Merge pull request #31 from nl2go/shard-support-doc

adding documentation following shard support

README.md

@@ -20,23 +20,23 @@ Server configuration directory.
 
 Server log level.
 
-    clickhouse_profiles: 
+    clickhouse_profiles:
       default:
         max_memory_usage: 10000000000
         use_uncompressed_cache: 0
         load_balancing: "random"
 
 A settings profile is a collection of [settings](https://clickhouse.tech/docs/en/operations/settings/settings-profiles/) grouped under the same name.
 
-    clickhouse_profiles: 
+    clickhouse_profiles:
       default:
         password: "qwerty"
         profile: "default"
         quota: "default"
 
 The `users` section of the `user.xml` configuration file contains user  [settings](https://clickhouse.tech/docs/en/operations/settings/settings-users/).
 
-    clickhouse_profiles: 
+    clickhouse_profiles:
       default:
         intervals:
           - duration: 3600
@@ -48,14 +48,14 @@ The `users` section of the `user.xml` configuration file contains user  [setting
 
 Quota [settings](https://clickhouse.tech/docs/en/operations/settings/settings-profiles/) is for limiting resource usage over a period of time or track the use of resources.
 
-    clickhouse_databases: 
+    clickhouse_databases:
       - these_metrics
       - other_metrics
       - imported_data
 
 **clickhouse_databases** is a list of the databases you want to create in ClickHouse.
 
-    clickhouse_zookeeper_hosts: 
+    clickhouse_zookeeper_hosts:
       - 1.2.3.4
       - 1.2.3.5
 
@@ -69,9 +69,37 @@ Quota [settings](https://clickhouse.tech/docs/en/operations/settings/settings-pr
 For debugging purposes it is better to start just with one host and then add more hosts.
 
 **clickhouse_zookeeper_credentials** in format ``username:password`` is used to authenticate against Zookeeper.
+
 **clickhouse_path**: Path to the directory where ClickHouse will store data
+
 **clickhouse_log_directory**: Path to the error and normal logs.
 
+**clickhouse_clusters**: is the list of clusters and also contains the shards that are part of each cluster.
+
+    clickhouse_clusters:
+      distributed_cluster_1:
+        shard1:
+          - { host: "host_1", port: 9000 }
+          - { host: "host_2", port: 9000 }
+        shard2:
+          - { host: "host_3", port: 9000 }
+          - { host: "host_4", port: 9000 }
+      distributed_cluster_2:
+        shard1:
+          - { host: "host_5", port: 9000 }
+          - { host: "host_6", port: 9000 }
+        shard2:
+          - { host: "host_7", port: 9000 }
+          - { host: "host_8", port: 9000 }
+
+For the file macros.xml, you need to fill the variables below:
+
+**clickhouse_cluster**: is the cluster name which is used for replication.
+
+**clickhouse_distributed_cluster**: is the distributed cluster name defined in **clickhouse_clusters**.
+
+**clickhouse_shard**: is the shard number (01, 02) that being part of the replication inside one shard.
+
 ## Dependencies
 
 None.

templates/macros.xml.j2

@@ -2,9 +2,9 @@
 <!-- {{ ansible_managed }} -->
 <yandex>
     <macros>
-        <cluster>{{ cluster }}</cluster>
-        <shard>{{ shard }}</shard>
+        <cluster>{{ clickhouse_cluster }}</cluster>
+        <shard>{{ clickhouse_shard }}</shard>
         <replica>{{ inventory_hostname }}</replica>
-        <distributed_clusters>{{ distributed_cluster }}</distributed_clusters>
+        <distributed_clusters>{{ clickhouse_distributed_cluster }}</distributed_clusters>
     </macros>
 </yandex>

vars/main.yml

@@ -1,3 +1,6 @@
 ---
 clickhouse_group: clickhouse
 clickhouse_user: clickhouse
+clickhouse_shard: 01
+clickhouse_cluster: clickhouse-cluster
+clickhouse_distributed_cluster: distributed_cluster