With Manticore, write transactions (such as INSERT, REPLACE, DELETE, TRUNCATE, UPDATE, COMMIT) can be replicated to other cluster nodes before the transaction is fully applied on the current node. Currently, replication is supported for percolate, rt and distributed tables in Linux and macOS.
Native Windows binaries for Manticore do not support replication. We recommend installing Manticore via WSL (Windows Subsystem for Linux).
On macOS, replication has limited support and is recommended only for development purposes.
Manticore's replication is powered by the Galera library and boasts several impressive features:
- True multi-master: read and write to any node at any time.
- Virtually synchronous replication no slave lag and no data loss after a node crash.
- Hot standby: no downtime during failover (since there is no failover).
- Tightly coupled: all nodes hold the same state and no diverged data between nodes is allowed.
- Automatic node provisioning: no need to manually backup the database and restore it on a new node.
- Easy to use and deploy.
- Detection and automatic eviction of unreliable nodes.
- Certification-based replication.
To set up replication in Manticore Search:
- The data_dir option must be set in the "searchd" section of the configuration file. Replication is not supported in plain mode.
- A listen directive must be specified, containing an IP address accessible by other nodes, or a node_address with an accessible IP address.
- Optionally, you can set unique values for server_id on each cluster node. If no value is set, the node will attempt to use the MAC address or a random number to generate the
server_id.
If authentication and authorization is enabled, cluster operations require the replication permission. Grant it to the user that should own replication operations for the cluster:
GRANT replication ON 'posts' TO 'repl_user';
CREATE CLUSTER and JOIN CLUSTER can specify that user with '<user>' AS user. The user must exist with matching stored authentication data on the nodes that take part in the operation. Creating the same user name and password independently on each node is not enough, because the stored authentication data can differ. If you change auth data outside the daemon, run RELOAD AUTH on the affected nodes before using the cluster operation.
Later ALTER CLUSTER ... ADD, ALTER CLUSTER ... DROP, ALTER CLUSTER ... UPDATE nodes, and DELETE CLUSTER use the stored cluster user. When authentication is enabled, a successful JOIN CLUSTER replaces all local authentication data on the joining node with the donor cluster's authentication data.
If there is no replication listen directive set, Manticore will use the first two free ports in the range of 200 ports after the default protocol listening port for each created cluster. To set replication ports manually, the listen directive (of replication type) port range must be defined and the address/port range pairs must not intersect between different nodes on the same server. As a rule of thumb, the port range should specify at least two ports per cluster. When you define a replication listener with a port range (e.g., listen = 192.168.0.1:9320-9328:replication), Manticore doesn't immediately start listening on these ports. Instead, it will take random free ports from the specified range only when you start using replication.
A replication cluster is a group of nodes in which a write transaction is replicated. Replication is set up on a per-table basis, meaning that one table can only belong to one cluster. There is no limit on the number of tables that a cluster can have. All transactions such as INSERT, REPLACE, DELETE, TRUNCATE on any percolate or real-time table that belongs to a cluster are replicated to all the other nodes in that cluster. Distributed tables can also be part of the replication process. Replication is multi-master, so writes to any node or multiple nodes simultaneously will work just as well.
To create a cluster, you can typically use the command create cluster with CREATE CLUSTER <cluster name>, and to join a cluster, you can use join cluster with JOIN CLUSTER <cluster name> at 'host:port'. However, in some rare cases, you may want to fine-tune the behavior of CREATE/JOIN CLUSTER. The available options are:
This option specifies the name of the cluster. It should be unique among all the clusters in the system.
Note: The maximum allowable hostname length for the JOIN command is 253 characters. If you exceed this limit, searchd will generate an error.
The path option specifies the data directory for write-set cache replication and other cluster provider files. It does not change where replicated tables are stored. Incoming replicated tables are stored in their normal table directory under data_dir. This value should be unique among all the clusters in the system and should be specified as a relative path to the data_dir directory. By default, it is set to the value of data_dir.
Breaking change: Older versions also stored incoming replicated table files under the cluster path. If you used a custom cluster path, upgrade carefully because replicated tables received by older versions might need to be moved or re-synchronized into the normal data_dir/<table> layout.
The nodes option is a list of address:port pairs for all the nodes in the cluster, separated by commas. This list should be obtained using the node's API interface and can include the address of the current node as well. It is used to join the node to the cluster and to rejoin it after a restart.
The options option allows you to pass additional options directly to the Galera replication plugin, as described in the Galera Documentation Parameters
When working with a replication cluster, all write statements such as INSERT, REPLACE, DELETE, TRUNCATE, UPDATE that modify the content of a cluster's table must use thecluster_name:table_name expression instead of the table name. This ensures that the changes are propagated to all replicas in the cluster. If the correct expression is not used, an error will be triggered.
In the JSON interface, the cluster property must be set along with the table name for all write statements to a cluster's table. Failure to set the cluster property will result in an error.
The Auto ID for a table in a cluster should be valid as long as the server_id is correctly configured.
