Fix race condition in relcache init file invalidation.

The previous code tried to synchronize by unlinking the init file twice,
but that doesn't actually work: it leaves a window wherein a third process
could read the already-stale init file but miss the SI messages that would
tell it the data is stale.  The result would be bizarre failures in catalog
accesses, typically "could not read block 0 in file ..." later during
startup.

Instead, hold RelCacheInitLock across both the unlink and the sending of
the SI messages.  This is more straightforward, and might even be a bit
faster since only one unlink call is needed.

This has been wrong since it was put in (in 2002!), so back-patch to all
supported releases.

src/backend/utils/cache/inval.c

@@ -767,10 +767,10 @@ inval_twophase_postcommit(TransactionId xid, uint16 info,
 			SendSharedInvalidMessage(msg);
 			break;
 		case TWOPHASE_INFO_FILE_BEFORE:
-			RelationCacheInitFileInvalidate(true);
+			RelationCacheInitFilePreInvalidate();
 			break;
 		case TWOPHASE_INFO_FILE_AFTER:
-			RelationCacheInitFileInvalidate(false);
+			RelationCacheInitFilePostInvalidate();
 			break;
 		default:
 			Assert(false);
@@ -817,7 +817,7 @@ AtEOXact_Inval(bool isCommit)
 		 * unless we committed.
 		 */
 		if (transInvalInfo->RelcacheInitFileInval)
-			RelationCacheInitFileInvalidate(true);
+			RelationCacheInitFilePreInvalidate();
 
 		AppendInvalidationMessages(&transInvalInfo->PriorCmdInvalidMsgs,
 								   &transInvalInfo->CurrentCmdInvalidMsgs);
@@ -826,7 +826,7 @@ AtEOXact_Inval(bool isCommit)
 									SendSharedInvalidMessage);
 
 		if (transInvalInfo->RelcacheInitFileInval)
-			RelationCacheInitFileInvalidate(false);
+			RelationCacheInitFilePostInvalidate();
 	}
 	else if (transInvalInfo != NULL)
 	{

src/backend/utils/cache/relcache.c

@@ -3642,8 +3642,8 @@ write_relcache_init_file(void)
 	 * updated by SI message processing, but we can't be sure whether what we
 	 * wrote out was up-to-date.)
 	 *
-	 * This mustn't run concurrently with RelationCacheInitFileInvalidate, so
+	 * This mustn't run concurrently with the code that unlinks an init file
-	 * grab a serialization lock for the duration.
+	 * and sends SI messages, so grab a serialization lock for the duration.
 	 */
 	LWLockAcquire(RelCacheInitLock, LW_EXCLUSIVE);
 
@@ -3707,49 +3707,55 @@ RelationIdIsInInitFile(Oid relationId)
  * changed one or more of the relation cache entries that are kept in the
  * init file.
  *
- * We actually need to remove the init file twice: once just before sending
+ * To be safe against concurrent inspection or rewriting of the init file,
- * the SI messages that include relcache inval for such relations, and once
+ * we must take RelCacheInitLock, then remove the old init file, then send
- * just after sending them.  The unlink before ensures that a backend that's
+ * the SI messages that include relcache inval for such relations, and then
- * currently starting cannot read the now-obsolete init file and then miss
+ * release RelCacheInitLock.  This serializes the whole affair against
- * the SI messages that will force it to update its relcache entries.  (This
+ * write_relcache_init_file, so that we can be sure that any other process
- * works because the backend startup sequence gets into the PGPROC array before
+ * that's concurrently trying to create a new init file won't move an
- * trying to load the init file.)  The unlink after is to synchronize with a
+ * already-stale version into place after we unlink.  Also, because we unlink
- * backend that may currently be trying to write an init file based on data
+ * before sending the SI messages, a backend that's currently starting cannot
- * that we've just rendered invalid.  Such a backend will see the SI messages,
+ * read the now-obsolete init file and then miss the SI messages that will
- * but we can't leave the init file sitting around to fool later backends.
+ * force it to update its relcache entries.  (This works because the backend
+ * startup sequence gets into the sinval array before trying to load the init
+ * file.)
  *
- * Ignore any failure to unlink the file, since it might not be there if
+ * We take the lock and do the unlink in RelationCacheInitFilePreInvalidate,
- * no backend has been started since the last removal.
+ * then release the lock in RelationCacheInitFilePostInvalidate.  Caller must
+ * send any pending SI messages between those calls.
  */
 void
-RelationCacheInitFileInvalidate(bool beforeSend)
+RelationCacheInitFilePreInvalidate(void)
 {
 	char		initfilename[MAXPGPATH];
 
 	snprintf(initfilename, sizeof(initfilename), "%s/%s",
 			 DatabasePath, RELCACHE_INIT_FILENAME);
 
-	if (beforeSend)
+	LWLockAcquire(RelCacheInitLock, LW_EXCLUSIVE);
-	{
+
-		/* no interlock needed here */
+	if (unlink(initfilename) < 0)
-		unlink(initfilename);
-	}
-	else
 	{
 		/*
-		 * We need to interlock this against write_relcache_init_file, to
+		 * The file might not be there if no backend has been started since
-		 * guard against possibility that someone renames a new-but-
+		 * the last removal.  But complain about failures other than ENOENT.
-		 * already-obsolete init file into place just after we unlink. With
+		 * Fortunately, it's not too late to abort the transaction if we
-		 * the interlock, it's certain that write_relcache_init_file will
+		 * can't get rid of the would-be-obsolete init file.
-		 * notice our SI inval message before renaming into place, or else
-		 * that we will execute second and successfully unlink the file.
 		 */
-		LWLockAcquire(RelCacheInitLock, LW_EXCLUSIVE);
+		if (errno != ENOENT)
-		unlink(initfilename);
+			ereport(ERROR,
-		LWLockRelease(RelCacheInitLock);
+					(errcode_for_file_access(),
+					 errmsg("could not remove cache file \"%s\": %m",
+							initfilename)));
 	}
 }
 
+void
+RelationCacheInitFilePostInvalidate(void)
+{
+	LWLockRelease(RelCacheInitLock);
+}
+
 /*
  * Remove the init file for a given database during postmaster startup.
  *

src/include/utils/relcache.h

@@ -68,7 +68,8 @@ extern void AtEOSubXact_RelationCache(bool isCommit, SubTransactionId mySubid,
  * Routines to help manage rebuilding of relcache init file
  */
 extern bool RelationIdIsInInitFile(Oid relationId);
-extern void RelationCacheInitFileInvalidate(bool beforeSend);
+extern void RelationCacheInitFilePreInvalidate(void);
+extern void RelationCacheInitFilePostInvalidate(void);
 extern void RelationCacheInitFileRemove(const char *dbPath);
 
 /* should be used only by relcache.c and catcache.c */