Devolutions Gateway Cookbook

February 17, 2026 · View on GitHub

Developer-oriented cookbook for testing purposes.

RDP routing

Devolutions Gateway can redirect RDP traffic authorized by a JWT (Json Web Token) signed (JWS) and optionally encrypted (JWE).

The key used to sign must be known by the Devolutions Gateway. This key is provided through the ProvisionerPublicKeyFile option in the configuration file. The provisioner can then use its private key to sign a JWT and authorize RDP routing.

Similarly, The key used for token decryption is provided through the DelegationPrivateKeyFile option. The public counterpart of the delegation key must then be used for token encryption.

JWT structure and claims

Devolutions Gateway is expecting signed claims using JWS (Json Web Signature) as a compact JWT. Use of RSASSA-PKCS-v1_5 using SHA-256 (RS256) is recommended.

Required claims:

  • dst_hst (String): target RDP host
  • jet_cm (String): identity connection mode used for Jet association This must be set to fwd.
  • jet_ap (string): application protocol used over Jet transport. This must be set to rdp.
  • exp (Integer): a UNIX timestamp for "expiration"
  • nbf (Integer): a UNIX timestamp for "not before"

This token may be encrypted and wrapped inside another JWT using JWE (Json Web Encryption), in compact form as well. Use of RSAES OAEP using SHA-256 and MGF1 with SHA-256 (RSA-OAEP-256) and AES GCM using 256-bit key (A256GCM) is recommended.

Token generation utilities

JWT generation should be facilitated by a provisioner (such as Devolutions Server or Devolutions Password Hub). However, you can easily generate a JWT for testing purposes by using CLI tools provided in /tools folder.

tokengen

A native CLI. No binary provided; you will need a Rust toolchain to build yourself. See Install Rust.

$ cargo build --package tokengen --release

The binary is produced inside a target/release folder.

Example:

$ ./tokengen --provisioner-key /path/to/provisioner/private/key.pem forward --dst-hst 192.168.122.70 --jet-ap rdp

Inject token in RDP connection using MSTSC

  1. Open MSTSC

  2. Enter a JET address in the "computer" field

  3. Press the "Save As..." button under the "Connection settings" panel to save ".RDP" file to you PC

  4. Open saved ".RDP" file with a text editor

  5. Append string "pcb:s:" to the end of the file (e.g: pcb:s:eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9.eyJleHAiOj...)

  6. Save file

  7. In MSTSC press "Open..." and select your edited file

  8. Done. You can start the connection

Inject token in RDP connection using FreeRdp

Using FreeRDP, token can be provided using /pcb argument with xfreerdp. (e.g: /pcb:eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9.eyJleHAiOj...)

WebSocket-to-TCP tunnel using jetsocat

Our CLI-based toolkit jetsocat can be used to create a network tunnel bridging a WebSocket connection with a TCP connection. This is useful when debugging the Devolutions Gateway service.

This section describes how to create the following tunnel:

(jetsocat as client) <--WS/TCP/IP--> (Devolutions Gateway) <--TCP/IP--> (jetsocat as server)

Devolutions Gateway service

Configure and start the Devolutions Gateway service (see top-level README.md file).

Server TCP endpoint

Start jetsocat to act as our server endpoint:

cargo run -p jetsocat -- forward tcp-listen://127.0.0.1:9999 -

Received payload will be printed to the standard output.

WebSocket client

Generate a session forwarding token using tokengen (or alternatively, the New-DGatewayToken cmdlet):

cargo run --manifest-path=./tools/tokengen/Cargo.toml --provisioner-key <path/to/provisioner.key> forward --dst-hst 127.0.0.1:9999 --jet-aid 123e4567-e89b-12d3-a456-426614174000

New-DGatewayToken -Type ASSOCIATION -DestinationHost 127.0.0.1:9999 -ApplicationProtocol unknown -AssociationId 123e4567-e89b-12d3-a456-426614174000

Finally, run the following command to connect to the Devolutions Gateway service and open a WebSocket-to-TCP tunnel:

cargo run -p jetsocat -- forward - "ws://127.0.0.1:7171/jet/fwd/tcp/123e4567-e89b-12d3-a456-426614174000?token=<TOKEN>"

Try entering text and see it printed on the other side.

Standalone web application custom authentication

This section demonstrates how to use curl to test the /jet/webapp/app-token and /jet/webapp/session-token endpoints.

The standalone web application must be enabled and configured to use the custom authentication mode.

