NotificationsYou must be signed in to change notification settings
Fork6
Star31

Commite86cfc4

committed

Don't allow relminmxid to go backwards during VACUUM FULL

We were allowing a table's pg_class.relminmxid value to move backwardswhen heaps were swapped by VACUUM FULL or CLUSTER. There is asimilar protection against relfrozenxid going backwards, which weneglected to clone when the multixact stuff was rejiggered by commit0ac5ad5.Backpatch to 9.3, where relminmxid was introduced.As reported by Heikki inhttp://www.postgresql.org/message-id/52401AEA.9000608@vmware.com

1 parent5c53844 commite86cfc4Copy full SHA for e86cfc4

File tree

1 file changed

-0

lines changed

src/backend/commands
- cluster.c

1 file changed

-0

lines changed

`‎src/backend/commands/cluster.c`

Lines changed: 6 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -873,6 +873,12 @@ copy_heap_data(Oid OIDNewHeap, Oid OIDOldHeap, Oid OIDOldIndex,`
`873`	`873`	`if (TransactionIdPrecedes(FreezeXid,OldHeap->rd_rel->relfrozenxid))`
`874`	`874`	`FreezeXid=OldHeap->rd_rel->relfrozenxid;`
`875`	`875`
	`876`	`+/*`
	`877`	`+ * MultiXactCutoff, similarly, shouldn't go backwards either.`
	`878`	`+ */`
	`879`	`+if (MultiXactIdPrecedes(MultiXactCutoff,OldHeap->rd_rel->relminmxid))`
	`880`	`+MultiXactCutoff=OldHeap->rd_rel->relminmxid;`
	`881`	`+`
`876`	`882`	`/* return selected values to caller */`
`877`	`883`	`*pFreezeXid=FreezeXid;`
`878`	`884`	`*pCutoffMulti=MultiXactCutoff;`

0 commit comments

Comments

(0)

Movatterモバイル変換

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Commite86cfc4

File tree

1 file changed

1 file changed

`‎src/backend/commands/cluster.c`

0 commit comments