RE: Some Dataguard is good, lots more must be better? (specifically, when do most actual failovers really occur?)

From: "Laimutis Nedzinskas" <Laimutis.Nedzinskas@xxxxxxxxxxxxx>
To: <oracle-l@xxxxxxxxxxxxx>
Date: Wed, 20 Sep 2006 17:07:52 -0000

Well, here come a few things:

- separate: use DG for database and OS snapshots for OS stuff.
- keep it clean: eliminate all kind of scripts from db server machine
except of tightly controlled. 

I find it quite logical. High Availability must take the whole system
into account not only database alone. 
What if application server goes down, etc? However DG is quite good in
keeping the most complicated part - database - protected. The other
parts must be protected somehow as well.

Now regarding DG with time lag. Well, it is not a good option for
maximum data protection(as Oracle defines it.) At least I do not know
how to protect redo logs in this mode. Which means that for maximum data
protection you have to have a standby database which is in complete
sync, i.e. running in maximum protection mode(at least most of the
time.) BUT I am not sure if file system snapshot technique can achieve
this either.

Which means that you need yet another database with time lag. This
configuration can protect from accidental table truncates (drops can be
undone in 10g with recycle bin - provided it is not too buggy.) 10g has
a flashback option. I've never tested but it may be that you can
flashback a pshysical standby which means that same standby can serve
both purposes(again - most of the time but not 100% of time). However
for that I believe you need a flashback area with flashback logs - the
question is how much space they take per 1h for example?

If you go into
http://www.scaleabilities.co.uk/book/ScalingOracle8i-CreativeCommons.htm
then you can find a nice example of HA design at the end of the book (in
case study chapter) It has 4(!) databases if my memory serves me right.






-----Original Message-----
From: oracle-l-bounce@xxxxxxxxxxxxx
[mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of Kevin Closson
Sent: 20. september 2006 16:33
To: oracle-l@xxxxxxxxxxxxx
Subject: RE: Some Dataguard is good, lots more must be better?
(specifically, when do most actual failovers really occur?)

 >>>The OS mirror issue means that system admin must 
>>>unfortunetely be quite involved in the failover process or DBA must 
>>>take some tasks from sysadmin. I used to work in such an environment,

>>>for example, for me as DBA it was no issue to keep in synch and 
>>>activate all scripts, jobs, etc at the standby machine.
>>>

herein lies the main concern I have.  DG is great for keeping a database
replicated, yes, but upon switchover there needs to be the total
environment there...all the stuff the site has **outside** the database
for the workflow.
Such things as UTL_FILE stuff, external tables, scripts, etc... seems
storage level would be the only way to have the operations environment
replicated.

--
//www.freelists.org/webpage/oracle-l


Fyrirvari/Disclaimer
http://www.landsbanki.is/disclaimer
--
//www.freelists.org/webpage/oracle-l

Follow-Ups:
- RE: Some Dataguard is good, lots more must be better? (specifically, when do most actual failovers really occur?)
  - From: Kevin Closson
- RE: Some Dataguard is good, lots more must be better? (specifically, when do most actual failovers really occur?)
  - From: Carel-Jan Engel

RE: Some Dataguard is good, lots more must be better? (specifically, when do most actual failovers really occur?)

Other related posts: