Fix possibility of self-deadlock in ResolveRecoveryConflictWithBufferPin().

anarazel · anarazel · commit 57c5ad168be1 · 2022-05-02T18:28:10.000-07:00
The tests added in 9f8a050 failed nearly reliably on FreeBSD in CI, and occasionally on the buildfarm. That turns out to be caused not by a bug in the test, but by a longstanding bug in recovery conflict handling. The standby timeout handler, used by ResolveRecoveryConflictWithBufferPin(), executed SendRecoveryConflictWithBufferPin() inside a signal handler. A bad idea, because the deadlock timeout handler (or a spurious latch set) could have interrupted ProcWaitForSignal(). If unlucky that could cause a self-deadlock on ProcArrayLock, if the deadlock check is in SendRecoveryConflictWithBufferPin()->CancelDBBackends(). To fix, set a flag in StandbyTimeoutHandler(), and check the flag in ResolveRecoveryConflictWithBufferPin(). Subsequently the recovery conflict tests will be backpatched. Discussion: https://postgr.es/m/20220413002626.udl7lll7f3o7nre7@alap3.anarazel.de Backpatch: 10-
diff --git a/src/backend/storage/ipc/standby.c b/src/backend/storage/ipc/standby.c
@@ -44,6 +44,7 @@ static HTAB *RecoveryLockLists;
 
 /* Flags set by timeout handlers */
 static volatile sig_atomic_t got_standby_deadlock_timeout = false;
+static volatile sig_atomic_t got_standby_delay_timeout = false;
 static volatile sig_atomic_t got_standby_lock_timeout = false;
 
 static void ResolveRecoveryConflictWithVirtualXIDs(VirtualTransactionId *waitlist,
@@ -603,10 +604,15 @@ ResolveRecoveryConflictWithBufferPin(void)
 		enable_timeouts(timeouts, cnt);
 	}
 
-	/* Wait to be signaled by UnpinBuffer() */
+	/*
+	 * Wait to be signaled by UnpinBuffer() or for the wait to be interrupted
+	 * by one of the timeouts established above.
+	 */
 	ProcWaitForSignal(PG_WAIT_BUFFER_PIN);
 
-	if (got_standby_deadlock_timeout)
+	if (got_standby_delay_timeout)
+		SendRecoveryConflictWithBufferPin(PROCSIG_RECOVERY_CONFLICT_BUFFERPIN);
+	else if (got_standby_deadlock_timeout)
 	{
 		/*
 		 * Send out a request for hot-standby backends to check themselves for
@@ -632,6 +638,7 @@ ResolveRecoveryConflictWithBufferPin(void)
 	 * individually, but that'd be slower.
 	 */
 	disable_all_timeouts(false);
+	got_standby_delay_timeout = false;
 	got_standby_deadlock_timeout = false;
 }
 
@@ -691,8 +698,8 @@ CheckRecoveryConflictDeadlock(void)
  */
 
 /*
- * StandbyDeadLockHandler() will be called if STANDBY_DEADLOCK_TIMEOUT
- * occurs before STANDBY_TIMEOUT.
+ * StandbyDeadLockHandler() will be called if STANDBY_DEADLOCK_TIMEOUT is
+ * exceeded.
  */
 void
 StandbyDeadLockHandler(void)
@@ -702,16 +709,11 @@ StandbyDeadLockHandler(void)
 
 /*
  * StandbyTimeoutHandler() will be called if STANDBY_TIMEOUT is exceeded.
- * Send out a request to release conflicting buffer pins unconditionally,
- * so we can press ahead with applying changes in recovery.
  */
 void
 StandbyTimeoutHandler(void)
 {
-	/* forget any pending STANDBY_DEADLOCK_TIMEOUT request */
-	disable_timeout(STANDBY_DEADLOCK_TIMEOUT, false);
-
-	SendRecoveryConflictWithBufferPin(PROCSIG_RECOVERY_CONFLICT_BUFFERPIN);
+	got_standby_delay_timeout = true;
 }
 
 /*
