Fix snapshot handling in logicalmsg_decode

tvondra · tvondra · commit 497f863f0598 · 2023-02-22T15:50:37.000+01:00
Whe decoding a transactional logical message, logicalmsg_decode called SnapBuildGetOrBuildSnapshot. But we may not have a consistent snapshot yet at that point. We don't actually need the snapshot in this case (during replay we'll have the snapshot from the transaction), so in practice this is harmless. But in assert-enabled build this crashes. Fixed by requesting the snapshot only in non-transactional case, where we are guaranteed to have SNAPBUILD_CONSISTENT. Backpatch to 11. The issue exists since 9.6. Backpatch-through: 11 Reviewed-by: Andres Freund Discussion: https://postgr.es/m/84d60912-6eab-9b84-5de3-41765a5449e8@enterprisedb.com
diff --git a/src/backend/replication/logical/decode.c b/src/backend/replication/logical/decode.c
@@ -517,7 +517,7 @@ DecodeLogicalMsgOp(LogicalDecodingContext *ctx, XLogRecordBuffer *buf)
 	TransactionId xid = XLogRecGetXid(r);
 	uint8		info = XLogRecGetInfo(r) & ~XLR_INFO_MASK;
 	RepOriginId origin_id = XLogRecGetOrigin(r);
-	Snapshot	snapshot;
+	Snapshot	snapshot = NULL;
 	xl_logical_message *message;
 
 	if (info != XLOG_LOGICAL_MESSAGE)
@@ -547,7 +547,17 @@ DecodeLogicalMsgOp(LogicalDecodingContext *ctx, XLogRecordBuffer *buf)
 			  SnapBuildXactNeedsSkip(builder, buf->origptr)))
 		return;
 
-	snapshot = SnapBuildGetOrBuildSnapshot(builder, xid);
+	/*
+	 * If this is a non-transactional change, get the snapshot we're expected
+	 * to use. We only get here when the snapshot is consistent, and the
+	 * change is not meant to be skipped.
+	 *
+	 * For transactional changes we don't need a snapshot, we'll use the
+	 * regular snapshot maintained by ReorderBuffer. We just leave it NULL.
+	 */
+	if (!message->transactional)
+		snapshot = SnapBuildGetOrBuildSnapshot(builder, xid);
+
 	ReorderBufferQueueMessage(ctx->reorder, xid, snapshot, buf->endptr,
 							  message->transactional,
 							  message->message, /* first part of message is
diff --git a/src/backend/replication/logical/reorderbuffer.c b/src/backend/replication/logical/reorderbuffer.c
@@ -622,6 +622,13 @@ ReorderBufferQueueMessage(ReorderBuffer *rb, TransactionId xid,
 
 		Assert(xid != InvalidTransactionId);
 
+		/*
+		 * We don't expect snapshots for transactional changes - we'll use the
+		 * snapshot derived later during apply (unless the change gets
+		 * skipped).
+		 */
+		Assert(!snapshot);
+
 		oldcontext = MemoryContextSwitchTo(rb->context);
 
 		change = ReorderBufferGetChange(rb);
@@ -640,6 +647,9 @@ ReorderBufferQueueMessage(ReorderBuffer *rb, TransactionId xid,
 		ReorderBufferTXN *txn = NULL;
 		volatile Snapshot snapshot_now = snapshot;
 
+		/* Non-transactional changes require a valid snapshot. */
+		Assert(snapshot_now);
+
 		if (xid != InvalidTransactionId)
 			txn = ReorderBufferTXNByXid(rb, xid, true, NULL, lsn, true);
 
