Re: weird Dataguard

In article <f50ea93c-b873-45f3-a42b-93f2ddefe328 @r35g2000prm.googlegroups.com>, joel-garry_at_home.com says...
> On Aug 21, 10:43 am, <poky..._at_yahoo.fr> wrote:
> > Hi,
> >
> > Environment: Oracle 10.2.0.3 on windows server 2003 with MAA dataguard.
> >
> > Problem: Only REDO LOG FILE on standby database are written.
> >
> > It seems that LNS process work fine on primary, but LGWR process does
> > not write anything on primary REDO LOG File.
> >
> > Please, could you tell me where can I investigate ?
> >
> > Thanks
>
> Please tell us how you are determining the primary redo is not being
> written. That would indeed be weird. "a transaction will not commit
> until the redo needed to recover that transaction is written to the
> local online redo log and to at least one remote standby redo log."
>
> I'm guessing you have a confusion between LNS and LGWR?
> http://download.oracle.com/docs/cd/B19306_01/server.102/b25159/configbp.htm#CHDJCBEF
>
> Also see bug Bug 5746174 if your db is hanging.
>
> jg
> --
> @home.com is bogus.
> Masters of Spin: http://www.signonsandiego.com/uniontrib/20080820/news_1b20sdge.html
>

With the windows explorer I can see "Date modified" has not changed since months on primary redo logs.
Redo log on standby has a modified date from today.

With strings utility (under cygwin) I can see Seq# 0000000000 inside the primary redo log.

Archived redo logs are a modified date from today on both machine !

I don't understand how can I see why primary redo log are NOT updated ? Received on Fri Aug 22 2008 - 13:37:26 CDT