Paul J Stevens wrote: > N Sj wrote: > >> In our dbmail environment dbmail_headervalue >> table has become very large and DBA is recommending running optimize table. >> Then I ran into a thread in dbmail archives that if the table >> dbmail_headervalue is dropped and recreated then it will get re-populated by >> running "dbmail-util -by", if that is possible then should we run OPTIMIZE >> TABLE or just drop and recreate the table. >> > > Dropping, re-creating and re-populating the table is a valid procedure. > However, you will end up with the same sized table. > > I've blogged about a better approach on blog.dbmail.eu. > > > I admit to not having read the post on blog.dbmail.eu yet, but the idea he's aiming for is something faster than OPTIMIZE TABLE dbmail_headervalue;
For which the obvious intent is to use less disc-space. I'm not sure this will be relevant, but it is his desire. To shrink the size of the database file itself. And yes, I know that if you delete from the table, the leftover spots in the file will get re-used. He may also want to run ANALYZE TABLE, if he hasn't already. Running both what you recommend (removing the unnecessary cached fields) and the OPTIMIZE will produce even _better_ results than either by itself. >> Also we run "dbmail-util- a -y -vv" every night. >> >> Thank you for your input. >> >> >> >> >> _______________________________________________ >> DBmail mailing list >> [email protected] >> http://mailman.fastxs.nl/cgi-bin/mailman/listinfo/dbmail >> >> > > >
signature.asc
Description: OpenPGP digital signature
_______________________________________________ DBmail mailing list [email protected] http://mailman.fastxs.nl/cgi-bin/mailman/listinfo/dbmail
