WebSocket Notification Service#
Added in version 1.1.
ESS provides a secure implementation of a WebSockets notification protocol. The ESS WebSocket service publishes asynchronous notifications upon create/update/delete operations on a Resource, including a Container.
Disambiguation
The ESS WebSocket service is not an implementation of the draft Solid WebSockets API Spec (version solid-0.1).
The ESS WebSocket service is based on an earlier version of the Solid Notifications Protocol.
Client Subscriptions#
To receive notifications, a client subscribes to a particular Resource or Resources. A client can subscribe to a Resource that does not yet exist.
Using the @inrupt/solid-client-notifications
library,
applications can subscribe to WebSocket notifications.
Important
To subscribe to a resource, the authenticated
user must have Read
access to the resource.
import { getDefaultSession, fetch } from "@inrupt/solid-client-authn-browser";
import {
WebsocketNotification,
} from "@inrupt/solid-client-notifications";
const resourceURL = ...; // URL of the Resource to which to subscribe
// ... authentication logic has been omitted
const websocket = new WebsocketNotification(
resourceURL,
{ fetch: fetch } // Authenticated fetch of a user with Read access
);
websocket.on("message", console.log);
websocket.connect();
For more information, see:
The process for establishing a WebSocket connection starts at the
Notification Gateway endpoint for ESS (e.g.,
https://notification.<DOMAIN>/
). This endpoint allows clients
to identify the protocols they support along with any particular
features that are required.
To determine your Notification Gateway endpoint, see /.well-known/solid.
Clients can
POST
the following JSON document to the ESS Notification Gateway; no authentication is required:{ "protocols": ["ws"] }
The response JSON contains the WebSocket endpoint that clients can access for login; for example:
{ "endpoint": "https://websocket.<DOMAIN>/", "features":[], "protocol":"ws" }
Clients can use the returned
endpoint
to subscribe to a resource. Authentication is required.Important
To subscribe to a resource, the authenticated user must have
Read
access to the resource.To subscribe, clients can
POST
(using standard Solid authentication headers) a JSON document that specifies the Resource to which to subscribe:{ "topic": "<URL of the resource to which to subscribe>" }
The response JSON returns the
endpoint
to establish the actual WebSocket connection:{ "endpoint": "wss://websocket.<DOMAIN>/?s=<session token>", "protocol": "ws", "subprotocol": "solid-0.2" }
Using the returned
endpoint
andsubprotocol
, a client can connect to the WebSocket service, passing in the entireendpoint
string andsubprotocol
string into the JavaScript WebSocket constructor):var ws = new WebSocket("<endpoint>", "<subprotocol>");
You must specify the subprotocol. Otherwise, the connection is dropped.
Note
The subprotocol string is subject to change as the ESS WebSocket notification protocol undergoes the Solid panel process.
Once the connection is established, all the standard WebSocket methods can be used from the JavaScript API.
When subscribed to a particular Resource, the client sees all create/update/delete events for that particular Resource.
If a client subscribes to a Container, the client receives events for the Container as well as for the Container’s child Resource; specifically, if a child Resource is created or deleted but not if a child Resource is modified.
Clients should expect long-lived WebSocket connections to be terminated
at any point in time. It is important that client code watch for
WebSocket.onclose
events and handle them by restarting the
connection flow.
Notification Event#
The ESS WebSocket service publishes each notification as a W3C ActivityStream message, serialized as JSON-LD:
{
"@context": [
"https://www.w3.org/ns/activitystreams",
{
"state": {
"@id": "http://www.w3.org/2011/http-headers#etag"
}
}
],
"id": "urn:uuid:<random>",
"type": [
"http://www.w3.org/ns/prov#Activity",
"<Action>"
],
"object": {
"id": "<resource URL>",
"type": [
"<LDP Resource Type>",
...
"http://www.w3.org/ns/ldp#Resource"
]
},
"published": "<YYYY-MM-DDThh:mm:ss.sTZD>"
}
Field |
Description |
||||
---|---|---|---|---|---|
|
String that contains an identifier for the event. |
||||
|
An array identifying the event type: [
"http://www.w3.org/ns/prov#Activity",
"<Action>"
]
Where
|
||||
|
The resource object:
|
||||
|
The date and time the event is published. |
||||
|
An array containing the JSON-LD contexts for the notification event itself. |
Configuration#
As part of the installation process, Inrupt provides base Kustomize overlays and associated files that require deployment-specific configuration inputs.
