Commit119fd40

committed

doc: document the need to analyze partitioned tables

Autovacuum does not do it.Reported-by: Justin PryzbyDiscussion:https://postgr.es/m/20210913035409.GA10647@telsasoft.comBackpatch-through: 11

1 parentb9bb026 commit119fd40Copy full SHA for 119fd40

File tree

-4

lines changed

-4

lines changed

Lines changed: 9 additions & 4 deletions

Original file line number	Diff line number	Diff line change
`@@ -860,10 +860,15 @@ analyze threshold = analyze base threshold + analyze scale factor * number of tu`
`860`	`860`	`</para>`
`861`	`861`
`862`	`862`	`<para>`
`863`		`- Partitioned tables are not processed by autovacuum. Statistics`
`864`		`- should be collected by running a manual <command>ANALYZE</command> when it is`
`865`		`- first populated, and again whenever the distribution of data in its`
`866`		`- partitions changes significantly.`
	`863`	`+ Partitioned tables do not directly store tuples and consequently`
	`864`	`+ are not processed by autovacuum. (Autovacuum does process table`
	`865`	`+ partitions just like other tables.) Unfortunately, this means that`
	`866`	`+ autovacuum does not run <command>ANALYZE</command> on partitioned`
	`867`	`+ tables, and this can cause suboptimal plans for queries that reference`
	`868`	`+ partitioned table statistics. You can work around this problem by`
	`869`	`+ manually running <command>ANALYZE</command> on partitioned tables`
	`870`	`+ when they are first populated, and again whenever the distribution`
	`871`	`+ of data in their partitions changes significantly.`
`867`	`872`	`</para>`
`868`	`873`
`869`	`874`	`<para>`

Comments

(0)