"WebApp": {
  "Enabled": true,
  "Authentication": "Custom"
}

A users.txt file is expected as well. For instance, with a user named David protected by the password abc:

David:$argon2id$v=19$m=16,t=2,p=1$U0tDR3NSSjlBaVJMRmV0Tg\$4KRKy3UsOganH/qTYVvOQg

It’s possible to retrieve a web application token using the POST /jet/webapp/app-token endpoint.

If the Authorization header is absent of the request, the server responds with a challenge:

$ curl -v http://127.0.0.1:7171/jet/webapp/app-token --json '{ "content_type": "WEBAPP", "subject": "David" }'
*   Trying 127.0.0.1:7171...
* Connected to 127.0.0.1 (127.0.0.1) port 7171
> POST /jet/webapp/app-token HTTP/1.1
> Host: 127.0.0.1:7171
> User-Agent: curl/8.5.0
> Content-Type: application/json
> Accept: application/json
> Content-Length: 48
>
< HTTP/1.1 401 Unauthorized
< www-authenticate: Basic realm="DGW Custom Auth", charset="UTF-8"
< access-control-allow-origin: *
< vary: origin
< vary: access-control-request-method
< vary: access-control-request-headers
< content-length: 0
< date: Fri, 22 Dec 2023 16:34:12 GMT
<
* Connection #0 to host 127.0.0.1 left intact

Notice the WWW-Authenticate header which advertises the configured authentication mode.

By requesting again with an appropriate Authorization header, a token is returned:

$ curl -v http://127.0.0.1:7171/jet/webapp/app-token --json '{ "content_type": "WEBAPP", "subject": "David" }' -H "Authorization: Basic RGF2aWQ6YWJj"
*   Trying 127.0.0.1:7171...
* Connected to 127.0.0.1 (127.0.0.1) port 7171
> POST /jet/webapp/app-token HTTP/1.1
> Host: 127.0.0.1:7171
> User-Agent: curl/8.5.0
> Authorization: Basic RGF2aWQ6YWJj
> Content-Type: application/json
> Accept: application/json
> Content-Length: 48
>
< HTTP/1.1 200 OK
< content-type: text/plain; charset=utf-8
< cache-control: no-cache, no-store
< content-length: 548
< access-control-allow-origin: *
< vary: origin
< vary: access-control-request-method
< vary: access-control-request-headers
< date: Fri, 22 Dec 2023 16:34:46 GMT
<
* Connection #0 to host 127.0.0.1 left intact
eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCIsImN0eSI6IldFQkFQUCJ9.eyJqdGkiOiIyODU5NjZhZi04M2VlLTRlNTEtYWYwOS01YWMwZTNjMzQyOTEiLCJpYXQiOjE3MDMyNjI4ODYsIm5iZiI6MTcwMzI2Mjg4NiwiZXhwIjoxNzAzMjkxNjg2LCJzdWIiOiJEYXZpZCJ9.ZO-bbuJpnoOMChbMEHsLj8gIXpcflJQ7DMIS4wo2dgEK4xnCxEJ4AdXVquYnZmGgg7-L1bhgKRi5EM35QFoYrnQDkMfSb6cVROGdp9Lg1-AgGA94Tw8Btq2bWXBJGES67cNFkdN-HJ07ixWKqpRz0wA4yZjn_8Z5B5K_S2_BP7IxfO7ckV_NqQzpaa94oH8XrdX_7dXwG6m-bXkNLOvAzyXHXFQkpb7l9-_CabJ6ZlJpdHcHJ4Tekx1_cHUW7haSyTd1Dp_VWIlnKhaqOcN3BRJ0aW9QaxR7JgSU1k9NWuZL3S5Au_SXUiYrOk2TdNkGDBptImkQhlSim6P4_OXacA

It’s then possible to retrieve a session token:

