Do not decode TOAST data for table rewrites

tvondra · tvondra · commit 4e7395d83faf · 2018-11-28T01:44:11.000+01:00
During table rewrites (VACUUM FULL and CLUSTER), the main heap is logged using XLOG / FPI records, and thus (correctly) ignored in decoding. But the associated TOAST table is WAL-logged as plain INSERT records, and so was logically decoded and passed to reorder buffer. That has severe consequences with TOAST tables of non-trivial size. Firstly, reorder buffer has to keep all those changes, possibly spilling them to a file, incurring I/O costs and disk space. Secondly, ReoderBufferCommit() was stashing all those TOAST chunks into a hash table, which got discarded only after processing the row from the main heap. But as the main heap is not decoded for rewrites, this never happened, so all the TOAST data accumulated in memory, resulting either in excessive memory consumption or OOM. The fix is simple, as commit e9edc1b already introduced infrastructure (namely HEAP_INSERT_NO_LOGICAL flag) to skip logical decoding of TOAST tables, but it only applied it to system tables. So simply use it for all TOAST data in raw_heap_insert(). That would however solve only the memory consumption issue - the TOAST changes would still be decoded and added to the reorder buffer, and spilled to disk (although without TOAST tuple data, so much smaller). But we can solve that by tweaking DecodeInsert() to just ignore such INSERT records altogether, using XLH_INSERT_CONTAINS_NEW_TUPLE flag, instead of skipping them later in ReorderBufferCommit(). Review: Masahiko Sawada Discussion: https://www.postgresql.org/message-id/flat/1a17c643-e9af-3dba-486b-fbe31bc1823a%402ndquadrant.com Backpatch: 9.4-, where logical decoding was introduced
diff --git a/src/backend/access/heap/rewriteheap.c b/src/backend/access/heap/rewriteheap.c
@@ -658,12 +658,11 @@ raw_heap_insert(RewriteState state, HeapTuple tup)
 			options |= HEAP_INSERT_SKIP_WAL;
 
 		/*
-		 * The new relfilenode's relcache entrye doesn't have the necessary
-		 * information to determine whether a relation should emit data for
-		 * logical decoding.  Force it to off if necessary.
+		 * While rewriting the heap for VACUUM FULL / CLUSTER, make sure data
+		 * for the TOAST table are not logically decoded.  The main heap is
+		 * WAL-logged as XLOG FPI records, which are not logically decoded.
 		 */
-		if (!RelationIsLogicallyLogged(state->rs_old_rel))
-			options |= HEAP_INSERT_NO_LOGICAL;
+		options |= HEAP_INSERT_NO_LOGICAL;
 
 		heaptup = toast_insert_or_update(state->rs_new_rel, tup, NULL,
 										 options);
diff --git a/src/backend/replication/logical/decode.c b/src/backend/replication/logical/decode.c
@@ -639,13 +639,23 @@ DecodeAbort(LogicalDecodingContext *ctx, XLogRecordBuffer *buf,
 static void
 DecodeInsert(LogicalDecodingContext *ctx, XLogRecordBuffer *buf)
 {
+	Size		datalen;
+	char	   *tupledata;
+	Size		tuplelen;
 	XLogReaderState *r = buf->record;
 	xl_heap_insert *xlrec;
 	ReorderBufferChange *change;
 	RelFileNode target_node;
 
 	xlrec = (xl_heap_insert *) XLogRecGetData(r);
 
+	/*
+	 * Ignore insert records without new tuples (this does happen when
+	 * raw_heap_insert marks the TOAST record as HEAP_INSERT_NO_LOGICAL).
+	 */
+	if (!(xlrec->flags & XLH_INSERT_CONTAINS_NEW_TUPLE))
+		return;
+
 	/* only interested in our database */
 	XLogRecGetBlockTag(r, 0, &target_node, NULL, NULL);
 	if (target_node.dbNode != ctx->slot->data.database)
@@ -664,17 +674,13 @@ DecodeInsert(LogicalDecodingContext *ctx, XLogRecordBuffer *buf)
 
 	memcpy(&change->data.tp.relnode, &target_node, sizeof(RelFileNode));
 
-	if (xlrec->flags & XLH_INSERT_CONTAINS_NEW_TUPLE)
-	{
-		Size		datalen;
-		char	   *tupledata = XLogRecGetBlockData(r, 0, &datalen);
-		Size		tuplelen = datalen - SizeOfHeapHeader;
+	tupledata = XLogRecGetBlockData(r, 0, &datalen);
+	tuplelen = datalen - SizeOfHeapHeader;
 
-		change->data.tp.newtuple =
-			ReorderBufferGetTupleBuf(ctx->reorder, tuplelen);
+	change->data.tp.newtuple =
+		ReorderBufferGetTupleBuf(ctx->reorder, tuplelen);
 
-		DecodeXLogTuple(tupledata, datalen, change->data.tp.newtuple);
-	}
+	DecodeXLogTuple(tupledata, datalen, change->data.tp.newtuple);
 
 	change->data.tp.clear_toast_afterwards = true;
 
diff --git a/src/backend/replication/logical/reorderbuffer.c b/src/backend/replication/logical/reorderbuffer.c
@@ -1599,17 +1599,12 @@ ReorderBufferCommit(ReorderBuffer *rb, TransactionId xid,
 						 * transaction's changes. Otherwise it will get
 						 * freed/reused while restoring spooled data from
 						 * disk.
-						 *
-						 * But skip doing so if there's no tuple-data. That
-						 * happens if a non-mapped system catalog with a toast
-						 * table is rewritten.
 						 */
-						if (change->data.tp.newtuple != NULL)
-						{
-							dlist_delete(&change->node);
-							ReorderBufferToastAppendChunk(rb, txn, relation,
-														  change);
-						}
+						Assert(change->data.tp.newtuple != NULL);
+
+						dlist_delete(&change->node);
+						ReorderBufferToastAppendChunk(rb, txn, relation,
+													  change);
 					}
 
 			change_done:
