Skip to main content
Version: 6.0-SNAPSHOT
Enterprise Edition
This feature requires Hazelcast Enterprise, a licensed edition of the Hazelcast Platform.

Client Authorization

To protect your members from a malicious client, you can allow them to identify clients and restrict their permissions to access data in data structures or use features such as user code deployment.

The Authentication overview describes how authentication is used for verifying credentials, and roles mapping. This section describes how the assigned role names are used to map permissions to clients.

Hazelcast client authorization is configured by a client permission policy. Hazelcast has a default permission policy implementation that uses permission configurations defined in the Hazelcast security configuration. Default policy permission checks are made against instance types (map, queue, etc.), instance names, instance actions (put, read, remove, add, etc.), the client endpoint address (ClusterEndpointPrincipal), and client roles (ClusterRolePrincipal).

The default permission policy allows you to use comma separated names in the principal attribute configuration.

note

Unless part of the role name, don't include spaces when adding names to the principal attribute.

Hazelcast doesn't automatically remove spaces in role names. If you include spaces that aren't part of the name, permission isn't granted to the intended role.

For example, if you configure permissions for the admin and devel roles using principal=" admin ,devel", the admin role isn't granted the permission.

You can define the instance and principal names as wildcards using the "*" character. For more information, see Using Wildcards.

The endpoint names can use range characters "-" and "*" as described in the Interfaces section.

xml
<hazelcast>
    ...
    <security enabled="true">
        <client-permissions>
            <!-- Principals 'admin' and 'root' from endpoint '127.0.0.1' have all permissions. -->
            <all-permissions principal="admin,root">
                <endpoints>
                    <endpoint>127.0.0.1</endpoint>
                </endpoints>
            </all-permissions>

            <!-- Principals named 'dev' from all endpoints have 'create', 'destroy',
            'put', 'read' permissions for map named 'myMap'. -->
            <map-permission name="myMap" principal="dev">
                <actions>
                    <action>create</action>
                    <action>destroy</action>
                    <action>put</action>
                    <action>read</action>
                </actions>
            </map-permission>

            <!-- All principals from endpoints '127.0.0.1' or matching to '10.10.*.*'
            have 'put', 'read', 'remove' permissions for map
            whose name matches to 'com.foo.entity.*'. -->
            <map-permission name="com.foo.entity.*">
                <endpoints>
                    <endpoint>10.10.*.*</endpoint>
                    <endpoint>127.0.0.1</endpoint>
                </endpoints>
                <actions>
                    <action>put</action>
                    <action>read</action>
                    <action>remove</action>
                </actions>
            </map-permission>

            <!-- Principals named 'dev' from endpoints matching either
            '192.168.1.1-100' or '192.168.2.*'
            have 'create', 'add', 'remove' permissions for all queues. -->
            <queue-permission name="*" principal="dev">
                <endpoints>
                    <endpoint>192.168.1.1-100</endpoint>
                    <endpoint>192.168.2.*</endpoint>
                </endpoints>
                <actions>
                    <action>create</action>
                    <action>add</action>
                    <action>remove</action>
                </actions>
            </queue-permission>

           <!-- All principals from all endpoints have transaction permission.-->
           <transaction-permission />
       </client-permissions>
    </security>
    ...
</hazelcast>

You can also define your own policy by implementing com.hazelcast.security.IPermissionPolicy.

java
package com.hazelcast.security;
/**
 * IPermissionPolicy is used to determine any Subject's
 * permissions to perform a security sensitive Hazelcast operation.
 *
 */
public interface IPermissionPolicy {
  void configure( SecurityConfig securityConfig, Properties properties );

  PermissionCollection getPermissions( Subject subject,
                                       Class<? extends Permission> type );

  void destroy();
}

Permission policy implementations can access client-permissions that are in the configuration by using SecurityConfig.getClientPermissionConfigs() when Hazelcast calls the configure(SecurityConfig securityConfig, Properties properties) method.

The IPermissionPolicy.getPermissions(Subject subject, Class<? extends Permission> type) method is used to determine a client request that has been granted permission to perform a security-sensitive operation.

The permission policy should return a PermissionCollection containing permissions of the given type for the given Subject. The Hazelcast access controller calls PermissionCollection.implies(Permission) on returning PermissionCollection and decides whether the current Subject has permission to access the requested resources.

Permissions​

All Hazelcast clients authorized to connect to a cluster are able to receive proxy information for data structures present on the cluster. This proxy only provides the client with the data structure's type and name. Permissions for other actions related to these data structures can be configured as required.

