Graph
The graph service provides the Graph API which is a RESTful web API used to access Infinite Scale resources. It is inspired by the Microsoft Graph API and can be used by clients or other services or extensions. Visit the Libre Graph API for a detailed specification of the API implemented by the graph service.
Sequence Diagram
The following image gives an overview of the scenario when a client requests to list available spaces the user has access to. To do so, the client is directed with his request automatically via the proxy service to the graph service.
Users and Groups API
The graph service provides endpoints for querying users and groups. It features two different backend implementations:
ldap
: This is currently the default backend. It queries user and group information from an LDAP server. Depending on the configuration, it can also be used to manage (create, update, delete) users and groups provided by an LDAP server.cs3
: This backend queries users and groups using the CS3 identity APIs as implemented by theusers
andgroups
service. This backend is currently still experimental and only implements a subset of the Libre Graph API. It should not be used in production.
LDAP Configuration
The LDAP backend is configured using a set of environment variables. A detailed list of all the
available configuration options can be found in the documentation.
The LDAP related options are prefixed with OCIS_LDAP_
(or GRAPH_LDAP_
for settings specific to graph service).
Read-Only Access to Existing LDAP Servers
To connect the graph service to an existing LDAP server, set OCIS_LDAP_SERVER_WRITE_ENABLED
to
false
to prevent the graph service from sending write operations to the LDAP server. Also set the
various OCIS_LDAP_*
environment variables to match the configuration of the LDAP server you are connecting
to. An example configuration for connecting oCIS to an instance of Microsoft Active Directory is
available here.
Using a Write Enabled LDAP Server
To use the graph service for managing (create, update, delete) users and groups, a write enabled LDAP
server is required. In the default configuration, the graph service will use the simple LDAP server
that is bundled with oCIS in the idm
service which provides all the required features.
It is also possible to setup up an external LDAP server with write access for use with oCIS. It is
recommend to use OpenLDAP for this. The LDAP server needs to fulfill a couple of requirements with
respect to the available schema:
- The LDAP server must provide the
inetOrgPerson
object class for users and thegroupOfNames
object class for groups. - The graph service maintains a few additional attributes for users and groups that are not available in the standard LDAP schema. An schema file, ready to use with OpenLDAP, defining those additional attributes is available here.
Query Filters Provided by the Graph API
Some API endpoints provided by the graph service allow to specify query filters. The filter syntax is based on the OData Specification. See the Libre Graph API for examples on the filters supported when querying users.
Caching
The graph
service can use a configured store via GRAPH_CACHE_STORE
. Possible stores are:
memory
: Basic in-memory store and the default.redis-sentinel
: Stores data in a configured Redis Sentinel cluster.nats-js-kv
: Stores data using key-value-store feature of nats jetstreamnoop
: Stores nothing. Useful for testing. Not recommended in production environments.
Other store types may work but are not supported currently.
Note: The service can only be scaled if not using memory
store and the stores are configured identically over all instances!
Note that if you have used one of the deprecated stores, you should reconfigure to one of the supported ones as the deprecated stores will be removed in a later version.
Store specific notes:
- When using
redis-sentinel
, the Redis master to use is configured via e.g.OCIS_CACHE_STORE_NODES
in the form of<sentinel-host>:<sentinel-port>/<redis-master>
like10.10.0.200:26379/mymaster
. - When using
nats-js-kv
it is recommended to setOCIS_CACHE_STORE_NODES
to the same value asOCIS_EVENTS_ENDPOINT
. That way the cache uses the same nats instance as the event bus. - When using the
nats-js-kv
store, it is possible to setOCIS_CACHE_DISABLE_PERSISTENCE
to instruct nats to not persist cache data on disc.
Keycloak Configuration For The Personal Data Export
If Keycloak is used for authentication, GDPR regulations require to add all personal identifiable information that Keycloak has about the user to the personal data export. To do this, the following environment variables must be set:
OCIS_KEYCLOAK_BASE_PATH
- The URL to the keycloak instance.OCIS_KEYCLOAK_CLIENT_ID
- The client ID of the client that is used to authenticate with keycloak, this client has to be able to list users and get the credential data.OCIS_KEYCLOAK_CLIENT_SECRET
- The client secret of the client that is used to authenticate with keycloak.OCIS_KEYCLOAK_CLIENT_REALM
- The realm the client is defined in.OCIS_KEYCLOAK_USER_REALM
- The realm the oCIS users are defined in.OCIS_KEYCLOAK_INSECURE_SKIP_VERIFY
- If set to true, the TLS certificate of the keycloak instance is not verified.