$ curl -v http://127.0.0.1:7171/jet/webapp/session-token --json '{ "content_type": "ASSOCIATION", "protocol": "rdp", "destination": "tcp://localhost:8888", "lifetime": 60, "session_id": "123e4567-e89b-12d3-a456-426614174000" }' -H "Authorization: Bearer eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCIsImN0eSI6IldFQkFQUCJ9.eyJqdGkiOiIyODU5NjZhZi04M2VlLTRlNTEtYWYwOS01YWMwZTNjMzQyOTEiLCJpYXQiOjE3MDMyNjI4ODYsIm5iZiI6MTcwMzI2Mjg4NiwiZXhwIjoxNzAzMjkxNjg2LCJzdWIiOiJEYXZpZCJ9.ZO-bbuJpnoOMChbMEHsLj8gIXpcflJQ7DMIS4wo2dgEK4xnCxEJ4AdXVquYnZmGgg7-L1bhgKRi5EM35QFoYrnQDkMfSb6cVROGdp9Lg1-AgGA94Tw8Btq2bWXBJGES67cNFkdN-HJ07ixWKqpRz0wA4yZjn_8Z5B5K_S2_BP7IxfO7ckV_NqQzpaa94oH8XrdX_7dXwG6m-bXkNLOvAzyXHXFQkpb7l9-_CabJ6ZlJpdHcHJ4Tekx1_cHUW7haSyTd1Dp_VWIlnKhaqOcN3BRJ0aW9QaxR7JgSU1k9NWuZL3S5Au_SXUiYrOk2TdNkGDBptImkQhlSim6P4_OXacA"
*   Trying 127.0.0.1:7171...
* Connected to 127.0.0.1 (127.0.0.1) port 7171
> POST /jet/webapp/session-token HTTP/1.1
> Host: 127.0.0.1:7171
> User-Agent: curl/8.5.0
> Authorization: Bearer eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCIsImN0eSI6IldFQkFQUCJ9.eyJqdGkiOiIyODU5NjZhZi04M2VlLTRlNTEtYWYwOS01YWMwZTNjMzQyOTEiLCJpYXQiOjE3MDMyNjI4ODYsIm5iZiI6MTcwMzI2Mjg4NiwiZXhwIjoxNzAzMjkxNjg2LCJzdWIiOiJEYXZpZCJ9.ZO-bbuJpnoOMChbMEHsLj8gIXpcflJQ7DMIS4wo2dgEK4xnCxEJ4AdXVquYnZmGgg7-L1bhgKRi5EM35QFoYrnQDkMfSb6cVROGdp9Lg1-AgGA94Tw8Btq2bWXBJGES67cNFkdN-HJ07ixWKqpRz0wA4yZjn_8Z5B5K_S2_BP7IxfO7ckV_NqQzpaa94oH8XrdX_7dXwG6m-bXkNLOvAzyXHXFQkpb7l9-_CabJ6ZlJpdHcHJ4Tekx1_cHUW7haSyTd1Dp_VWIlnKhaqOcN3BRJ0aW9QaxR7JgSU1k9NWuZL3S5Au_SXUiYrOk2TdNkGDBptImkQhlSim6P4_OXacA
> Content-Type: application/json
> Accept: application/json
> Content-Length: 161
>
< HTTP/1.1 200 OK
< content-type: text/plain; charset=utf-8
< cache-control: no-cache, no-store
< content-length: 762
< access-control-allow-origin: *
< vary: origin
< vary: access-control-request-method
< vary: access-control-request-headers
< date: Fri, 22 Dec 2023 16:35:51 GMT
<
* Connection #0 to host 127.0.0.1 left intact
eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCIsImN0eSI6IkFTU09DSUFUSU9OIn0.eyJkc3RfYWx0IjpbXSwiZHN0X2hzdCI6InRjcDovL2xvY2FsaG9zdDo4ODg4IiwiZXhwIjoxNzAzMjYzMDExLCJpYXQiOjE3MDMyNjI5NTEsImpldF9haWQiOiIxMjNlNDU2Ny1lODliLTEyZDMtYTQ1Ni00MjY2MTQxNzQwMDAiLCJqZXRfYXAiOiJyZHAiLCJqZXRfY20iOiJmd2QiLCJqZXRfZmx0IjpmYWxzZSwiamV0X3JlYyI6ZmFsc2UsImpldF90dGwiOjAsImp0aSI6ImMyZjAzMmU4LWNlZGMtNDk5Zi05ODYyLWExZWFlNjU5NGNiNCIsIm5iZiI6MTcwMzI2Mjk1MX0.WRwnQR-o6UNvIDCiskvOPiQ5XStriaGl4c4UfhZPdZY9hSN4nLajP_inWjbVR8V8h-WcuWZEo_p-s_0Ze6OnEpJ94HRw8e_ANEJ3JWCMrWB7MypWT4V3khPCk-SL29V-if2VUpwPq6Oc9ugpatCxHAJRcUD4FYxr1cy85jU__E3DwOceqGL1OUStfPVw5zqZvJQmZ2ndNO8K_6NhfC2PRSwmMYPPR_vKDeBFShSFQSHCWv2-X3Og5Mjm6R7vyMbvfKY7fs2zRQxwZBoUEaLhEimhqeVcsDH3dF8deN5DbnQ1nq2Eu_eWoJ4y3tBmwaZPMvIDHPq3STZRgehFkY5pqw