The following configuration options are available for the service and may be set as part of updating the inputs for your deployment. The Inrupt-provided base Kustomize overlays may be using updated configuration values that differ from the default values.
Required#
- INRUPT_WEBSOCKET_BASE_URI#
The URI of the service; e.g.,
wss://websocket.example/
- QUARKUS_GRPC_CLIENTS_AUTHZ_HOST#
The host where the ESS Authorization Service is running.
- SMALLRYE_JWT_SIGN_KEY_LOCATION#
The path to the signing key used for negotiating subscriptions.
Kafka Configuration#
Tip
See also ESS’ Kafka Configuration.
- INRUPT_KAFKA_SOLIDRESOURCE_CIPHER_PASSWORD#
The symmetric key to use when decrypting messages (see
MP_MESSAGING_INCOMING_SOLIDRESOURCE_VALUE_DESERIALIZER
).Warning
Set to a strong password.
- KAFKA_BOOTSTRAP_SERVERS#
Default:
localhost:9092
Comma-delimited list of Kafka broker servers for use by ESS services, including this service.
Setting
KAFKA_BOOTSTRAP_SERVERS
configures ESS to use the same Kafka instance(s) for all its Kafka message channels (e.g.,solidresource
andauditv1out
message channels). This service uses thesolidresource
andauditv1out
channels.Note
Inrupt-provided overlays default to using
KAFKA_BOOTSTRAP_SERVERS
.To use a different Kafka instance for the
solidresource
andauditv1out
channels, use specific message channel configuration.See also ESS’ Kafka Configuration.
- MP_MESSAGING_INCOMING_SOLIDRESOURCE_SSL_TRUSTSTORE_LOCATION#
The location of the Kafka connection truststore
- MP_MESSAGING_INCOMING_SOLIDRESOURCE_SSL_TRUSTSTORE_PASSWORD#
The password for the Kafka connection truststore.
- MP_MESSAGING_INCOMING_SOLIDRESOURCE_VALUE_DESERIALIZER#
Default:
org.apache.kafka.common.serialization.StringDeserializer
The deserializer to use to consume Kafka messages sent by the Pod services.
Supported values are:
org.apache.kafka.common.serialization.StringDeserializer
Set to this value to consume notification messages sent by services that use
org.apache.kafka.common.serialization.StringSerializer
as theirMP_MESSAGING_OUTGOING_SOLIDRESOURCE_VALUE_SERIALIZER
com.inrupt.components.kafka.encryption.DecryptMessageDeserializer
Set to this value to consume encrypted notification messages sent by services that use
com.inrupt.components.kafka.encryption.EncryptMessageSerializer
as theirMP_MESSAGING_OUTGOING_SOLIDRESOURCE_VALUE_SERIALIZER
- MP_MESSAGING_OUTGOING_AUDITV1OUT_BOOTSTRAP_SERVERS#
Default:
localhost:9092
Comma-delimited list of Kafka broker servers used for the outgoing audit v1 messages.
These messages are sent over the
auditv1out
message channel.Note
To configure ESS to use the same Kafka instances for all its Kafka message channels, use
KAFKA_BOOTSTRAP_SERVERS
option instead. Inrupt-provided overlays default to usingKAFKA_BOOTSTRAP_SERVERS
.
Optional#
Configuration Logging#
Starting in 2.2, ESS services log their startup configuration.
- INRUPT_LOGGING_CONFIGURATION_PREFIX_ALLOW#
Default: inrupt,smallrye.jwt.sign.key.location
A comma-separated list of configuration property prefixes (case-sensitive) that determine which configurations are logged:
If the list is empty, NO configuration property is logged.