The following is the list of client permissions that can be configured on the member:

note

In YAML configuration, permission types use one of two formats:

  • Single object format (used by all, config, and transaction): These singleton permissions can only have one configuration entry, so they are defined directly as an object without the - list indicator.
  • Array format (used by all other permission types): These permissions support multiple entries with different name attributes or configurations, so each entry begins with a - to indicate it is a list item.

For example, all: principal: admin (single object) versus map: - name: myMap (array).

If you see an error like expected type: JSONArray, found: JSONObject, check whether you are using the correct format for the permission type.

All permissions​

<all-permissions> grants clients access to all data and features. The client specified in the following code example is used by Hazelcast Management Center when it connects to clusters. To learn more about this client, see Cluster connections.

xml
<all-permissions principal="principal">
    <endpoints>
        ...
    </endpoints>
</all-permissions>

Management Permission​

<management-permission> defines which client principals/endpoints are allowed to perform management tasks.

warning

For Hazelcast Management Center to work properly, management-permission is not sufficient. all-permissions is required.

xml
<management-permission principal="mcadmin">
    <endpoints>
        ...
    </endpoints>
</management-permission>

Map permission​

Actions: all, create, destroy, index, intercept, listen, lock, put, read, remove, projection, aggregate.

xml
<map-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</map-permission>

Queue permission​

Actions: add, all, create, destroy, listen, read, remove.

xml
<queue-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</queue-permission>

MultiMap permission​

Actions: all, create, destroy, listen, lock, put, read, remove.

xml
<multimap-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
     </actions>
</multimap-permission>

Replicated map permission​

Actions: all, create, destroy, index, intercept, listen, lock, put, read, remove.

xml
<replicatedmap-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
     </actions>
</replicatedmap-permission>

Topic permission​

Actions: create, destroy, listen, publish.

xml
<topic-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</topic-permission>

Reliable topic permission​

Actions: create, destroy, listen, publish.

xml
<reliable-topic-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</reliable-topic-permission>

List permission​

Actions: add, all, create, destroy, listen, read, remove.

xml
<list-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</list-permission>

Set permission​

Actions: add, all, create, destroy, listen, read, remove.

xml
<set-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</set-permission>

Ringbuffer permission​

Actions: add, put, read, create, destroy.

xml
<ring-buffer-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</ring-buffer-permission>

Lock permission​

Actions: all, create, destroy, lock, read.

xml
<lock-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</lock-permission>

AtomicLong permission​

Actions: all, create, destroy, modify, read.

xml
<atomic-long-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</atomic-long-permission>

AtomicReference permission​

Actions: all, create, destroy, modify, read.

xml
<atomic-reference-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</atomic-reference-permission>

CountDownLatch permission​

Actions: all, create, destroy, modify, read.

xml
<countdown-latch-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</countdown-latch-permission>

FlakeIdGenerator permission​

Actions: all, create, destroy, modify, read.

xml
<flake-id-generator-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</flake-id-generator-permission>

Semaphore permission​

Actions: all, acquire, create, destroy, read, release.

xml
<semaphore-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</semaphore-permission>

Executor Service Permission​

Actions: all, create, destroy

xml
<executor-service-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</executor-service-permission>

Durable executor service permission​

Actions: all, create, destroy.

xml
<durable-executor-service-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</durable-executor-service-permission>

Scheduled executor service permission​

Actions: all, create, destroy, read, modify.

xml
<scheduled-executor-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</scheduled-executor-permission>

Cardinality estimator permission​

Actions: all, create, destroy, read, modify.

xml
<cardinality-estimator-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</cardinality-estimator-permission>

PN counter permission​

Actions: all, create, destroy, read, modify.

xml
<pn-counter-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</pn-counter-permission>

Transaction permission​

xml
<transaction-permission principal="principal">
    <endpoints>
        ...
    </endpoints>
</transaction-permission>

Cache permission​

Actions: all, create, destroy, listen, put, read, remove.

xml
<cache-permission name="/hz/cache-name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</cache-permission>
note

The name provided in cache-permission must be the Hazelcast distributed object name that corresponds to the Cache as described in the JCache - Hazelcast Instance Integration section.

