RE: Real-time Apply in DataGuard
Date: Tue, 21 Sep 2010 21:16:08 +0200
Message-ID: <4814386347E41145AAE79139EAA39898102B654376_at_ws03-exch07.iconos.be>
Allen,
With lgwr async you still have the possibility to impact your primary datatase (although you need to have a really bad situation, before it will emerge):
With lgwr async your lns process is reading from the online redo logfile and sends the redo vectors to the standby database. If all is fine, the lns process is just behind the poin the point where the lgwr is writing. However, if your network latency to the standby db is to high for the amount of redo you are generating, the lns process can fall behind. In the case that the lns process is so far behind that it is still reading from logfile n, while the lgwr is already writing to logfile n+2, then an optimization kicks in and the lns process will skip the logfile n+1 (it will be transferred via the arch process). However, if the lns process is even further behind and the lgwr wants to start writing to the logfile from which the lns process is still reading, then the database will halt until the lns process has finished reading that logfile.
Regards,
Freek D'Hooge
Uptime
Oracle Database Administrator
email: freek.dhooge_at_uptime.be
tel +32(0)3 451 23 82
http://www.uptime.be
disclaimer: www.uptime.be/disclaimer
--
From: oracle-l-bounce_at_freelists.org [mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Allen, Brandon
Sent: dinsdag 21 september 2010 19:42
To: Andrew Kerber; oracle-l_at_freelists.org
Subject: Real-time Apply in DataGuard
I'm starting a new thread specific to real-time apply.
That is my fear - that by using real-time apply we might impact the performance or availability of the production database if there are network problems between the primary and standby. We had problems with that the last time I tried to implement Data Guard in 9.2.0.4, but according to the documentation it sounds like that shouldn't happen anymore in 10.2. For example, from page 22 of the doc linked below: "In Oracle Database 10g Release 2, the new LGWR ASYNC behavior eliminates the potential to stall the production database"
Has anyone done any testing to confirm if network problems can impact performance of the primary when using real-time apply in 10.2 (10.2.0.4.3 specifically in my case)? I'll be testing it myself soon, but am still curious to know what others have found.
Thanks,
Brandon
From: Andrew Kerber [mailto:andrew.kerber_at_gmail.com] I have never implemented real-time apply, but it is my understanding that there are performance issues when there are problems with the network
On Tue, Sep 21, 2010 at 11:16 AM, Allen, Brandon <Brandon.Allen_at_oneneck.com> wrote: Why not use real-time apply instead?
Privileged/Confidential Information may be contained in this message or attachments hereto. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. Opinions, conclusions and other information in this message that do not relate to the official business of this company shall be understood as neither given nor endorsed by it.
--
http://www.freelists.org/webpage/oracle-l Received on Tue Sep 21 2010 - 14:16:08 CDT