If a configuration property starts with a listed prefix (case-sensitive), the configuration property and its value are logged unless the configuration also matches a prefix in
INRUPT_LOGGING_CONFIGURATION_PREFIX_DENY
(which acts as a filter onINRUPT_LOGGING_CONFIGURATION_PREFIX_ALLOW
list).As such, if the configuration matches prefix in both
INRUPT_LOGGING_CONFIGURATION_PREFIX_ALLOW
andINRUPT_LOGGING_CONFIGURATION_PREFIX_DENY
, theINRUPT_LOGGING_CONFIGURATION_PREFIX_DENY
takes precedence and the configuration is not logged. For example, ifinrupt.
is an allow prefix, butinrupt.kafka.
is a deny prefix, all configurations that start withinrupt.kafka.
are excluded from the logs.
When specifying the prefixes, you can specify the prefixes using one of two formats:
using dot notation (e.g.,
inrupt.foobar.
), orusing the MicroProfile Config environmental variables conversion value (e.g.,
INRUPT_FOOBAR_
).
Warning
Use the same format for both
INRUPT_LOGGING_CONFIGURATION_PREFIX_ALLOW
andINRUPT_LOGGING_CONFIGURATION_PREFIX_DENY
.For example, if you change the format of
INRUPT_LOGGING_CONFIGURATION_PREFIX_ALLOW
, change the format ofINRUPT_LOGGING_CONFIGURATION_PREFIX_DENY
as well.Tip
To avoid allowing more than desired configurations, specify as much of the prefix as possible. If the prefix specifies the complete prefix term, include the term delineator. For example:
If using dot-notation, if you want to match configuration properties of the form
foobar.<xxxx>...
, specifyfoobar.
(including the dot.
) instead of, for example,foo
orfoobar
.If using converted form, if you want to match configuration properties of the form
FOOBAR_<XXXX>...
, specifyFOOBAR_
(including the underscore_
) instead of, for example,FOO
orFOOBAR
.
Added in version 2.2.0.
- INRUPT_LOGGING_CONFIGURATION_PREFIX_DENY#
Default: inrupt.kafka
A comma-separated list of configuration name prefixes (case-sensitive) that determines which configurations (that would otherwise match the
INRUPT_LOGGING_CONFIGURATION_PREFIX_ALLOW
) are not logged. That is,INRUPT_LOGGING_CONFIGURATION_PREFIX_DENY
acts as a filter onINRUPT_LOGGING_CONFIGURATION_PREFIX_ALLOW
. For example:If
foobar.
is an allowed prefix, to suppressfoobar.private.<anything>
, you can specifyfoobar.private.
to the deny list.If
foobar.
is not an allowed prefix, no property starting withfoobar.
is logged. As such, you do not need to specifyfoobar.private
to the deny list.
When specifying the prefixes, you can specify the prefixes using one of two formats:
using dot notation (e.g.,
inrupt.foobar.
), orusing the MicroProfile Config environmental variables conversion value (e.g.,
INRUPT_FOOBAR_
).
Warning
Use the same format for both
INRUPT_LOGGING_CONFIGURATION_PREFIX_ALLOW
andINRUPT_LOGGING_CONFIGURATION_PREFIX_DENY
.For example, if you change the format of
INRUPT_LOGGING_CONFIGURATION_PREFIX_ALLOW
, change the format ofINRUPT_LOGGING_CONFIGURATION_PREFIX_DENY
as well.Added in version 2.2.0.
Logging Redaction#
- INRUPT_LOGGING_REDACTION_{NAME}_ACTION#
Default:
REPLACE
Type of the redaction to perform. Supported values are:
Action
Description
REPLACE
Default. Replaces the matching text with a specified replacement.
PLAIN
Leaves the matching field unprocessed. Only available if the redaction target is a field (i.e.,
INRUPT_LOGGING_REDACTION_{NAME}_FIELD
).DROP
Suppresses the matching field. Only available if the redaction target is a field (i.e.,
INRUPT_LOGGING_REDACTION_{NAME}_FIELD
).PRIORITIZE
Changes the log level of the matching message.
SHA256
Replaces the matching text with its hash.
If the action is
REPLACE
(default), see alsoINRUPT_LOGGING_REDACTION_{NAME}_REPLACEMENT
.If the action is to
PRIORITIZE
, see alsoINRUPT_LOGGING_REDACTION_{NAME}_LEVEL
.
For more information on log redaction, see Logging Redaction.
Added in version 2.2.0.