Vector collection permission (Beta)​

Actions: all, create, destroy, put, read, remove, optimize.

xml
<vector-collection-permission name="name" principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</vector-collection-permission>

User Code Deployment permission​

Actions: all, deploy/

caution

User Code Deployment has been deprecated and will be removed in the next major version. To continue deploying your user code after this time, Community Edition users can either upgrade to Enterprise Edition, or add their resources to the Hazelcast member class paths. Hazelcast recommends that Enterprise Edition users migrate their user code to use User Code Namespaces. For further information on migrating from User Code Deployment to User Code Namespaces, see Migrate from User Code Deployment.

xml
<user-code-deployment-permission principal="principal">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</user-code-deployment-permission>

If you have migrated to User Code Namespaces, use the following permissions:

xml
<user-code-namespace-permission principal="principal" name="my_namespace">
    <endpoints>
        ...
    </endpoints>
    <actions>
        ...
    </actions>
</user-code-namespace-permission>

Configuration permission​

This permission defines which client principals/endpoints are allowed to add data structure configurations at runtime.

xml
<config-permission principal="principal">
    <endpoints>
        <endpoint>...</endpoint>
    </endpoints>
</config-permission>

Job permission​

Actions:

  • submit: Submit a new job, without uploading resources.

  • cancel: Cancel a running job.

  • read: Get or list information about a job (by ID or name) such as job configuration, job status, and submission time.

    warning

    When you query a streaming source with SQL, Hazelcast runs that query as a job. As a result, clients with the read permission for jobs can see the SQL query and any parameters.

  • restart: Suspend and resume a running job.

  • export-snapshot: Export or read snapshots.

  • add-resources: Upload resources and classes as well as jobs to members.

    warning

    Hazelcast can't check permissions in code that's uploaded with a job, If you enable this permission, clients can upload custom code that ignores any configured permissions.

  • all: Enable all actions.

All actions for job permissions also enable the read action. For example, if you enable the create action, the read action is automatically enabled as well.

xml
<job-permission principal="principal">
    <actions>
        <action>...</action>
    </actions>
</job-permission>

Connector permission​

You can give permissions to the following connectors:

  • File
  • Socket

Actions:

  • read: Read data from sources.
  • write: Write data to sinks.
  • all: Enable all actions.
File Connector
xml
<!-- It is currently only possible to give access to a whole directory, not to a single file. -->
<connector-permission name="file:directory_name">
    <actions>
        <action>...</action>
    </actions>
</connector-permission>
Socket Connector
xml
<connector-permission name="socket:host:port">
    <actions>
        <action>...</action>
    </actions>
</connector-permission>
warning

To protect external systems from being reached by external connectors (JDBC, Mongo, S3, …​), use other means than Hazelcast client permissions. Traditionally, this is done by enabling authentication on the external system and/or setting up firewall rules.

SQL permission​

You can give clients permission to use the following SQL statements:

Actions:

  • create: Use the CREATE MAPPING statement to create new mappings or replace existing ones.
  • destroy: Use the DROP MAPPING statement to delete mappings.
  • create-index: Use the CREATE INDEX statement to create a new index for a map.
  • create-view: Use the CREATE VIEW statement to create new views or replace existing ones.
  • drop-view: Use the DROP VIEW statement to delete an existing view.
  • create-dataconnection: Use the CREATE DATA CONNECTION statement to create new data connections or replace existing ones.
  • drop-dataconnection: Use the DROP DATA CONNECTION statement to delete data connections.
  • view-dataconnection: Use the SHOW RESOURCES statement to view the resources and data types accessible via data connections.
  • all: Enable all actions.

To apply permissions to certain mappings or data connections, provide their names in the name attribute. Or, you can apply permissions to all mappings and data connections using the * wildcard.

Apply permissions to a mapping
xml
<sql-permission name="mapping_name">
  <actions>
    <action>create</action>
    <action>destroy</action>
  </actions>
</sql-permission>
Apply permissions to all mappings
xml
<sql-permission name="*">
  <actions>
    <action>create</action>
    <action>destroy</action>
  </actions>
</sql-permission>
Apply permissions to a data connection
xml
<sql-permission name="data_connection_name">
  <actions>
    <action>drop-dataconnection</action>
    <action>view-dataconnection</action>
  </actions>
</sql-permission>

Handling permissions when a new member joins​

By default, the set of permissions defined in the leader member of a cluster is distributed to new members that join, overriding their own permission configurations, if any. However, you can configure a new member to join but keep its own set of permissions and even send these to the existing members in the cluster. This can be done dynamically, without needing to restart the cluster, using either one of the following configuration options:

  • the on-join-operation configuration attribute
  • the setOnJoinPermissionOperation() method

You can use these options to choose whether a new member joining to a cluster will apply the client permissions stored in its own configuration, or use the ones defined in the cluster. The behaviors that you specify with the configuration are RECEIVE, SEND and NONE, which are described after the examples below.

