database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-11-16 15:02:33 +03:00
Code Activity

Avoid early reuse of btree pages, causing incorrect query results.

Browse Source 
When we allowed read-only transactions to skip assigning XIDs
we introduced the possibility that a fully deleted btree page
could be reused. This broke the index link sequence which could
then lead to indexscans silently returning fewer rows than would
have been correct. The actual incidence of silent errors from
this is thought to be very low because of the exact workload
required and locking pre-conditions. Fix is to remove pages only
if index page opaque->btpo.xact precedes RecentGlobalXmin.

Noah Misch, reviewed by Simon Riggs
This commit is contained in:
Simon Riggs
2012-06-01 12:21:45 +01:00
parent 3e4d3a32b2
commit d3abbbebe5
3 changed files with 20 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
src/backend/access/nbtree/nbtpage.c
View File

@@ -558,19 +558,9 @@ _bt_getbuf(Relation rel, BlockNumber blkno, int access)
*/
if (XLogStandbyInfoActive())
{
TransactionId latestRemovedXid;
BTPageOpaque opaque = (BTPageOpaque) PageGetSpecialPointer(page);
/*
* opaque->btpo.xact is the threshold value not the
* value to measure conflicts against. We must retreat
* by one from it to get the correct conflict xid.
*/
latestRemovedXid = opaque->btpo.xact;
TransactionIdRetreat(latestRemovedXid);
_bt_log_reuse_page(rel, blkno, latestRemovedXid);
_bt_log_reuse_page(rel, blkno, opaque->btpo.xact);
}
/* Okay to use page. Re-initialize and return it */
@@ -685,7 +675,6 @@ bool
_bt_page_recyclable(Page page)
{
BTPageOpaque opaque;
TransactionId cutoff;
/*
* It's possible to find an all-zeroes page in an index --- for example, a
@@ -698,18 +687,11 @@ _bt_page_recyclable(Page page)
/*
* Otherwise, recycle if deleted and too old to have any processes
* interested in it. If we are generating records for Hot Standby
* defer page recycling until RecentGlobalXmin to respect user
* controls specified by vacuum_defer_cleanup_age or hot_standby_feedback.
* interested in it.
*/
if (XLogStandbyInfoActive())
cutoff = RecentGlobalXmin;
else
cutoff = RecentXmin;
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
if (P_ISDELETED(opaque) &&
TransactionIdPrecedesOrEquals(opaque->btpo.xact, cutoff))
TransactionIdPrecedes(opaque->btpo.xact, RecentGlobalXmin))
return true;
return false;
}
@@ -1376,7 +1358,13 @@ _bt_pagedel(Relation rel, Buffer buf, BTStack stack)
/*
* Mark the page itself deleted. It can be recycled when all current
* transactions are gone.
* transactions are gone. Storing GetTopTransactionId() would work, but
* we're in VACUUM and would not otherwise have an XID. Having already
* updated links to the target, ReadNewTransactionId() suffices as an
* upper bound. Any scan having retained a now-stale link is advertising
* in its PGXACT an xmin less than or equal to the value we read here. It
* will continue to do so, holding back RecentGlobalXmin, for the duration
* of that scan.
*/
page = BufferGetPage(buf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);