- INRUPT_LOGGING_REDACTION_{NAME}_ENABLED#
Default:
true
A boolean that determines whether the redaction configurations with the specified
INRUPT_LOGGING_REDACTION_{NAME}_
prefix is enabled.For more information on log redaction, see Logging Redaction.
Added in version 2.2.0.
- INRUPT_LOGGING_REDACTION_{NAME}_EXCEPTION#
Fully qualified name of the exception class to match in the log messages (includes inner exception). Configure to target an exception message class.
For more information on log redaction, see Logging Redaction.
Added in version 2.2.0.
- INRUPT_LOGGING_REDACTION_{NAME}_FIELD#
Exact name of the field to match in the log messages. Configure to target a specific log message field for redaction.
For more information on log redaction, see Logging Redaction.
Added in version 2.2.0.
- INRUPT_LOGGING_REDACTION_{NAME}_LEVEL#
Default:
DEBUG
A new log level to use for the log message if the
INRUPT_LOGGING_REDACTION_{NAME}_ACTION
isPRIORITIZE
.Added in version 2.2.0.
- INRUPT_LOGGING_REDACTION_{NAME}_PATTERN#
A regex (see Java regex pattern) to match in the log messages. Configure to target log message text that matches a specified pattern.
For more information on log redaction, see Logging Redaction.
Added in version 2.2.0.
- INRUPT_LOGGING_REDACTION_{NAME}_REPLACEMENT#
Default:
[REDACTED]
Replacement text to use if the
INRUPT_LOGGING_REDACTION_{NAME}_ACTION
isREPLACE
.If unspecified, defaults to
[REDACTED]
.For more information on log redaction, see Logging Redaction.
Added in version 2.2.0.
Application-Defined Metadata Propagation#
- INRUPT_AUDIT_PRODUCER_REQUEST_METADATA_ALLOW#
A comma-separated list [1] of application-defined properties that can be included in the associated audit events (unless specified in the corresponding
INRUPT_AUDIT_PRODUCER_REQUEST_METADATA_DENY
).This configuration is case-sensitive to the propagated properties in the baggage.
Tip
To include a propagated property that was added via the
INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_ALLOW
configuration, ensure that the cases of these properties match.See:
Manage Application-Defined Metadata Propagation to configure.
Application-Defined Metadata for more information.
Added in version 2.2.0.
- INRUPT_AUDIT_PRODUCER_REQUEST_METADATA_DENY#
A comma-separated list [1] of application-defined properties to exclude from the associated audit messages. This setting takes precedence over
INRUPT_AUDIT_PRODUCER_REQUEST_METADATA_ALLOW
.This configuration is case-sensitive to the propagated properties in the baggage.
Tip
To exclude a propagated property that was added via the
INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_ALLOW
configuration, ensure that the cases of these properties match.See:
Manage Application-Defined Metadata Propagation to configure.
Application-Defined Metadata for more information.
Added in version 2.2.0.
- INRUPT_LOGGING_REQUEST_METADATA_ALLOW#
A comma-separated list [1] of application-defined properties that can be included in the associated log messages (unless specified in the corresponding
INRUPT_LOGGING_REQUEST_METADATA_DENY
).This configuration is case-sensitive to the propagated properties in the baggage.
Tip
To include a propagated property that was added via the
INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_ALLOW
configuration, ensure that the cases of these properties match.See:
Manage Application-Defined Metadata Propagation to configure.
Application-Defined Metadata for more information.
Added in version 2.2.0.
- INRUPT_LOGGING_REQUEST_METADATA_DENY#
A comma-separated list [1] of application-defined properties to exclude from the associated log messages. This setting takes precedence over
INRUPT_LOGGING_REQUEST_METADATA_ALLOW
.This configuration is case-sensitive to the propagated properties in the baggage.
Tip
To exclude a propagated property that was added via the
INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_ALLOW
configuration, ensure that the cases of these properties match.See:
Manage Application-Defined Metadata Propagation to configure.
Application-Defined Metadata for more information.
Added in version 2.2.0.
- INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_ALLOW#
A comma-separated list [1] of non-baggage request headers to add to the baggage (unless specified in the corresponding
INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_DENY
); i.e., include these non-baggage request headers as application-defined properties.The configuration is case-insensitive; i.e., the listed headers do not need to match the case of the client request headers. For example, a list that includes
x-correlation-id
can matchx-correlation-id
header,X-CoRrElAtIoN-Id
header, etc.See:
Manage Application-Defined Metadata Propagation to configure.
