Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Usenet -> c.d.o.server -> Re: data replication tool
On Thu, 10 Jul 2003 20:41:03 +0200, "Christoph Seidel"
<christoph.seidel1_at_gmx.de> wrote:
>Sybrand Bakker wrote:
>> On Thu, 10 Jul 2003 19:58:52 +0200, "Christoph Seidel"
>> <christoph.seidel1_at_gmx.de> wrote:
>>
>>> Well after reading a bit and playing a bit with it: it seems that it
>>> can only copy complete tables. What I need is:
>>>
>>> The user selects record 1, 30 and 100 from a table and only these
>>> records and the data they depend on are copied.
>>>
>>
>>
>> Not true.
>> Read again
>
>Argh... please gimme some more hints, the docs are really huge.
>
Let's discuss a simple snapshot first
By a simple snapshot I mean select * from table, no joins, no group
by's no order by nothing.
Whem you have a snapshot only, you can only issue a *complete*
refresh.
Once you set up a snapshot *log* at the master site, Oracle will track
the changes on the table for you.
You now can issue a *fast* refresh which will only read the snapshot
log and update the affected rows at the slave.
As far as I know this will still work when you create your snapshot
with a simple where clause, generating a straight subset of your data.
Of course you'll need to have *formal* criteria (and the example you
provided seemed to pick random records)
With respect to the *complex* snapshot (with group by's, order by's,
joins and the like) you can't setup a snapshot log. Hence you will be
limited to complete refresh for complex snapshots
Hope this helps
Sybrand Bakker, Senior Oracle DBA
To reply remove -verwijderdit from my e-mail address Received on Thu Jul 10 2003 - 15:38:30 CDT
![]() |
![]() |