The following examples show how to use both approaches:

Declarative configuration:

xml
<hazelcast>
    ...
    <security enabled="true">
        <client-permissions on-join-operation="SEND">
            <!-- ... -->
        </client-permissions>
    </security>
    ...
</hazelcast>

Programmatic configuration:

java
Config config = new Config();
config.getSecurityConfig()
    .setEnabled(true)
    .setOnJoinPermissionOperation(OnJoinPermissionOperationName.SEND);

The behaviors are explained below:

  • RECEIVE: Applies the permissions from the leader member in the cluster before joining. This is the default value.

  • SEND: Doesn't apply the permissions from the leader member before joining. If security is enabled, then it refreshes or replaces the cluster wide permissions with the ones in the new member after it joins. This option is suitable for scenarios where you need to replace the cluster-wide permissions without restarting the cluster.

  • NONE: Neither applies pre-join permissions nor sends the local permissions to the other members. It means that the new member doesn't send its own permission definitions to the cluster, but keeps them when it joins. However, after the join, when you update the permissions in the other cluster members, those updates are also sent to the new member. Therefore, this option is suitable for scenarios where you need to elevate privileges temporarily on a single member (preferably a lite member) for a limited time period. The clients which need to use these temporary permissions have to access the cluster through this single new member, meaning that you need to configure the SINGLE_MEMBER cluster routing mode for such clients.

    Note that the create and destroy permissions won't work when using the NONE option, since the distributed objects need to be created/destroyed on all the members.

    The following is an example for a scenario where NONE is used:

    java
    // temporary member, in the below case a lite member
    Config config = new Config().setLiteMember(true);
    PermissionConfig allPermission = new PermissionConfig(PermissionType.ALL, "*", null);
    config.getSecurityConfig()
      .setEnabled(true)
      .setOnJoinPermissionOperation(OnJoinPermissionOperationName.NONE)
      .addClientPermissionConfig(allPermission);
    HazelcastInstance hzLite = Hazelcast.newHazelcastInstance(config);

    // temporary client connecting only to the lite member
    String memberAddr = ...;
    ClientConfig clientConfig = new ClientConfig();
    clientConfig.getNetworkConfig().setSmartRouting(false)
      .addAddress(memberAddr);
    HazelcastInstance client = HazelcastClient.newHazelcastClient(clientConfig);

    // do operations with escalated privileges:
    client.getMap("protectedConfig").put("master.resolution", "1920");

    // shutdown the client and lite member
    client.shutdown();
    hzLite.shutdown();

Deny permissions​

Hazelcast employs additive access control as its default security mechanism. When a client connects to a security-enabled cluster, it is initially granted no permissions. As a result, access to protected resources is inherently denied unless explicit permissions are configured and granted to specific roles.

The additive access control approach has limited expression capabilities and is not well-suited for configurations involving simple exclusions. For example, it's challenging to allow access to all maps except the one named "private".

To address this limitation, Hazelcast introduces the concept of deny permissions (or deny rules).

Within the permission configuration, there is a boolean flag called deny that enables permission subtraction.

xml
<hazelcast>
    ...
    <security enabled="true">
        <client-permissions>
            <!-- Grant access to all maps. -->
            <map-permission name="*">
                <actions>
                    <action>all</action>
                </actions>
            </map-permission>
            <!-- Deny access to the "private" map. -->
            <map-permission name="private" deny="true">
                <actions>
                    <action>all</action>
                </actions>
            </map-permission>
       </client-permissions>
    </security>
    ...
</hazelcast>

Priority of grant and deny permissions​

By default, when a permission is both granted and denied, the denial takes precedence. In other words, if conflicting permissions exist, denial prevails.

In certain scenarios, it might be beneficial to reverse this behavior and give higher priority to permission grants. Hazelcast supports this by introducing the boolean flag priority-grant, which can be set to true.

xml
<hazelcast>
    ...
    <security enabled="true">
        <client-permissions  priority-grant="true">
        ...
       </client-permissions>
    </security>
    ...
</hazelcast>

Permission evaluation table​

The table below illustrates how permission evaluation changes when priority-grant is configured:

Permission Implicationpriority-grant=false (default)priority-grant=true
No Grant or Deny Implication❌✅
Implication from Grant only✅✅
Implication from Deny only❌❌
Both Grant and Deny Imply❌✅
Edit this page