ALTER TABLE ... DETACH PARTITION ... CONCURRENTLY

Allow a partition be detached from its partitioned table without
blocking concurrent queries, by running in two transactions and only
requiring ShareUpdateExclusive in the partitioned table.

Because it runs in two transactions, it cannot be used in a transaction
block.  This is the main reason to use dedicated syntax: so that users
can choose to use the original mode if they need it.  But also, it
doesn't work when a default partition exists (because an exclusive lock
would still need to be obtained on it, in order to change its partition
constraint.)

In case the second transaction is cancelled or a crash occurs, there's
ALTER TABLE .. DETACH PARTITION .. FINALIZE, which executes the final
steps.

The main trick to make this work is the addition of column
pg_inherits.inhdetachpending, initially false; can only be set true in
the first part of this command.  Once that is committed, concurrent
transactions that use a PartitionDirectory will include or ignore
partitions so marked: in optimizer they are ignored if the row is marked
committed for the snapshot; in executor they are always included.  As a
result, and because of the way PartitionDirectory caches partition
descriptors, queries that were planned before the detach will see the
rows in the detached partition and queries that are planned after the
detach, won't.

A CHECK constraint is created that duplicates the partition constraint.
This is probably not strictly necessary, and some users will prefer to
remove it afterwards, but if the partition is re-attached to a
partitioned table, the constraint needn't be rechecked.

Author: Álvaro Herrera <[email protected]>
Reviewed-by: Amit Langote <[email protected]>
Reviewed-by: Justin Pryzby <[email protected]>
Discussion: https://postgr.es/m/[email protected]

1 files changed, 30 insertions, 8 deletions

diff --git a/src/backend/catalog/partition.c b/src/backend/catalog/partition.c
index af7754d6ab7..790f4ccb927 100644
--- a/src/backend/catalog/partition.c
+++ b/src/backend/catalog/partition.c
@@ -32,7 +32,8 @@
 #include "utils/rel.h"
 #include "utils/syscache.h"
 
-static Oid	get_partition_parent_worker(Relation inhRel, Oid relid);
+static Oid	get_partition_parent_worker(Relation inhRel, Oid relid,
+										bool *detach_pending);
 static void get_partition_ancestors_worker(Relation inhRel, Oid relid,
 										   List **ancestors);
 
@@ -42,23 +43,32 @@ static void get_partition_ancestors_worker(Relation inhRel, Oid relid,
  *
  * Returns inheritance parent of a partition by scanning pg_inherits
  *
+ * If the partition is in the process of being detached, an error is thrown,
+ * unless even_if_detached is passed as true.
+ *
  * Note: Because this function assumes that the relation whose OID is passed
  * as an argument will have precisely one parent, it should only be called
  * when it is known that the relation is a partition.
  */
 Oid
-get_partition_parent(Oid relid)
+get_partition_parent(Oid relid, bool even_if_detached)
 {
 	Relation	catalogRelation;
 	Oid			result;
+	bool		detach_pending;
 
 	catalogRelation = table_open(InheritsRelationId, AccessShareLock);
 
-	result = get_partition_parent_worker(catalogRelation, relid);
+	result = get_partition_parent_worker(catalogRelation, relid,
+										 &detach_pending);
 
 	if (!OidIsValid(result))
 		elog(ERROR, "could not find tuple for parent of relation %u", relid);
 
+	if (detach_pending && !even_if_detached)
+		elog(ERROR, "relation %u has no parent because it's being detached",
+			 relid);
+
 	table_close(catalogRelation, AccessShareLock);
 
 	return result;
@@ -68,15 +78,20 @@ get_partition_parent(Oid relid)
  * get_partition_parent_worker
  *		Scan the pg_inherits relation to return the OID of the parent of the
  *		given relation
+ *
+ * If the partition is being detached, *detach_pending is set true (but the
+ * original parent is still returned.)
  */
 static Oid
-get_partition_parent_worker(Relation inhRel, Oid relid)
+get_partition_parent_worker(Relation inhRel, Oid relid, bool *detach_pending)
 {
 	SysScanDesc scan;
 	ScanKeyData key[2];
 	Oid			result = InvalidOid;
 	HeapTuple	tuple;
 
+	*detach_pending = false;
+
 	ScanKeyInit(&key[0],
 				Anum_pg_inherits_inhrelid,
 				BTEqualStrategyNumber, F_OIDEQ,
@@ -93,6 +108,9 @@ get_partition_parent_worker(Relation inhRel, Oid relid)
 	{
 		Form_pg_inherits form = (Form_pg_inherits) GETSTRUCT(tuple);
 
+		/* Let caller know of partition being detached */
+		if (form->inhdetachpending)
+			*detach_pending = true;
 		result = form->inhparent;
 	}
 
@@ -134,10 +152,14 @@ static void
 get_partition_ancestors_worker(Relation inhRel, Oid relid, List **ancestors)
 {
 	Oid			parentOid;
+	bool		detach_pending;
 
-	/* Recursion ends at the topmost level, ie., when there's no parent */
-	parentOid = get_partition_parent_worker(inhRel, relid);
-	if (parentOid == InvalidOid)
+	/*
+	 * Recursion ends at the topmost level, ie., when there's no parent; also
+	 * when the partition is being detached.
+	 */
+	parentOid = get_partition_parent_worker(inhRel, relid, &detach_pending);
+	if (parentOid == InvalidOid || detach_pending)
 		return;
 
 	*ancestors = lappend_oid(*ancestors, parentOid);
@@ -170,7 +192,7 @@ index_get_partition(Relation partition, Oid indexId)
 		ReleaseSysCache(tup);
 		if (!ispartition)
 			continue;
-		if (get_partition_parent(partIdx) == indexId)
+		if (get_partition_parent(partIdx, false) == indexId)
 		{
 			list_free(idxlist);
 			return partIdx;