pgstat: scaffolding for transactional stats creation / drop.

One problematic part of the current statistics collector design is that there
is no reliable way of getting rid of statistics entries. Because of that
pgstat_vacuum_stat() (called by [auto-]vacuum) matches all stats for the
current database with the catalog contents and tries to drop now-superfluous
entries. That's quite expensive. What's worse, it doesn't work on physical
replicas, despite physical replicas collection statistics entries.

This commit introduces infrastructure to create / drop statistics entries
transactionally, together with the underlying catalog objects (functions,
relations, subscriptions). pgstat_xact.c maintains a list of stats entries
created / dropped transactionally in the current transaction. To ensure the
removal of statistics entries is durable dropped statistics entries are
included in commit / abort (and prepare) records, which also ensures that
stats entries are dropped on standbys.

Statistics entries created separately from creating the underlying catalog
object (e.g. when stats were previously lost due to an immediate restart)
are *not* WAL logged. However that can only happen outside of the transaction
creating the catalog object, so it does not lead to "leaked" statistics
entries.

For this to work, functions creating / dropping functions / relations /
subscriptions need to call into pgstat. For subscriptions this was already
done when dropping subscriptions, via pgstat_report_subscription_drop() (now
renamed to pgstat_drop_subscription()).

This commit does not actually drop stats yet, it just provides the
infrastructure. It is however a largely independent piece of infrastructure,
so committing it separately makes sense.

Bumps XLOG_PAGE_MAGIC.

Author: Andres Freund <andres@anarazel.de>
Reviewed-By: Thomas Munro <thomas.munro@gmail.com>
Reviewed-By: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Discussion: https://postgr.es/m/20220303021600.hs34ghqcw6zcokdh@alap3.anarazel.de

1 files changed, 65 insertions, 3 deletions

diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 3596a7d7345..bf2fc08d940 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -1285,6 +1285,8 @@ RecordTransactionCommit(void)
 	RelFileNode *rels;
 	int			nchildren;
 	TransactionId *children;
+	int			ndroppedstats = 0;
+	xl_xact_stats_item *droppedstats = NULL;
 	int			nmsgs = 0;
 	SharedInvalidationMessage *invalMessages = NULL;
 	bool		RelcacheInitFileInval = false;
@@ -1303,6 +1305,7 @@ RecordTransactionCommit(void)
 	/* Get data needed for commit record */
 	nrels = smgrGetPendingDeletes(true, &rels);
 	nchildren = xactGetCommittedChildren(&children);
+	ndroppedstats = pgstat_get_transactional_drops(true, &droppedstats);
 	if (XLogStandbyInfoActive())
 		nmsgs = xactGetCommittedInvalidationMessages(&invalMessages,
 													 &RelcacheInitFileInval);
@@ -1317,10 +1320,12 @@ RecordTransactionCommit(void)
 		/*
 		 * We expect that every RelationDropStorage is followed by a catalog
 		 * update, and hence XID assignment, so we shouldn't get here with any
-		 * pending deletes.  Use a real test not just an Assert to check this,
-		 * since it's a bit fragile.
+		 * pending deletes. Same is true for dropping stats.
+		 *
+		 * Use a real test not just an Assert to check this, since it's a bit
+		 * fragile.
 		 */
-		if (nrels != 0)
+		if (nrels != 0 || ndroppedstats != 0)
 			elog(ERROR, "cannot commit a transaction that deleted files but has no xid");
 
 		/* Can't have child XIDs either; AssignTransactionId enforces this */
@@ -1395,6 +1400,7 @@ RecordTransactionCommit(void)
 
 		XactLogCommitRecord(xactStopTimestamp,
 							nchildren, children, nrels, rels,
+							ndroppedstats, droppedstats,
 							nmsgs, invalMessages,
 							RelcacheInitFileInval,
 							MyXactFlags,
@@ -1518,6 +1524,8 @@ cleanup:
 	/* Clean up local data */
 	if (rels)
 		pfree(rels);
+	if (ndroppedstats)
+		pfree(droppedstats);
 
 	return latestXid;
 }
@@ -1698,6 +1706,8 @@ RecordTransactionAbort(bool isSubXact)
 	TransactionId latestXid;
 	int			nrels;
 	RelFileNode *rels;
+	int			ndroppedstats = 0;
+	xl_xact_stats_item *droppedstats = NULL;
 	int			nchildren;
 	TransactionId *children;
 	TimestampTz xact_time;
@@ -1734,6 +1744,7 @@ RecordTransactionAbort(bool isSubXact)
 	/* Fetch the data we need for the abort record */
 	nrels = smgrGetPendingDeletes(false, &rels);
 	nchildren = xactGetCommittedChildren(&children);
+	ndroppedstats = pgstat_get_transactional_drops(false, &droppedstats);
 
 	/* XXX do we really need a critical section here? */
 	START_CRIT_SECTION();
@@ -1750,6 +1761,7 @@ RecordTransactionAbort(bool isSubXact)
 	XactLogAbortRecord(xact_time,
 					   nchildren, children,
 					   nrels, rels,
+					   ndroppedstats, droppedstats,
 					   MyXactFlags, InvalidTransactionId,
 					   NULL);
 