INSERT INTO posts:weekly_table VALUES ( 'iphone case' )
TRUNCATE TABLE click_query:weekly_table
UPDATE INTO posts:rt_tags SET tags=(101, 302, 304) WHERE MATCH ('use') AND id IN (1,101,201)
DELETE FROM clicks:rt WHERE MATCH ('dumy') AND gid>206
POST /insert -d '
{
"cluster":"posts",
"table":"weekly_table",
"doc":
{
"title" : "iphone case",
"price" : 19.85
}
}'
POST /delete -d '
{
"cluster":"posts",
"table": "weekly_table",
"id":1
}'
$table->addDocuments([
1, ['title' => 'iphone case', 'price' => 19.85]
]);
$table->deleteDocument(1);
indexApi.insert({"cluster":"posts","table":"weekly_table","doc":{"title":"iphone case","price":19.85}})
indexApi.delete({"cluster":"posts","table":"weekly_table","id":1})
await indexApi.insert({"cluster":"posts","table":"weekly_table","doc":{"title":"iphone case","price":19.85}})
await indexApi.delete({"cluster":"posts","table":"weekly_table","id":1})
res = await indexApi.insert({"cluster":"posts","table":"weekly_table","doc":{"title":"iphone case","price":19.85}});
res = await indexApi.delete({"cluster":"posts","table":"weekly_table","id":1});
InsertDocumentRequest newdoc = new InsertDocumentRequest();
HashMap<String,Object> doc = new HashMap<String,Object>(){{
put("title","Crossbody Bag with Tassel");
put("price",19.85);
}};
newdoc.table("weekly_table").cluster("posts").id(1L).setDoc(doc);
sqlresult = indexApi.insert(newdoc);
DeleteDocumentRequest deleteRequest = new DeleteDocumentRequest();
deleteRequest.table("weekly_table").cluster("posts").setId(1L);
indexApi.delete(deleteRequest);
Dictionary<string, Object> doc = new Dictionary<string, Object>();
doc.Add("title", "Crossbody Bag with Tassel");
doc.Add("price", 19.85);
InsertDocumentRequest newdoc = new InsertDocumentRequest(table: "weekly_table", cluster:posts, id: 1, doc: doc);
var sqlresult = indexApi.Insert(newdoc);
DeleteDocumentRequest deleteDocumentRequest = new DeleteDocumentRequest(table: "weekly_table", cluster: "posts", id: 1);
indexApi.Delete(deleteDocumentRequest);
let mut doc = HashMap::new();
doc.insert("title".to_string(), serde_json::json!("Crossbody Bag with Tassel"));
doc.insert("price".to_string(), serde_json::json!(19.85));
let insert_req = InsertDocumentRequest {
table: serde_json::json!("weekly_table"),
doc: serde_json::json!(doc),
cluster: serde_json::json!("posts"),
id: serde_json::json!(1),
};
let insert_res = index_api.insert(insert_req).await;
let delete_req = DeleteDocumentRequest {
table: serde_json::json!("weekly_table"),
cluster: serde_json::json!("posts"),
id: serde_json::json!(1),
};
index_api.delete(delete_req).await;
Read statements such as SELECT, CALL PQ, DESCRIBE can either use regular table names that are not prepended with a cluster name, or they can use the cluster_name:table_nameformat. If the latter is used, the cluster_name component is ignored.
When using the HTTP endpoint json/search, the cluster property can be specified if desired, but it can also be omitted.
SELECT * FROM weekly_table
CALL PQ('posts:weekly_table', 'document is here')
POST /search -d '
{
"cluster":"posts",
"table":"weekly_table",
"query":{"match":{"title":"keyword"}}
}'
POST /search -d '
{
"table":"weekly_table",
"query":{"match":{"title":"keyword"}}
}'
SET CLUSTER click_query GLOBAL 'pc.bootstrap' = 1
POST /cli -d "
SET CLUSTER click_query GLOBAL 'pc.bootstrap' = 1
"
It's possible for replicated nodes to diverge from one another, leading to a state where all nodes are labeled as non-primary. This can occur as a result of a network split between nodes, a cluster crash, or if the replication plugin experiences an exception when determining the primary component. In such a scenario, it's necessary to select a node and promote it to the role of primary component.
To identify the node that needs to be promoted, you should compare the last_committed cluster status variable value on all nodes. If all the servers are currently running, there's no need to restart the cluster. Instead, you can simply promote the node with the highest last_committed value to the primary component using the SET statement (as demonstrated in the example).
The other nodes will then reconnect to the primary component and resynchronize their data based on this node.
SET CLUSTER posts GLOBAL 'pc.bootstrap' = 1
POST /cli -d "
SET CLUSTER posts GLOBAL 'pc.bootstrap' = 1
"
To use replication, you need to define one listen port for SphinxAPI protocol and one listen for replication address and port range in the configuration file. Also, specify the data_dir folder to store incoming replicated tables.
searchd {
listen = 9312
listen = 192.168.1.101:9360-9370:replication
data_dir = /var/lib/manticore/
...
