Re: LONG delete with LOTS of FK's - Mailing list pgsql-general

From Tom Lane
Subject Re: LONG delete with LOTS of FK's
Date
Msg-id 25119.1367507317@sss.pgh.pa.us
Whole thread Raw
In response to Re: LONG delete with LOTS of FK's  (Larry Rosenman <ler@lerctr.org>)
Responses Re: LONG delete with LOTS of FK's  (Larry Rosenman <ler@lerctr.org>)
Re: LONG delete with LOTS of FK's  (Larry Rosenman <ler@lerctr.org>)
List pgsql-general
Larry Rosenman <ler@lerctr.org> writes:
> Question: Do all these need to have a bare index just on the account_id
> column, or is a multicolumn index with account_id first
> sufficient for the check to be reasonably quick?

I would think that such an index would be sufficient, but you could
check for yourself by doing something like

    explain select 1 from <table> where account_id = <something>

and verifying that you get an indexscan not a seqscan, for each
dependent table.

            regards, tom lane


pgsql-general by date:

Previous
From: Larry Rosenman
Date:
Subject: Re: LONG delete with LOTS of FK's
Next
From: Larry Rosenman
Date:
Subject: Re: LONG delete with LOTS of FK's