Avoid having autovacuum workers wait for relation locks.

Waiting for relation locks can lead to starvation - it pins down an
autovacuum worker for as long as the lock is held.  But if we're doing
an anti-wraparound vacuum, then we still wait; maintenance can no longer
be put off.

To assist with troubleshooting, if log_autovacuum_min_duration >= 0,
we log whenever an autovacuum or autoanalyze is skipped for this reason.

Per a gripe by Josh Berkus, and ensuing discussion.

doc/src/sgml/config.sgml

@@ -4080,7 +4080,10 @@ COPY postgres_log FROM '/full/path/to/logfile.csv' WITH csv;
         all autovacuum actions. Minus-one (the default) disables logging
         autovacuum actions.  For example, if you set this to
         <literal>250ms</literal> then all automatic vacuums and analyzes that run
-        250ms or longer will be logged.  Enabling this parameter can be helpful
+        250ms or longer will be logged.  In addition, when this parameter is
+        set to any value other than <literal>-1</literal>, a message will be
+        logged if an autovacuum action is skipped due to the existence of a
+        conflicting lock.  Enabling this parameter can be helpful
         in tracking autovacuum activity.  This setting can only be set in
         the <filename>postgresql.conf</> file or on the server command line.
        </para>

src/backend/commands/analyze.c

@@ -36,6 +36,7 @@
 #include "pgstat.h"
 #include "postmaster/autovacuum.h"
 #include "storage/bufmgr.h"
+#include "storage/lmgr.h"
 #include "storage/proc.h"
 #include "storage/procarray.h"
 #include "utils/acl.h"
@@ -148,7 +149,19 @@ analyze_rel(Oid relid, VacuumStmt *vacstmt,
 	 * matter if we ever try to accumulate stats on dead tuples.) If the rel
 	 * has been dropped since we last saw it, we don't need to process it.
 	 */
-	onerel = try_relation_open(relid, ShareUpdateExclusiveLock);
+	if (!(vacstmt->options & VACOPT_NOWAIT))
+		onerel = try_relation_open(relid, ShareUpdateExclusiveLock);
+	else if (ConditionalLockRelationOid(relid, ShareUpdateExclusiveLock))
+		onerel = try_relation_open(relid, NoLock);
+	else
+	{
+		onerel = NULL;
+		if (IsAutoVacuumWorkerProcess() && Log_autovacuum_min_duration >= 0)
+			ereport(LOG,
+					(errcode(ERRCODE_LOCK_NOT_AVAILABLE),
+					 errmsg("skipping analyze of \"%s\" --- lock not available",
+						vacstmt->relation->relname)));
+	}
 	if (!onerel)
 		return;
 

src/backend/commands/vacuum.c

@@ -61,7 +61,7 @@ static BufferAccessStrategy vac_strategy;
 /* non-export function prototypes */
 static List *get_rel_oids(Oid relid, const RangeVar *vacrel);
 static void vac_truncate_clog(TransactionId frozenXID);
-static void vacuum_rel(Oid relid, VacuumStmt *vacstmt, bool do_toast,
+static bool vacuum_rel(Oid relid, VacuumStmt *vacstmt, bool do_toast,
 		   bool for_wraparound, bool *scanned_all);
 
 
@@ -226,8 +226,11 @@ vacuum(VacuumStmt *vacstmt, Oid relid, bool do_toast,
 			bool		scanned_all = false;
 
 			if (vacstmt->options & VACOPT_VACUUM)
-				vacuum_rel(relid, vacstmt, do_toast, for_wraparound,
+			{
-						   &scanned_all);
+				if (!vacuum_rel(relid, vacstmt, do_toast, for_wraparound,
+								&scanned_all))
+					continue;
+			}
 