}
To replicate tables, you must create a cluster on the server that has the local tables to be replicated.
POST /cli -d "
CREATE CLUSTER posts
"
$params = [
'cluster' => 'posts'
]
];
$response = $client->cluster()->create($params);
utilsApi.sql('CREATE CLUSTER posts')
await utilsApi.sql('CREATE CLUSTER posts')
res = await utilsApi.sql('CREATE CLUSTER posts');
utilsApi.sql("CREATE CLUSTER posts");
utilsApi.Sql("CREATE CLUSTER posts");
utils_api.sql("CREATE CLUSTER posts", Some(true)).await;
Add these local tables to the cluster
ALTER CLUSTER posts ADD pq_title
ALTER CLUSTER posts ADD pq_clicks
POST /cli -d "
ALTER CLUSTER posts ADD pq_title
"
POST /cli -d "
ALTER CLUSTER posts ADD pq_clicks
"
$params = [
'cluster' => 'posts',
'body' => [
'operation' => 'add',
'table' => 'pq_title'
]
];
$response = $client->cluster()->alter($params);
$params = [
'cluster' => 'posts',
'body' => [
'operation' => 'add',
'table' => 'pq_clicks'
]
];
$response = $client->cluster()->alter($params);
utilsApi.sql('ALTER CLUSTER posts ADD pq_title')
utilsApi.sql('ALTER CLUSTER posts ADD pq_clicks')
await utilsApi.sql('ALTER CLUSTER posts ADD pq_title')
await utilsApi.sql('ALTER CLUSTER posts ADD pq_clicks')
res = await utilsApi.sql('ALTER CLUSTER posts ADD pq_title');
res = await utilsApi.sql('ALTER CLUSTER posts ADD pq_clicks');
utilsApi.sql("ALTER CLUSTER posts ADD pq_title");
utilsApi.sql("ALTER CLUSTER posts ADD pq_clicks");
utilsApi.Sql("ALTER CLUSTER posts ADD pq_title");
utilsApi.Sql("ALTER CLUSTER posts ADD pq_clicks");
utils_api.sql("ALTER CLUSTER posts ADD pq_title", Some(true)).await;
utils_api.sql("ALTER CLUSTER posts ADD pq_clicks", Some(true)).await;
All other nodes that wish to receive a replica of the cluster's tables should join the cluster as follows:
JOIN CLUSTER posts AT '192.168.1.101:9312'
POST /cli -d "
JOIN CLUSTER posts AT '192.168.1.101:9312'
"
$params = [
'cluster' => 'posts',
'body' => [
'192.168.1.101:9312'
]
];
$response = $client->cluster->join($params);
utilsApi.sql('JOIN CLUSTER posts AT \'192.168.1.101:9312\'')
await utilsApi.sql('JOIN CLUSTER posts AT \'192.168.1.101:9312\'')
res = await utilsApi.sql('JOIN CLUSTER posts AT \'192.168.1.101:9312\'');
utilsApi.sql("JOIN CLUSTER posts AT '192.168.1.101:9312'");
utilsApi.Sql("JOIN CLUSTER posts AT '192.168.1.101:9312'");
utils_api.sql("JOIN CLUSTER posts AT '192.168.1.101:9312'", Some(true)).await;
When running queries, prepend the table name with the cluster name posts: or use the cluster property for HTTP request object.
INSERT INTO posts:pq_title VALUES ( 3, 'test me' )
POST /insert -d '
{
"cluster":"posts",
"table":"pq_title",
"id": 3
"doc":
{
"title" : "test me"
}
}'
$table->addDocuments([
3, ['title' => 'test me']
]);
indexApi.insert({"cluster":"posts","table":"pq_title","id":3"doc":{"title":"test me"}})
await indexApi.insert({"cluster":"posts","table":"pq_title","id":3"doc":{"title":"test me"}})
res = await indexApi.insert({"cluster":"posts","table":"pq_title","id":3"doc":{"title":"test me"}});
InsertDocumentRequest newdoc = new InsertDocumentRequest();
HashMap<String,Object> doc = new HashMap<String,Object>(){{
put("title","test me");
}};
newdoc.table("pq_title").cluster("posts").id(3L).setDoc(doc);
sqlresult = indexApi.insert(newdoc);
Dictionary<string, Object> doc = new Dictionary<string, Object>();
doc.Add("title", "test me");
InsertDocumentRequest newdoc = new InsertDocumentRequest(table: "pq_title", cluster: "posts", id: 3, doc: doc);
var sqlresult = indexApi.Insert(newdoc);
let mut doc = HashMap::new();
doc.insert("title".to_string(), serde_json::json!("test me"));
let insert_req = InsertDocumentRequest {
table: serde_json::json!("pq_title"),
doc: serde_json::json!(doc),
cluster: serde_json::json!("posts"),
id: serde_json::json!(3),
};
let insert_res = index_api.insert(insert_req).await;
All queries that modify tables in the cluster are now replicated to all nodes in the cluster.