On Mon, Dec 12, 2005 at 11:09:01PM -0500, Tom Lane wrote: > "Jim C. Nasby" <jnasby@xxxxxxxxxxxxx> writes: > > On Mon, Dec 12, 2005 at 06:26:37PM -0500, Tom Lane wrote: > >> No. Avoiding that would require a new approach to > >> vacuum-vs-ordinary-indexscan interlocking, so it won't happen until > >> someone has a Bright Idea (tm). > > > Plus there is a TODO to only vacuum pages that are known to have dead > > tuples, which should hopefully mean no more index-scans during vacuum as > > well. > > No such luck. You delete any tuples, you need to scan the indexes. Even though you can see what the index values were for the now-dead tuple? -- Jim C. Nasby, Sr. Engineering Consultant jnasby@xxxxxxxxxxxxx Pervasive Software http://pervasive.com work: 512-231-6117 vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
