[Fwd: Re: 2 Gb file limit problem]

I would have fed the spooled output (flat file) into a named pipe & set its output thru split.
This is similar to what I do with exports for tables that are larger than 2GB.
-------- Original Message --------

Subject: 

Re: 2 Gb file limit problem

Date: 

Mon, 30 Jul 2001 12:00:32 -0800

From: 

"Satish Iyer" <Satish.Iyer_at_ci.seattle.wa.us>

Reply-To: 

ORACLE-L_at_fatcity.com

Organization: 

Fat City Network Services, San Diego, California

To: 

Multiple recipients of list ORACLE-L <ORACLE-L_at_fatcity.com>

 Thnaks Joe,Yes that was what
I did as a work-around yesterday and I had to be around for a long time on a week-end. I have most of these  processes automated and it works fine with 95% of the tables . Doing this for about 600 tables. So this would mean going back to change code again. Was wondering if there was any straightforward options that I was missing. Satish 
>>> "JOE TESTA" <JTESTA_at_longaberger.com> 07/30/01 11:44AM >>>how
about this: (avg_row_size
+ delimiters) * number_of_rows = total bytes. total bytes / 1900000000 = number of pieces. number_of_rows / number_of_pieces = number of rows per piece select number of rows needed multiple times, spooling each one individually. then sqlldr all the pieces. joe  
>>> Satish.Iyer_at_ci.seattle.wa.us

07/30/01 02:20PM >>>Hi List, I
need to transport few tables from one instance to another and of course found the sqlldr method much faster than the exp/imp.But the problems is for large tables .When I spool such input tables to a flat file , it stops spooling into it after about  2 Gb. Any possible solutions to get around it. I am on AIX 4.3.3/8.1.5 My ulimits on AIX  aretime(seconds)       
unlimited
file(blocks)        
unlimited
data(kbytes)        
unlimited
stack(kbytes)       
unlimited
memory(kbytes)      
unlimited
coredump(blocks)    
unlimited
nofiles(descriptors) 2000 Thanks Satish Received on Mon Jul 30 2001 - 15:43:57 CDT