For more details see the User-Triggered GDPR Report in the ocis admin documentation.
Keycloak Client Configuration
The client that is used to authenticate with keycloak has to be able to list users and get the credential data. To do this, the following roles have to be assigned to the client and they have to be about the realm that contains the oCIS users:
view-users
view-identity-providers
view-realm
view-clients
view-events
view-authorization
Note that these roles are only available to assign if the client is in the master
realm.
Translations
The graph
service has embedded translations sourced via transifex to provide a basic set of translated languages. These embedded translations are available for all deployment scenarios. In addition, the service supports custom translations, though it is currently not possible to just add custom translations to embedded ones. If custom translations are configured, the embedded ones are not used. To configure custom translations, the GRAPH_TRANSLATION_PATH
environment variable needs to point to a base folder that will contain the translation files. This path must be available from all instances of the graph service, a shared storage is recommended. Translation files must be of type .po or .mo. For each language, the filename needs to be graph.po
(or graph.mo
) and stored in a folder structure defining the language code. In general the path/name pattern for a translation file needs to be:
{GRAPH_TRANSLATION_PATH}/{language-code}/LC_MESSAGES/graph.po
The language code pattern is composed of language[_territory]
where language
is the base language and _territory
is optional and defines a country.
For example, for the language de
, one needs to place the corresponding translation files to {GRAPH_TRANSLATION_PATH}/de_DE/LC_MESSAGES/graph.po
.
Important: For the time being, the embedded ownCloud Web frontend only supports the main language code but does not handle any territory. When strings are available in the language code language_territory
, the web frontend does not see it as it only requests language
. In consequence, any translations made must exist in the requested language
to avoid a fallback to the default.
Translation Rules
- If a requested language code is not available, the service tries to fall back to the base language if available. For example, if the requested language-code
de_DE
is not available, the service tries to fall back to translations in thede
folder. - If the base language
de
is also not available, the service falls back to the system's default English (en
), which is the source of the texts provided by the code.
Default Language
The default language can be defined via the OCIS_DEFAULT_LANGUAGE
environment variable. See the settings
service for a detailed description.
Unified Role Management
Unified Roles are roles granted a user for sharing and can be enabled or disabled. A CLI command is provided to list existing roles and their state among other data.
{{< hint info >}} Note that a disabled role does not lose previously assigned permissions. It only means that the role is not available for new assignments. {{< /hint >}}
The following roles are enabled by default:
UnifiedRoleViewerID
UnifiedRoleSpaceViewer
UnifiedRoleEditor
UnifiedRoleSpaceEditor
UnifiedRoleFileEditor
UnifiedRoleEditorLite
UnifiedRoleManager
The following role is disabled by default:
UnifiedRoleSecureViewer
To enable disabled roles like the UnifiedRoleSecureViewer
, you must provide the UID(s) by one of the following methods:
- Using the
GRAPH_AVAILABLE_ROLES
environment variable. - Setting the
available_roles
configuration value.
The following CLI command simplifies the process of finding out which UID belongs to which role:
ocis graph list-unified-roles
The output of this command includes the following information for each role:
UID
The unique identifier of the role.Enabled
Whether the role is enabled or not.Description
A short description of the role.Condition
Allowed resource actions
Example output (shortned)
+--------------------------------------+----------+--------------------------------+--------------------------------+------------------------------------------+
| UID | ENABLED | DESCRIPTION | CONDITION | ALLOWED RESOURCE ACTIONS |
+--------------------------------------+----------+--------------------------------+--------------------------------+------------------------------------------+
| a8d5fe5e-96e3-418d-825b-534dbdf22b99 | enabled | View and download. | exists @Resource.Root | libre.graph/driveItem/path/read |
| | | | | libre.graph/driveItem/quota/read |
| | | | | libre.graph/driveItem/content/read |
| | | | | libre.graph/driveItem/permissions/read |
| | | | | libre.graph/driveItem/children/read |
| | | | | libre.graph/driveItem/deleted/read |
| | | | | libre.graph/driveItem/basic/read |
+--------------------------------------+----------+--------------------------------+--------------------------------+------------------------------------------+