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: Import 8.1.6 to 9.2.0.1 weirdness

RE: Import 8.1.6 to 9.2.0.1 weirdness

From: Karen Morton <oracledba_at_morton-consulting.com>
Date: Fri, 16 May 2003 06:51:50 -0800
Message-ID: <F001.0059AA13.20030516065150@fatcity.com>


I'll try the larger buffer and can definitely see how that would make a difference particularly with tables that don't use LOBs. The Oracle doc says that for tables containing LOB type columns, rows are inserted individually. For other tables, the size of the buffer must be large enough to contain the entire row (but it specifically says "except for LOB and LONG columns"). It then says that if the buffer cannot hold the longest row in a table, Import attempts to allocate a larger buffer.

So I'm wondering, how does it allocate the larger buffer (i.e. how does it determine the appropriate size and is there a way to see that) and how/why should it make a difference on tables using LOB datatypes if it will only do inserts 1 row at a time for those tables anyway?

Also, to answer a few other questions asked:
* there are no errors in the alert log

Thanks for all the responses,

Karen

-----Original Message-----
Sent: Thursday, May 15, 2003 7:37 PM
To: ORACLE-L_at_fatcity.com
Cc: oracledba_at_morton-consulting.com
Importance: High

I don't know if this may have any bearing on the problem you are experiencing,
but a 64k buffer is rather small. I like to use 10 meg on imports eg. buffer=10485760

When it is hanging, have you logged in to the target database and checked to
see what the import session is waiting on via v$session_wait?

Jared

"Karen Morton" <oracledba_at_morton-consulting.com> Sent by: root_at_fatcity.com
 05/15/2003 04:47 PM
 Please respond to ORACLE-L

        To:     Multiple recipients of list ORACLE-L <ORACLE-L_at_fatcity.com>
        cc:
        Subject:        Import 8.1.6 to 9.2.0.1 weirdness


Would anyone have any idea what might be going on with the following import
situation?

I'm importing a 8.1.6 database into a 9.2 database. There are almost 400 tables and the import seems to chug along fine until I get to a table with a
CLOB column in it. At that point, the import loads in anywhere between a few hundred and a few thousand rows of that table and then "hangs". I end up having to kill the import session and start over. Actually what has mostly happened is that I can't even simply restart but end up having to shutdown the database and restart it before I can continue (don't worry it's
a new test box so all the ups and downs aren't an issue). Since I'm using commit=y and the tables involved all have unique keys, I simply have been restarting the import and letting it take off from there.

Here's my imp command:
imp id/pswd file=exp816.dmp commit=y ignore=y analyze=n buffer=16384 fromuser=user1 touser=user2

I've tried changing the buffer setting from 4K to 64K with no apparent effect. I've also tried various other parameter combinations other than those listed with no help. The command you see is what seems to work best.

The real weird thing is that if I just keep restarting the database and then
restarting the import, the table will eventually load. For one table that had 30000 rows, I had to start the import and let it run until it hung, then
shutdown the database and restart the import 10 times to get the full table
to load. Once it loads, the import continues on just fine until it hits another table that has a CLOB datatype. Actually it goes one step further than just being a table with a CLOB, it's a table with a CLOB that's over 2K
or so. There are a couple of tables with CLOB columns but the max size of the data stored in that column is relatively small. The tables that seem to
have troubles have max size in the CLOB column of over 2K.

I just need to get this data loaded for test purposes so I can put up with the process for the moment, but I can't for the life of me figure out why it's acting this way. I've never seen behavior like this.

Does anybody have any ideas/suggestions or think of anything I may be missing somewhere that's causing this problem?

Many thanks,
Karen



Karen Morton
karen_at_morton-consulting.com
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: Karen Morton
  INET: oracledba_at_morton-consulting.com

Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
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.net
-- 
Author: Karen Morton
  INET: oracledba_at_morton-consulting.com

Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
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 Fri May 16 2003 - 09:51:50 CDT

Original text of this message

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