Preflight API

Generate a scope token with the scope gateway.preflight or * using tokengen (or alternatively, the New-DGatewayToken cmdlet):

tokengen --provisioner-key <path/to/provisioner.key> scope 'gateway.preflight'

New-DGatewayToken -Type SCOPE -Scope 'gateway.preflight'

Perform preflight operations using curl:

$ curl "127.0.0.1:7171/jet/preflight?token=$(cargo run --manifest-path ./tools/tokengen/Cargo.toml '--' sign --provisioner-key ./config/provisioner.key scope "gateway.preflight")" \
  -X POST -H "Content-Type: application/json" \
  --data '[
    {"id": "a86ae982-e4be-4f84-8ff2-893d66df9bdd", "kind": "get-version"},
    {"id": "ef1a3ae9-e55d-48b8-92b0-ae67c29b2e4e", "kind": "provision-token", "token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJqdGkiOiI1ZTNlODMzZi04NGM3LTQ1NDEtYjY3Ni1hY2MzMjk5ZTM5YjgifQ.1qECGlrW7y9HWFArc6GPHLGTOY7PhAvzKJ5XMRBg4k4"},
    {"id": "55821d24-d1df-481c-8b88-66c06f879835", "kind": "resolve-host", "host_to_resolve": "devolutions.net"},
    {"id": "8ec4ab6b-39a5-411d-b191-54df2d976820", "kind": "get-running-session-count"},
    {"id": "e02d8678-1bc8-4548-b555-548d663ffa1e", "kind": "unexisting-operation"}
  ]'

And here is how the response may look like:

[
  {"operation_id":"e02d8678-1bc8-4548-b555-548d663ffa1e","kind":"alert","alert_status":"unsupported-operation","alert_message":"unsupported operation: unexisting-operation"},
  {"operation_id":"a86ae982-e4be-4f84-8ff2-893d66df9bdd","kind":"version","version":"2025.1.4"},
  {"operation_id":"ef1a3ae9-e55d-48b8-92b0-ae67c29b2e4e","kind":"ack"}
  {"operation_id":"8ec4ab6b-39a5-411d-b191-54df2d976820","kind":"running-session-count","running_session_count":0},
  {"operation_id":"55821d24-d1df-481c-8b88-66c06f879835","kind":"resolved-host","resolved_host":"devolutions.net","resolved_addresses":["20.239.34.78"]}
]

Network monitoring API

Basic monitors can be set up to scan servers on an interval. Currently only ping is supported. Managing the configuration and storing the logs is expected to be done by an external server.

Set up a ping monitor for example.com which fires every 10 seconds and times out after 5 seconds.

curl -v http://127.0.0.1:7171/jet/net/monitor/config --json '{"monitors":[{"id":"monitor1","probe":"ping","address":"example.com","interval":10,"timeout":5}]}' -H "Authorization: Bearer $dgwkey"

The monitor will start immediately. Calling this API again will overwrite the configuration, stopping any monitors no longer present. A body is returned which may contain a list of monitors that could not be started due their type (set in the field probe) being unsupported.

Retrieve the logs:

curl -v http://127.0.0.1:7171/jet/net/monitor/log/drain -X POST -H "Authorization: Bearer $dgwkey"

The response will look similar to this:

{"entries":[{"monitor_id":"monitor1","request_start_time":"2025-08-22T17:07:34.3370521Z","response_success":true,"response_time":0.0585181}]}

Each log entry is only returned once. After you make this request, the existing log is deleted from memory.

