artem-budnikov commented on a change in pull request #8131:
URL: https://github.com/apache/ignite/pull/8131#discussion_r469093563
##########
File path: docs/_docs/thin-clients/java-thin-client.adoc
##########
@@ -170,19 +181,96 @@ NOTE: The `getAll()` method retrieves the results from
the cursor and closes it.
Read more about using `SqlFieldsQuery` and SQL API in the
link:SQL/sql-api[Using SQL API] section.
-== Handling Exceptions
+== Cluster API
-=== Handling Node Failures
+Cluster functionality (APIs for accessing cluster and nodes) is provided via
the `ClientCluster` interface. You can get an instance of `ClientCluster` from
`IgniteClient` as follows:
+[source, java]
+-------------------------------------------------------------------------------
+include::{sourceCodeFile}[tag=client-cluster,indent=0]
+-------------------------------------------------------------------------------
-When you provide the addresses of multiple nodes in the client configuration,
the client automatically switches to the next node if the current connection
fails and retries any ongoing operation.
+Through the `ClientCluster` interface you can:
+
+* Get or set cluster state
+* Get a list of all cluster members
+* Create logical Cluster Groups
-In the case of atomic operations, failover to another node is transparent to
the user. However, if you execute a scan query or a SELECT query, it may
return duplicate results. This can happen because queries return data in pages,
and if the node that the client is connected to goes down while the client
retrieves the pages, the client connects to another node and executes the query
again. To avoid this, you have to write some code in your application that
checks if the entries returned by the client are duplicated. Consider the
following code:
+== Cluster groups
Review comment:
I think this section should be a subsection of "Cluster API" and should
give an example of how to get a group of nodes for a specific purpose. Also,
"Cluster groups" is an incorrect term. I know it's used in the API itself, but
it should be "Node groups". In fact, you don't have to use the term, you can
simple say something like "here is how you can group nodes based on some
characteristic (server nodes, clients, attributes, etc.)." and give an example.
##########
File path: docs/_docs/thin-clients/java-thin-client.adoc
##########
@@ -170,19 +181,96 @@ NOTE: The `getAll()` method retrieves the results from
the cursor and closes it.
Read more about using `SqlFieldsQuery` and SQL API in the
link:SQL/sql-api[Using SQL API] section.
-== Handling Exceptions
+== Cluster API
-=== Handling Node Failures
+Cluster functionality (APIs for accessing cluster and nodes) is provided via
the `ClientCluster` interface. You can get an instance of `ClientCluster` from
`IgniteClient` as follows:
+[source, java]
+-------------------------------------------------------------------------------
+include::{sourceCodeFile}[tag=client-cluster,indent=0]
+-------------------------------------------------------------------------------
-When you provide the addresses of multiple nodes in the client configuration,
the client automatically switches to the next node if the current connection
fails and retries any ongoing operation.
+Through the `ClientCluster` interface you can:
+
+* Get or set cluster state
+* Get a list of all cluster members
+* Create logical Cluster Groups
-In the case of atomic operations, failover to another node is transparent to
the user. However, if you execute a scan query or a SELECT query, it may
return duplicate results. This can happen because queries return data in pages,
and if the node that the client is connected to goes down while the client
retrieves the pages, the client connects to another node and executes the query
again. To avoid this, you have to write some code in your application that
checks if the entries returned by the client are duplicated. Consider the
following code:
+== Cluster groups
+API to create logical groups of cluster nodes within your cluster is provided
by `ClientClusterGroup` interface. Instance of this interface can be obtained
from a parent `ClientClusterGroup` instance using node-filtering methods.
[source, java]
-------------------------------------------------------------------------------
-include::{sourceCodeFile}[tag=results-to-map,indent=0]
+include::{sourceCodeFile}[tag=client-cluster-groups,indent=0]
-------------------------------------------------------------------------------
+The `ClientCluster` instance (see link:#cluster-api[Cluster API]) it's a root
cluster group, which includes all nodes in the cluster.
+
+////
+*TODO: fix link*
+////
+
+Refer to link:clustering/cluster-groups[Cluster groups] to get more
information about logical nodes grouping.
Review comment:
Please change the link: as of now, the cluster groups page is located
here: distributed-computing/cluster-groups. I know this is not the best
location, I'll deal with it later ( I'm thinking of creating the "Cluster API"
section at the top level of the docs).
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
