]
RFS: Not using real application clusters

I'm wondering if it is because FAL_CLIENT and FAL_SERVER have values in them? Normally I wouldn't expect this from a database in PRIMARY role (select database_role from v$database).

Is there a way to shut it off?

Mike

Report message to a moderator

Re: RFS error after failover [message #246895 is a reply to message #246890]

Fri, 22 June 2007 06:56

mlgoins
Messages: 115
Registered: March 2007
Location: Denver, CO

Senior Member

Nevermind -- I figured it out (answered my own question). Cool

What's going on is that this was Failover Testing, and I brought the old Primary Database up after the test, so folks could access both databases and make comparisons. The old Primary Database still had log_archive_dest_state_2 set to "ENABLE", and was trying to send LGWR data to the "new" Primary Database. It tries that every 6 minutes, and complains that the destination database must be a Standby Database.

So, I just set log_archive_dest_state_2=DEFER on the original Primary Database, and the errors go away.

Mike

Report message to a moderator

Re: RFS error after failover [message #246947 is a reply to message #246895]

Fri, 22 June 2007 10:30

DreamzZ
Messages: 1666
Registered: May 2007
Location: Dreamzland

Senior Member

Quote:

The old Primary Database still had log_archive_dest_state_2 set to "ENABLE", and was trying to send LGWR data to the "new" Primary Database.

If i'm getting you right.sceneraio
your old PRIMARY is STANDBY and
your STANDBY is PRIMARY now.

when you use role transition

your Acutal PRIMARY BECOMES STANDBY and STANDBY goes to PRIMARY.

and when this happened then your old PRIMARY wouldnt generate archive log. Because your PRIMARY is NOW STANDBY.

So the Situation you defined is bit confusing.

Waiting for Response.

Report message to a moderator

Re: RFS error after failover [message #246982 is a reply to message #246890]

Fri, 22 June 2007 13:34

mlgoins
Messages: 115
Registered: March 2007
Location: Denver, CO

Senior Member

I'm in a testing situation. When I do a failover, the original Primary database is "gone". So the Standby becomes Primary. Once everyone's satisfied that they can get to the database, I turn the server back on and bring up the original Primary as well, so that they can check and see that all of their transactions made it to the new server. That's all,

Mike

Report message to a moderator

Previous Topic:	Rebuild Standby Database fails
Next Topic:	Looking for documentation on how quarterly patches are applied in a DG environ...

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

]

Current Time: Sat May 18 00:17:30 CDT 2013

Total time taken to generate the page: 0.14302 seconds