HTTP/SOCKS proxy configuration

Devolutions Gateway and Devolutions Agent support HTTP/HTTPS/SOCKS proxy configuration for outbound requests. Proxies are used for requests such as:

  • Subscriber API notifications (Gateway)
  • AI Gateway provider requests (Gateway)
  • Update downloads and version checks (Agent)

The proxy configuration supports three modes:

  • Off: Never use a proxy, ignore environment variables
  • System: Auto-detect proxy from environment variables or system settings (default)
  • Manual: Use explicitly configured proxy URLs

System proxy detection (default)

By default (Mode: System), proxy configuration is automatically detected from system settings:

  • Environment variables: HTTP_PROXY, HTTPS_PROXY, ALL_PROXY, NO_PROXY
  • Linux sysconfig: /etc/sysconfig/proxy on RHEL/SUSE systems
  • Windows Registry: Per-user and machine-wide settings with WinHTTP fallback
  • macOS System Configuration: SCDynamicStoreCopyProxies()

Gateway configuration example:

{
  "Proxy": {
    "Mode": "System"
  }
}

Agent configuration example:

{
  "Proxy": {
    "Mode": "System"
  }
}

Since System is the default mode, you can omit the Proxy section entirely to use system proxy detection.

Manual proxy configuration

You can override auto-detection by setting Mode: Manual and specifying explicit proxy URLs. The configuration supports protocol-specific proxies (Http, Https) and a fallback proxy (All).

Protocol-specific proxies:

{
  "Proxy": {
    "Mode": "Manual",
    "Http": "http://proxy.corp:8080",
    "Https": "http://proxy.corp:8443"
  }
}

SOCKS5 proxy as fallback for all protocols:

{
  "Proxy": {
    "Mode": "Manual",
    "All": "socks5://proxy.corp:1080"
  }
}

Corporate proxy with authentication:

{
  "Proxy": {
    "Mode": "Manual",
    "Http": "http://username:password@proxy.corp:8080",
    "Https": "http://username:password@proxy.corp:8080"
  }
}

Combined configuration with exclude list:

{
  "Proxy": {
    "Mode": "Manual",
    "Http": "http://proxy.corp:8080",
    "Https": "http://proxy.corp:8080",
    "Exclude": ["localhost", ".internal.net", "192.168.1.0/24"]
  }
}

Disabling proxy

To completely disable proxy usage (ignoring system settings):

{
  "Proxy": {
    "Mode": "Off"
  }
}

Exclusions

When using Mode: Manual, specify exclusions in the Exclude array:

{
  "Proxy": {
    "Mode": "Manual",
    "Http": "http://proxy.corp:8080",
    "Exclude": [
      "localhost",
      ".internal.corp",
      "devolutions.net",
      "127.0.0.1",
      "192.168.0.0/16"
    ]
  }
}

The Exclude list supports NO_PROXY semantics:

  • Wildcard: * (bypass proxy for all targets)
  • Exact hostname: localhost, example.com
  • Domain suffix: .corp.local (matches foo.corp.local)
  • IP address: 127.0.0.1
  • CIDR range: 10.0.0.0/8, 192.168.0.0/16

Proxy selection priority

For Mode: Manual, the proxy is selected based on the target URL scheme:

  1. For http:// requests → use Http proxy, fallback to All
  2. For https:// requests → use Https proxy, fallback to All
  3. For other schemes → use All proxy

If the target matches an entry in the Exclude list, no proxy is used.

Testing proxy configuration

Test Gateway subscriber notifications with a proxy:

  1. Configure the proxy in gateway.json:
{
  "Proxy": {
    "Mode": "Manual",
    "Https": "http://proxy.corp:8080"
  },
  "Subscriber": {
    "Url": "https://example.com/notifications",
    "Token": "your-token-here"
  }
}
  1. Monitor Gateway logs for proxy-related messages:
tail -f /var/log/devolutions-gateway/gateway.log | grep -i proxy

Test Agent updates with a proxy:

  1. Configure the proxy in agent.json:
{
  "Proxy": {
    "Mode": "Manual",
    "Https": "http://proxy.corp:8080"
  },
  "Updater": {
    "Enabled": true
  }
}
  1. Trigger an update check and monitor logs:
# View agent logs
tail -f /var/log/devolutions-agent/agent.log | grep -i "download\|proxy"

Proxy-based credentials injection for RDP

How it works

  • Perform two-way forwarding between the client and the target until the TLS security upgrade.
  • Separately perform the TLS upgrade for both the client and the server, effectively acting as a man-in-the-middle.
    • The client must trust the TLS certificate configured in the Devolutions Gateway.
  • Separately perform CredSSP authentication as server with the client, and as client with the target.
    • The fake, proxy credentials are used with the client.
    • The real, target credentials are used with the target.
  • Proceed with the usual two-way forwarding (except we can actually see and inspect all the traffic)

Prerequisites

  • Generate some tokens. You can use tokengen or the PowerShell cmdlet.
    • Generate a session token for the RDP session.
    • Generate a scope token for the preflight API.
  • Configure the TLS certificate and private key. Optionally, configure CredSspCertificateFile and CredSspPrivateKeyFile to use a different certificate for CredSSP.
  • Run the Devolutions Gateway.
    • We’ll assume it runs on localhost, and it listens for HTTP on 7171 and TCP on 8181.
    • Adjust to your needs.

