hi,
You would export to a device that does not support seeking such as a tape (not recommended, really slow) or a pipe.
Using compression AND split to make the export be in many managable sized file (500meg is my chosen size) will solve the problem.
Basically, you would create a pipe in the OS via:
$ mknod somefilename p
and then export to that pipe. you would set up another process in the background that 'eats' the contents of this pipe and puts it somewhere. I use split, you could use 'cat' to just put it into another file (if cat supports files >2 gig -- thats the problem here, most utilities do not, you need to use a special file io api to 2 gig file support).
Here is a script you can use as a template.
------------------------------
#!/bin/csh -vx
setenv UID /
setenv FN exp.`date +%j_%Y`.dmp
setenv PIPE /tmp/exp_tmp_ora8i.dmp
setenv MAXSIZE 500m
setenv EXPORT_WHAT "full=y COMPRESS=n"
echo $FN
cd /nfs/slx2/expbkup_ora8i
ls -l
rm expbkup.log export.test exp.*.dmp* $PIPE
mknod $PIPE p
date > expbkup.log
( gzip < $PIPE ) | split -b $MAXSIZE - $FN. &
# uncomment this to just SPLIT the file, not compress and split
#split -b $MAXSIZE $PIPE $FN. &
exp userid=$UID buffer=20000000 file=$PIPE $EXPORT_WHAT >>& expbkup.log
date >> expbkup.log
date > export.test
cat `echo $FN.* | sort` | gunzip > $PIPE &
# uncomment this to just SPLIT the file, not compress and split
#cat `echo $FN.* | sort` > $PIPE &
imp userid=sys/change_on_install file=$PIPE show=y full=y >>& export.test
date >> export.test
tail expbkup.log
tail export.test
ls -l
rm -f $PIPE
--------------------------------------------------
This also always does an 'integrity' check of the export right after it is done with an import show=y, that shows how to use these split files with import.
hope this helps!
Regards
Yogeeraj
No person was ever honoured for what he received. Honour has been the reward for what he gave (clavin coolidge)
