RE: Stand-by (Oracle9i Data Guard) vs. Replication

From: Freeman, Robert <Robert_Freeman_at_csx.com>
Date: Mon, 25 Mar 2002 11:48:04 -0800
Message-ID: <F001.004326C3.20020325114804@fatcity.com>

Yea yea yea.... well we don't have quite that level of definition at this point, and I have yet to see absolute numbers come my way. Isn't my project, in fact this is an outside vendor and I'm coming in on the tail end of the projects design phase and being asked, do you agree with this design and I'm saying, now's a fine time to ask.

I've asked for additional facts and figures, but pointing my little digits sometimes only get's them bitten off by the big machine that is.

The requirements as defined thus far to me are:

1. No more than 5 minutes outage.
2. No data loss
3. each site has to be able to act autonomously.

three different kinds of data....

1. Network wide data (used by all sites) - critical, no data loss
2. Regional data (specific to a region, not used by other sites) - critical, no data loss
3. Easily recoverable non-critical data. Data loss is tolerable.

That being said, they want to replicate between the four sites, and they want to use replication conflict resolution rules to deal with outages and resync of the databases. They claim it will work great, I'm thinking, yea right sounds like a bear to manage. I'd much rather do 9i data guard just off the hip, but as you point out, I do not have lots of info as yet, so we shall see what comes down the pike.

Robert G. Freeman - Oracle8i OCP
Oracle DBA Technical Lead
CSX Midtier Database Administration

The Cigarette Smoking Man: Anyone who can appease a man's conscience can take his freedom away from him.

-----Original Message-----
Sent: Monday, March 25, 2002 1:23 PM
To: Multiple recipients of list ORACLE-L

Sorry for interrupting...
but our SLAs (requirements docs) do not have such loose language... Things like 'ASAP' and 'as little latency as possible' must be specified in absolute numbers (minutes, hours, days etc.) If it is not spelled out, it's a wide open game of mud slinging.... ;)

• Kirti

-----Original Message-----
Sent: Monday, March 25, 2002 11:49 AM
To: Multiple recipients of list ORACLE-L

Pretty stringent. They want as little latency as possible. Changes at a master should be available to all sites ASAP. Now, they could all go to one central site, and thats ok as long as our networking is healthy, but if it goes down, there is a requirement that they be able to work independently (there are 4-5 sites) and then all changes need to be synchronized. Data loss is secondary to availability however.

These requirements smack of trouble to me.

Robert G. Freeman - Oracle8i OCP
Oracle DBA Technical Lead
CSX Midtier Database Administration

The Cigarette Smoking Man: Anyone who can appease a man's conscience can take his freedom away from him.

-----Original Message-----
Sent: Monday, March 25, 2002 11:48 AM
To: Multiple recipients of list ORACLE-L

What type of requirement or SLA do you have in regards to keeping the instances in sync?

-Joe

• "Freeman, Robert " <Robert_Freeman_at_csx.com> wrote:
  > Stand-by (Oracle9i Data Guard) vs. Replication
  >
  > Folks,
  >
  > I have a mission critical system we are architectonic right now.
  > There is some argument of the merits of replication vs. using
  > Standby database going on.
  >
  > Current we have 4 sites that we will be replicating data back and
  > forth between. There are 3 kinds of data:
  >
  > 1. Network Critical data (must be available for entire network)
  > 2. Regional Critical data (only used for a given region. site =
  > region).
  > 3. Regional non-critical data (this is data that is easily
  > recovered from
  > other operational data stores).
  >
  > I can load you up with details, but for now this is the general
  > requirement. We want a given site to be able to work independently
  > of the other sites in the event of network failure (WAN).
  >
  > What I'm looking for is your experience with using replication
  > for HA solutions vs. stand-by databases. I've also considered using
  > standby databases as a possible solution to this problem, along
  > with
  > using transportable tablespaces to re-sync the databases once
  > everything
  > comes up. I'm concerned with replication in that there is allot to
  > break,
  > and I'm concerned about synchronization issues in general with
  > either
  > solution.
  >
  > Thoughts?
  >
  > RF
  >
  >
  -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Deshpande, Kirti INET: kirti.deshpande_at_verizon.com

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------

To REMOVE yourself from this mailing list, send an E-Mail message to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Freeman, Robert 
  INET: Robert_Freeman_at_csx.com

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).

Received on Mon Mar 25 2002 - 13:48:04 CST