Push the credentials

curl "127.0.0.1:7171/jet/preflight?token=<SCOPE_TOKEN>" \
  -X POST -H "Content-Type: application/json" \
  --data '[
    {"id": "ef1a3ae9-e55d-48b8-92b0-ae67c29b2e4e", "kind": "provision-credentials", "token": "<SESSION_TOKEN>",
      "proxy_credential": { "kind": "username-password", "username": "FakeUser", "password": "FakePassword" },
      "target_credential": { "kind": "username-password", "username": "RealUser", "password": "RealPassword" } }
  ]'

Connect using the fake (proxy) credentials

xfreerdp3 /v:127.0.0.1:8181 /u:'FakeUser' /p:'FakePassword' /pcb:<SESSION_TOKEN>

You may also add the option /cert:ignore if the certificate you configured is not trusted.

Demo

proxy-based-credentials-injection-prototype.webm

Traffic Audit API

The Traffic Audit API provides endpoints to claim and acknowledge traffic events for external processing. This enables integration with external auditing and compliance systems.

Overview

The traffic audit system uses a lease-based claim/acknowledgment pattern:

  1. Claim events with a lease duration to prevent concurrent processing
  2. Process the events in your external system
  3. Acknowledge the events to remove them from the queue

Authentication

Both endpoints require a SCOPE token with appropriate scopes:

  • /jet/traffic/claim requires gateway.traffic.claim or * scope
  • /jet/traffic/ack requires gateway.traffic.ack or * scope

Use tokengen:

tokengen --provisioner-key <path/to/provisioner.key> scope 'gateway.traffic.claim'
tokengen --provisioner-key <path/to/provisioner.key> scope 'gateway.traffic.ack'

Or alternatively, the New-DGatewayToken cmdlet:

New-DGatewayToken -Type SCOPE -Scope 'gateway.traffic.claim'
New-DGatewayToken -Type SCOPE -Scope 'gateway.traffic.ack'

Claiming Traffic Events

Claim up to a specified number of traffic events with a lease duration.

# Basic claim request (uses defaults: lease_ms=300000, max=100)
curl -X POST "https://gateway.example.com/jet/traffic/claim" \
  -H "Authorization: Bearer <TOKEN>"

# Claim with custom parameters
curl -X POST "https://gateway.example.com/jet/traffic/claim?lease_ms=60000&max=50" \
  -H "Authorization: Bearer <TOKEN>"

Query Parameters:

  • lease_ms (optional): Lease duration in milliseconds (1000-3600000, default: 300000 = 5 minutes)
  • max (optional): Maximum number of events to claim (1-1000, default: 100)

Response:

[
  {
    "id": "01JFQH8V5HCZN8XKZJ3Y6M4W9P",
    "session_id": "550e8400-e29b-41d4-a716-446655440000",
    "outcome": "normal_termination",
    "protocol": "tcp",
    "target_host": "example.com",
    "target_ip": "192.168.1.100",
    "target_port": 80,
    "connect_at_ms": 1672531200000,
    "disconnect_at_ms": 1672531260000,
    "active_duration_ms": 60000,
    "bytes_tx": 1024,
    "bytes_rx": 2048
  }
]

Acknowledging Traffic Events

Acknowledge processed events to remove them from the queue.

# Acknowledge specific event IDs (ULID format)
curl -X POST "https://gateway.example.com/jet/traffic/ack" \
  -H "Authorization: Bearer <TOKEN>" \
  --json '{"ids": ["01JFQH8V5HCZN8XKZJ3Y6M4W9P", "01JFQH8V5JD2N7XKZJ4Y7M5W0Q", "01JFQH8V5KE3N9XKZJ5Y8M6W1R"]}'

Response:

{
  "deleted_count": 3
}

Notes

  • Event IDs are ULIDs (Universally Unique Lexicographically Sortable Identifiers)
  • Events are returned in ascending ID order (ULIDs are time-ordered)
  • Claimed events are locked for the specified lease duration
  • If not acknowledged before lease expiry, events become available for reclaiming
  • Events are permanently deleted after acknowledgment (no retention)