docs: remove citus.shard_max_size GUC

develop/api_guc.rst

@@ -263,18 +263,15 @@ Sets the commit protocol to use when performing COPY on a hash distributed table
 
 * **1pc:** The transactions in which COPY is performed on the shard placements are committed in a single round. Data may be lost if a commit fails after COPY succeeds on all placements (rare).
 
+.. _citus.shard_count:
+
 citus.shard_count (integer)
 $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
 
 Sets the shard count for hash-partitioned tables and defaults to 32. This value is used by
 the :ref:`create_distributed_table <create_distributed_table>` UDF when creating
 hash-partitioned tables. This parameter can be set at run-time and is effective on the coordinator. 
 
-citus.shard_max_size (integer)
-$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
-
-Sets the maximum size to which a shard will grow before it gets split and defaults to 1GB. When the source file's size (which is used for staging) for one shard exceeds this configuration value, the database ensures that a new shard gets created. This parameter can be set at run-time and is effective on the coordinator.
-
 .. _replicate_reference_tables_on_activate:
 
 citus.replicate_reference_tables_on_activate (boolean)

develop/reference_ddl.rst

@@ -71,9 +71,9 @@ distribution column and create the worker shards.
 
 This function informs Citus that the github_events table should be distributed
 on the repo_id column (by hashing the column value). The function also creates
-shards on the worker nodes using the citus.shard_count configuration value.
+shards on the worker nodes using the :ref:`citus.shard_count <citus.shard_count>` configuration value.
 
-This example would create a total of citus.shard_count number of shards where
+This example would create a total of :ref:`citus.shard_count <citus.shard_count>` number of shards where
 each shard owns a portion of a hash token space. Once the shards are created,
 this function saves all distributed metadata on the coordinator.