 			if (vacstmt->options & VACOPT_ANALYZE)
 			{
@@ -764,7 +767,7 @@ vac_truncate_clog(TransactionId frozenXID)
  *
  *		At entry and exit, we are not inside a transaction.
  */
-static void
+static bool
 vacuum_rel(Oid relid, VacuumStmt *vacstmt, bool do_toast, bool for_wraparound,
 		   bool *scanned_all)
 {
@@ -835,14 +838,29 @@ vacuum_rel(Oid relid, VacuumStmt *vacstmt, bool do_toast, bool for_wraparound,
 	 *
 	 * There's a race condition here: the rel may have gone away since the
 	 * last time we saw it.  If so, we don't need to vacuum it.
+	 *
+	 * If we've been asked not to wait for the relation lock, acquire it
+	 * first in non-blocking mode, before calling try_relation_open().
 	 */
-	onerel = try_relation_open(relid, lmode);
+	if (!(vacstmt->options & VACOPT_NOWAIT))
+		onerel = try_relation_open(relid, lmode);
+	else if (ConditionalLockRelationOid(relid, lmode))
+		onerel = try_relation_open(relid, NoLock);
+	else
+	{
+		onerel = NULL;
+		if (IsAutoVacuumWorkerProcess() && Log_autovacuum_min_duration >= 0)
+			ereport(LOG,
+					(errcode(ERRCODE_LOCK_NOT_AVAILABLE),
+					 errmsg("skipping vacuum of \"%s\" --- lock not available",
+						vacstmt->relation->relname)));
+	}
 
 	if (!onerel)
 	{
 		PopActiveSnapshot();
 		CommitTransactionCommand();
-		return;
+		return false;
 	}
 
 	/*
@@ -873,7 +891,7 @@ vacuum_rel(Oid relid, VacuumStmt *vacstmt, bool do_toast, bool for_wraparound,
 		relation_close(onerel, lmode);
 		PopActiveSnapshot();
 		CommitTransactionCommand();
-		return;
+		return false;
 	}
 
 	/*
@@ -890,7 +908,7 @@ vacuum_rel(Oid relid, VacuumStmt *vacstmt, bool do_toast, bool for_wraparound,
 		relation_close(onerel, lmode);
 		PopActiveSnapshot();
 		CommitTransactionCommand();
-		return;
+		return false;
 	}
 
 	/*
@@ -905,7 +923,7 @@ vacuum_rel(Oid relid, VacuumStmt *vacstmt, bool do_toast, bool for_wraparound,
 		relation_close(onerel, lmode);
 		PopActiveSnapshot();
 		CommitTransactionCommand();
-		return;
+		return false;
 	}
 
 	/*
@@ -989,6 +1007,9 @@ vacuum_rel(Oid relid, VacuumStmt *vacstmt, bool do_toast, bool for_wraparound,
 	 * Now release the session-level lock on the master table.
 	 */
 	UnlockRelationIdForSession(&onerelid, lmode);
+
+	/* Report that we really did it. */
+	return true;
 }
 
 

src/backend/postmaster/autovacuum.c

@@ -2671,19 +2671,27 @@ autovacuum_do_vac_analyze(autovac_table *tab,
 						  BufferAccessStrategy bstrategy)
 {
 	VacuumStmt	vacstmt;
+	RangeVar	rangevar;
 
-	/* Set up command parameters --- use a local variable instead of palloc */
+	/* Set up command parameters --- use local variables instead of palloc */
 	MemSet(&vacstmt, 0, sizeof(vacstmt));
+	MemSet(&rangevar, 0, sizeof(rangevar));
+
+	rangevar.schemaname = tab->at_nspname;
+	rangevar.relname = tab->at_relname;
+	rangevar.location = -1;
 
 	vacstmt.type = T_VacuumStmt;
-	vacstmt.options = 0;
+	if (!tab->at_wraparound)
+		vacstmt.options = VACOPT_NOWAIT;
 	if (tab->at_dovacuum)
 		vacstmt.options |= VACOPT_VACUUM;
 	if (tab->at_doanalyze)
 		vacstmt.options |= VACOPT_ANALYZE;
 	vacstmt.freeze_min_age = tab->at_freeze_min_age;
 	vacstmt.freeze_table_age = tab->at_freeze_table_age;
-	vacstmt.relation = NULL;	/* not used since we pass a relid */
+	/* we pass the OID, but might need this anyway for an error message */
+	vacstmt.relation = &rangevar;
 	vacstmt.va_cols = NIL;
 
 	/* Let pgstat know what we're doing */

src/include/nodes/parsenodes.h

@@ -2332,7 +2332,8 @@ typedef enum VacuumOption
 	VACOPT_ANALYZE = 1 << 1,	/* do ANALYZE */
 	VACOPT_VERBOSE = 1 << 2,	/* print progress info */
 	VACOPT_FREEZE = 1 << 3,		/* FREEZE option */
-	VACOPT_FULL = 1 << 4		/* FULL (non-concurrent) vacuum */
+	VACOPT_FULL = 1 << 4,		/* FULL (non-concurrent) vacuum */
+	VACOPT_NOWAIT = 1 << 5
 } VacuumOption;
 
 typedef struct VacuumStmt