Application-Defined Metadata for more information.
Added in version 2.2.0.
- INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_DENY#
A comma-separated list [1] of non-baggage request headers to exclude from being added to the baggage; i.e., excludes these headers as application-defined properties. This setting takes precedence over
INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_ALLOW
.The configuration is case-insensitive; i.e., the listed headers do not need to match the case of the client request headers. For example, a list that includes
x-correlation-id
can match (and exclude)x-correlation-id
header,X-CoRrElAtIoN-Id
header, etc.See:
Manage Application-Defined Metadata Propagation to configure.
Application-Defined Metadata for more information.
Added in version 2.2.0.
- INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_OVERRIDES#
A flag that determines ESS behavior when metadata property is defined both as a header and as a baggage entry:
If
true
, ESS updates/overrides the baggage entry with the header value.If
false
(the default), ESS keeps the baggage entry.
For details, Duplicate Property Definition.
Added in version 2.2.0.
- INRUPT_REQUEST_METADATA_REFLECTOR_HEADER_ALLOW#
A comma-separated list [1] of application-defined properties that can return as response headers (unless specified in the corresponding
INRUPT_REQUEST_METADATA_REFLECTOR_HEADER_DENY
).This configuration is case-sensitive to the propagated properties in the baggage.
Tip
To return a propagated property that was added via the
INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_ALLOW
configuration, ensure that the cases of these properties match.You may need to update
QUARKUS_HTTP_CORS_EXPOSED_HEADERS
to extend the list of CORS-safelisted response headers.
See:
Manage Application-Defined Metadata Propagation to configure.
Application-Defined Metadata for more information.
Added in version 2.2.0.
- INRUPT_REQUEST_METADATA_REFLECTOR_HEADER_DENY#
A comma-separated list [1] of application-defined properties to exclude from returning as response headers. This setting takes precedence over
INRUPT_REQUEST_METADATA_REFLECTOR_HEADER_ALLOW
.This configuration is case-sensitive to the propagated properties in the baggage.
Tip
To exclude a propagated property that was added via the
INRUPT_REQUEST_METADATA_PROPAGATOR_HEADER_ALLOW
configuration, ensure that the cases of these properties match.See:
Manage Application-Defined Metadata Propagation to configure.
Application-Defined Metadata for more information.
Added in version 2.2.0.
General#
- INRUPT_JWT_ALLOWED_SIGNATURE_ALGORITHMS#
Default:
ES256
,RS256
A comma-separated list that specifies the allowed encryption algorithms used to sign ID tokens.
- INRUPT_JWT_ISSUER_ALLOW_LIST#
A comma-separated list of trusted Solid-OIDC issuers (i.e., identity providers).
If unset, the service accepts all Solid-OIDC issuers with the exception of those in the
INRUPT_JWT_ISSUER_DENY_LIST
.If set, the service accepts only those Solid-OIDC issuers in the list with the following exception:
If an issuer is in both
INRUPT_JWT_ISSUER_ALLOW_LIST
andINRUPT_JWT_ISSUER_DENY_LIST
, theINRUPT_JWT_ISSUER_DENY_LIST
supersedes theINRUPT_JWT_ISSUER_ALLOW_LIST
and the issuer is not accepted by ESS.
See also
INRUPT_JWT_ISSUER_DENY_LIST
.
- INRUPT_JWT_ISSUER_DENY_LIST#
A comma-separated list of disallowed Solid-OIDC issuers.
If unset, the service accepts all Solid-OIDC issuers unless
INRUPT_JWT_ISSUER_ALLOW_LIST
is set, in which case, the service only accepts those in theINRUPT_JWT_ISSUER_ALLOW_LIST
.If set, the service disallows the Solid-OIDC issuers in the list. If
INRUPT_JWT_ISSUER_ALLOW_LIST
is also set, issuers not in theINRUPT_JWT_ISSUER_ALLOW_LIST
are also disallowed.
- QUARKUS_LOG_LEVEL#
Default:
INFO
Logging level.
Additional Information#
See also https://quarkus.io/guides/all-config.