Derive oldestActiveXid at correct time for Hot Standby.

There was a timing window between when oldestActiveXid was derived
and when it should have been derived that only shows itself under
heavy load. Move code around to ensure correct timing of derivation.
No change to StartupSUBTRANS() code, which is where this failed.

Bug report by Chris Redekop

5 files changed, 71 insertions, 6 deletions

diff --git a/src/backend/access/transam/xlog.c b/src/backend/access/transam/xlog.c
index bf57b3bf49..0944aa4162 100644
--- a/src/backend/access/transam/xlog.c
+++ b/src/backend/access/transam/xlog.c
@@ -7637,6 +7637,16 @@ CreateCheckPoint(int flags)
 	checkPoint.time = (pg_time_t) time(NULL);
 
 	/*
+	 * For Hot Standby, derive the oldestActiveXid before we fix the redo pointer.
+	 * This allows us to begin accumulating changes to assemble our starting
+	 * snapshot of locks and transactions.
+	 */
+	if (!shutdown && XLogStandbyInfoActive())
+		checkPoint.oldestActiveXid = GetOldestActiveTransactionId();
+	else
+		checkPoint.oldestActiveXid = InvalidTransactionId;
+
+	/*
 	 * We must hold WALInsertLock while examining insert state to determine
 	 * the checkpoint REDO pointer.
 	 */
@@ -7822,9 +7832,7 @@ CreateCheckPoint(int flags)
 	 * Update checkPoint.nextXid since we have a later value
 	 */
 	if (!shutdown && XLogStandbyInfoActive())
-		LogStandbySnapshot(&checkPoint.oldestActiveXid, &checkPoint.nextXid);
-	else
-		checkPoint.oldestActiveXid = InvalidTransactionId;
+		LogStandbySnapshot(&checkPoint.nextXid);
 
 	START_CRIT_SECTION();
 
diff --git a/src/backend/storage/ipc/procarray.c b/src/backend/storage/ipc/procarray.c
index 0284ac4082..1a48485f97 100644
--- a/src/backend/storage/ipc/procarray.c
+++ b/src/backend/storage/ipc/procarray.c
@@ -1621,6 +1621,63 @@ GetRunningTransactionData(void)
 }
 
 /*
+ * GetOldestActiveTransactionId()
+ *
+ * Similar to GetSnapshotData but returns just oldestActiveXid. We include
+ * all PGPROCs with an assigned TransactionId, even VACUUM processes.
+ * We look at all databases, though there is no need to include WALSender
+ * since this has no effect on hot standby conflicts.
+ *
+ * This is never executed during recovery so there is no need to look at
+ * KnownAssignedXids.
+ *
+ * We don't worry about updating other counters, we want to keep this as
+ * simple as possible and leave GetSnapshotData() as the primary code for
+ * that bookkeeping.
+ */
+TransactionId
+GetOldestActiveTransactionId(void)
+{
+	ProcArrayStruct *arrayP = procArray;
+	TransactionId oldestRunningXid;
+	int			index;
+
+	Assert(!RecoveryInProgress());
+
+	LWLockAcquire(ProcArrayLock, LW_SHARED);
+
+	oldestRunningXid = ShmemVariableCache->nextXid;
+
+	/*
+	 * Spin over procArray collecting all xids and subxids.
+	 */
+	for (index = 0; index < arrayP->numProcs; index++)
+	{
+		volatile PGPROC *proc = arrayP->procs[index];
+		TransactionId xid;
+
+		/* Fetch xid just once - see GetNewTransactionId */
+		xid = proc->xid;
+
+		if (!TransactionIdIsNormal(xid))
+			continue;
+
+		if (TransactionIdPrecedes(xid, oldestRunningXid))
+			oldestRunningXid = xid;
+
+		/*
+		 * Top-level XID of a transaction is always less than any of its
+		 * subxids, so we don't need to check if any of the subxids are
+		 * smaller than oldestRunningXid
+		 */
+	}
+
+	LWLockRelease(ProcArrayLock);
+
+	return oldestRunningXid;
+}
+
+/*
  * GetTransactionsInCommit -- Get the XIDs of transactions that are committing
  *
  * Constructs an array of XIDs of transactions that are currently in commit
diff --git a/src/backend/storage/ipc/standby.c b/src/backend/storage/ipc/standby.c
index 72c6b97b22..3f2fc81368 100644
--- a/src/backend/storage/ipc/standby.c
+++ b/src/backend/storage/ipc/standby.c
@@ -815,7 +815,7 @@ standby_desc(StringInfo buf, uint8 xl_info, char *rec)
  * making WAL entries.
  */
 void
-LogStandbySnapshot(TransactionId *oldestActiveXid, TransactionId *nextXid)
+LogStandbySnapshot(TransactionId *nextXid)
 {
 	RunningTransactions running;
 	xl_standby_lock *locks;
@@ -845,7 +845,6 @@ LogStandbySnapshot(TransactionId *oldestActiveXid, TransactionId *nextXid)
 	/* GetRunningTransactionData() acquired XidGenLock, we must release it */
 	LWLockRelease(XidGenLock);
 
-	*oldestActiveXid = running->oldestRunningXid;
 	*nextXid = running->nextXid;
 }
 
diff --git a/src/include/storage/procarray.h b/src/include/storage/procarray.h
index 71c82437cd..3e80cc5c07 100644
--- a/src/include/storage/procarray.h
+++ b/src/include/storage/procarray.h
@@ -50,6 +50,7 @@ extern RunningTransactions GetRunningTransactionData(void);
 extern bool TransactionIdIsInProgress(TransactionId xid);
 extern bool TransactionIdIsActive(TransactionId xid);
 extern TransactionId GetOldestXmin(bool allDbs, bool ignoreVacuum);
+extern TransactionId GetOldestActiveTransactionId(void);
 
 extern int	GetTransactionsInCommit(TransactionId **xids_p);
 extern bool HaveTransactionsInCommit(TransactionId *xids, int nxids);
diff --git a/src/include/storage/standby.h b/src/include/storage/standby.h
index 6ebac62db5..e587a5c4a4 100644
--- a/src/include/storage/standby.h
+++ b/src/include/storage/standby.h
@@ -111,6 +111,6 @@ typedef RunningTransactionsData *RunningTransactions;
 extern void LogAccessExclusiveLock(Oid dbOid, Oid relOid);
 extern void LogAccessExclusiveLockPrepare(void);
 
-extern void LogStandbySnapshot(TransactionId *oldestActiveXid, TransactionId *nextXid);
+extern void LogStandbySnapshot(TransactionId *nextXid);
 
 #endif   /* STANDBY_H */