Extents

IDS7.3 I need to expand the extents for most or all of our tables in our db. What would be the best way to go about this? The only obvious idea I have for now is to do a dbexport, edit the output to reflect larger extents, drop all tables and then do a dbimport? We are talking about over 100 tables and this would be a very tedious proccess. Any better ideas? Thanks! Steve

> I need to expand the extents for most or all of our tables in our db. What > would be the best way to go about this? > > The only obvious idea I have for now is to do a dbexport, edit the output to > reflect larger extents, drop all tables and then do a dbimport? Yes, definitely. Or you can use such a things like 'ALTER TABLE FRAGMENT INIT' for table or 'ALTER INDEX TO CLUSTER' for any index related to reorganised table. If you'll choose dbexport/dbimport, you don't need to drop all your tables one by one. Just drop your database and import it with dbimport. If you have big tables, use HPL for loading/unloading data into that tables. HTH. ------------------------------------------------- With best regards, Yuri Dovgart SAP R/3, Informix technical consultant, Informix Certified Professional, Senior System Consultant System Architecture and High Availability Systems, 'Telecominvest' company Email y_dovgart@tci.ukrtel.net ICQ 39284285

Steve Schroeder wrote: > > IDS7.3 > > I need to expand the extents for most or all of our tables in our db. What > would be the best way to go about this? > > The only obvious idea I have for now is to do a dbexport, edit the output to > reflect larger extents, drop all tables and then do a dbimport? > > We are talking about over 100 tables and this would be a very tedious > proccess. Any better ideas? Use the ALTER FRAGMENT ON TABLE tablename INIT IN dbspacename; to reorg your tables it is faster and does not change the tabid or require a full index rebuild, indexes are updated in-place. Just update the table's next extent size to a new value and reog. You can ease the load for large nubers of tables by making a perl or awk script, like those I posted in utils4_ak, to post process dbschema/myschema output. This works for non-fragmented tables and for fragmented tables just replace the IN dbspacename clause with the appropriate fragmentation scheme. Hmm this looks like a nice addition to myschema's capabilities! Most of the code is already in place. OK folk watch for the announcement. Art S. Kagel