Transportable Tablespace, SCNs and ORA-1555s.

  • From: Daniel Fink <Daniel.Fink@xxxxxxx>
  • To: oracle-l@xxxxxxxxxxxxx
  • Date: Wed, 17 Mar 2004 15:37:49 -0700

This is not in response to an actual problem, it is just the
result of a brain that goes off on tangents way to often...

How does a transaction know NOT to build a consistent version of
data that has been transported?

Sequence of events
1) Insert records into table1. A Commit SCN of 100000 is
assigned. Due to the size of the insert, there are blocks in the
table that still have an Uncommitted state.
2) Alter tablespace1 (containing table1) to read only and
transport it to database2.
3) Start a query in database2 (with an SCN of 1000) that
accesses the newly transported table1.

Will the query get a 1555?

Yes - It cannot create a read consistent view to SCN 1000
No - It knows that this data came from another database, so it
does not attempt to build a read-consistent version

Thoughts on which is correct? I know, I know...build a damn
testcase and see for myself. I'm working on it, but I wanted to
see what you all thought first.

Daniel


----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to:  oracle-l-request@xxxxxxxxxxxxx
put 'unsubscribe' in the subject line.
--
Archives are at //www.freelists.org/archives/oracle-l/
FAQ is at //www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------

Other related posts:

  1. Home
  2. oracle-l
  3. Archive
  4. 03-2004