I agree. Tony, thanks, that was close enough to what I was looking for. I have to make some modifications to get_shell, but most of the rest is right on. If anyone is interested, I can post the finished product after we test it (meaning, get rid of my typos). On Dec 19, 2007 5:30 PM, Jared Still <jkstill@xxxxxxxxx> wrote: > That's the way it *should* be done. > > 1 queue, multiple servers. > > > On Dec 19, 2007 11:14 AM, Tony Aponte <Tony_Aponte@xxxxxxxxx> wrote: > > > > > > > > > > Try this to see if it fits your needs. I've used it to do pretty much > the > > same task you describe. The script accepts a degree of parallelism and > an > > input file containing commands to execute. > > > > > > > > http://sysadmintalk.com/showthread.php?threadid=762 > > > > > > > > > > > > Tony Aponte > > > > > > > > ________________________________ > > > > > > From: oracle-l-bounce@xxxxxxxxxxxxx [mailto: > oracle-l-bounce@xxxxxxxxxxxxx] > > On Behalf Of Charles Schultz > > Sent: Tuesday, December 18, 2007 4:44 PM > > To: Oracle-L > > Subject: Script to build indexes in parallel > > > > > > > > > > Anyone have a script to read a DDL file and build indexes using parallel > > worker processes? Something in perl or shell would be preferable. I am > > trying to work around data pump's slow "feature" of building indexes > with > > one query coordinator by extracting the metadata ddl and running them > with > > real parallelization. I was just getting started with a perl script > using > > fork but wanted to check here so as not to reinvent the wheel. > > > > -- > > Charles Schultz > > > > -- > Jared Still > Certifiable Oracle DBA and Part Time Perl Evangelist > -- Charles Schultz