Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> Re: Is RAC really HA on Linux

Re: Is RAC really HA on Linux

From: Stephen Evans <evans036_at_mc.duke.edu>
Date: Mon, 13 Sep 2004 16:14:12 -0400
Message-ID: <OF41F047A9.80EFC81F-ON85256F0E.006EAABA@notes.duke.edu>


paul,
yes - archive log dest is mandatory. we only have one of them, but we are very carefull to monitor our alert logs for any potential invalidation of archive log dest. we also have a job that compares the currency of the standby databases with their production counterparts to ensure that all archived logs are successfully applied. if there is any significant lag, the help desk will get an alert and they will call database firewatch (day or night :-( )

is the consensus that RAC is too immature for primetime use? What about other non-linux unix platforms - aix, solaris maybe?

thanks,

steve

Paul Drake <bdbafh_at_gmail.com>
Sent by: oracle-l-bounce_at_freelists.org
09/13/2004 03:49 PM
Please respond to bdbafh  

        To:     oracle-l_at_freelists.org
        cc: 
        Subject:        Re: Is RAC really HA on Linux


Steve,

Those archivelog destinations are all mandatory, right? Non-mandatory destinations shouldn't really be counted on and may do more harm than good.

Paul

On Mon, 13 Sep 2004 15:38:47 -0400, Stephen Evans <evans036_at_mc.duke.edu> wrote:

> we are a hospital & somes apps are so mission critical that patient care
> is compromised for every minute of downtime we incur.

oh. that sounds like a serious requirement which would likely be able to carry a budget to support it.

> we are extensively using data guard right now. we have also achieved
zero
> data loss on the san (against a SINGLE point of failure) by using a
> combination of dataguard and triplexing our active redos on two stroage
> controllers (in different locations) and on local unix storage. We are
> almost certain that we will always be able to obtain at least one copy
of
> the active redo if any part of the system goes belly-up (hence zero
> dataloss once applied to our physical standby).
>

--
To unsubscribe - mailto:oracle-l-request_at_freelists.org&subject=unsubscribe 
To search the archives - http://www.freelists.org/archives/oracle-l/




--
To unsubscribe - mailto:oracle-l-request_at_freelists.org&subject=unsubscribe 
To search the archives - http://www.freelists.org/archives/oracle-l/
Received on Mon Sep 13 2004 - 15:09:47 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US