@@ -1796,6 +1808,8 @@ RecordTransactionAbort(bool isSubXact)
 	/* And clean up local data */
 	if (rels)
 		pfree(rels);
+	if (ndroppedstats)
+		pfree(droppedstats);
 
 	return latestXid;
 }
@@ -5573,6 +5587,7 @@ XLogRecPtr
 XactLogCommitRecord(TimestampTz commit_time,
 					int nsubxacts, TransactionId *subxacts,
 					int nrels, RelFileNode *rels,
+					int ndroppedstats, xl_xact_stats_item *droppedstats,
 					int nmsgs, SharedInvalidationMessage *msgs,
 					bool relcacheInval,
 					int xactflags, TransactionId twophase_xid,
@@ -5583,6 +5598,7 @@ XactLogCommitRecord(TimestampTz commit_time,
 	xl_xact_dbinfo xl_dbinfo;
 	xl_xact_subxacts xl_subxacts;
 	xl_xact_relfilenodes xl_relfilenodes;
+	xl_xact_stats_items xl_dropped_stats;
 	xl_xact_invals xl_invals;
 	xl_xact_twophase xl_twophase;
 	xl_xact_origin xl_origin;
@@ -5640,6 +5656,12 @@ XactLogCommitRecord(TimestampTz commit_time,
 		info |= XLR_SPECIAL_REL_UPDATE;
 	}
 
+	if (ndroppedstats > 0)
+	{
+		xl_xinfo.xinfo |= XACT_XINFO_HAS_DROPPED_STATS;
+		xl_dropped_stats.nitems = ndroppedstats;
+	}
+
 	if (nmsgs > 0)
 	{
 		xl_xinfo.xinfo |= XACT_XINFO_HAS_INVALS;
@@ -5696,6 +5718,14 @@ XactLogCommitRecord(TimestampTz commit_time,
 						 nrels * sizeof(RelFileNode));
 	}
 
+	if (xl_xinfo.xinfo & XACT_XINFO_HAS_DROPPED_STATS)
+	{
+		XLogRegisterData((char *) (&xl_dropped_stats),
+						 MinSizeOfXactStatsItems);
+		XLogRegisterData((char *) droppedstats,
+						 ndroppedstats * sizeof(xl_xact_stats_item));
+	}
+
 	if (xl_xinfo.xinfo & XACT_XINFO_HAS_INVALS)
 	{
 		XLogRegisterData((char *) (&xl_invals), MinSizeOfXactInvals);
@@ -5729,6 +5759,7 @@ XLogRecPtr
 XactLogAbortRecord(TimestampTz abort_time,
 				   int nsubxacts, TransactionId *subxacts,
 				   int nrels, RelFileNode *rels,
+				   int ndroppedstats, xl_xact_stats_item *droppedstats,
 				   int xactflags, TransactionId twophase_xid,
 				   const char *twophase_gid)
 {
@@ -5736,6 +5767,7 @@ XactLogAbortRecord(TimestampTz abort_time,
 	xl_xact_xinfo xl_xinfo;
 	xl_xact_subxacts xl_subxacts;
 	xl_xact_relfilenodes xl_relfilenodes;
+	xl_xact_stats_items xl_dropped_stats;
 	xl_xact_twophase xl_twophase;
 	xl_xact_dbinfo xl_dbinfo;
 	xl_xact_origin xl_origin;
@@ -5773,6 +5805,12 @@ XactLogAbortRecord(TimestampTz abort_time,
 		info |= XLR_SPECIAL_REL_UPDATE;
 	}
 
+	if (ndroppedstats > 0)
+	{
+		xl_xinfo.xinfo |= XACT_XINFO_HAS_DROPPED_STATS;
+		xl_dropped_stats.nitems = ndroppedstats;
+	}
+
 	if (TransactionIdIsValid(twophase_xid))
 	{
 		xl_xinfo.xinfo |= XACT_XINFO_HAS_TWOPHASE;
@@ -5834,6 +5872,14 @@ XactLogAbortRecord(TimestampTz abort_time,
 						 nrels * sizeof(RelFileNode));
 	}
 
+	if (xl_xinfo.xinfo & XACT_XINFO_HAS_DROPPED_STATS)
+	{
+		XLogRegisterData((char *) (&xl_dropped_stats),
+						 MinSizeOfXactStatsItems);
+		XLogRegisterData((char *) droppedstats,
+						 ndroppedstats * sizeof(xl_xact_stats_item));
+	}
+
 	if (xl_xinfo.xinfo & XACT_XINFO_HAS_TWOPHASE)
 	{
 		XLogRegisterData((char *) (&xl_twophase), sizeof(xl_xact_twophase));
@@ -5967,6 +6013,14 @@ xact_redo_commit(xl_xact_parsed_commit *parsed,
 		DropRelationFiles(parsed->xnodes, parsed->nrels, true);
 	}
 
+	if (parsed->nstats > 0)
+	{
+		/* see equivalent call for relations above */
+		XLogFlush(lsn);
+
+		pgstat_execute_transactional_drops(parsed->nstats, parsed->stats, true);
+	}
+
 	/*
 	 * We issue an XLogFlush() for the same reason we emit ForceSyncCommit()
 	 * in normal operation. For example, in CREATE DATABASE, we copy all files
@@ -6069,6 +6123,14 @@ xact_redo_abort(xl_xact_parsed_abort *parsed, TransactionId xid,
 
 		DropRelationFiles(parsed->xnodes, parsed->nrels, true);
 	}
+
+	if (parsed->nstats > 0)
+	{
+		/* see equivalent call for relations above */
+		XLogFlush(lsn);
+
+		pgstat_execute_transactional_drops(parsed->nstats, parsed->stats, true);
+	}
 }
 
 void