[🐛 Bug]: SE_NODE_SESSION_TIMEOUT ignored; session stays active > configured timeout

What happened?

Even though I’ve set SE_NODE_SESSION_TIMEOUT=1200 (20 minutes) and SE_OPTS=--session-timeout 1200 --log-level SEVERE in the Node container, browser sessions continue to show durations > 1 hour in the Selenium Grid UI and do not automatically terminate after idle timeout.

To Reproduce
Steps to reproduce the behavior:

Start Selenium Hub + Node (e.g. Docker, Kubernetes) with the following environment settings:

`- name: SE_NODE_SESSION_TIMEOUT
value: "1200"

• name: SE_OPTS
  value: "--session-timeout 1200 --log-level SEVERE"
• name: SE_SESSION_REQUEST_TIMEOUT
  value: "600"
  `
  Run a WebDriver test that purposely hangs (e.g. driver.get("https://slowpage.example.com") that doesn’t complete).

Watch the Grid UI (Sessions table) — note that even after 20 minutes of inactivity, the “Duration” for the session continues increasing (e.g. 1.3h).

Expected behavior:
After ~20 minutes of inactivity (no WebDriver commands), the session should be terminated and removed from grid — i.e., “Duration” should stop / disappear.

Actual behavior:
Sessions persist indefinitely (or until manual kill via API), showing durations much longer than configured idle timeout.

Environment:

Docker-Selenium version / tag (e.g. selenium/node-chrome:4.23.0-20241008)

kubernetes yaml for chrome node:
apiVersion: apps/v1 kind: Deployment metadata: name: {{ .Release.Name }} labels: app: {{ .Release.Name }} spec: replicas: 40 selector: matchLabels: app: {{ .Release.Name }} template: metadata: labels: app: {{ .Release.Name }} spec: volumes: - name: dshm emptyDir: medium: Memory sizeLimit: 2Gi containers: - name: {{ .Release.Name }} image: selenium/node-chrome:4.21.0-20240521 ports: - containerPort: 5555 - containerPort: 5900 #vnc port env: - name: SELENIUM_LOG_LEVEL value: "ERROR" - name: SE_EVENT_BUS_HOST value: "selenium-hub-c-test" - name: SE_EVENT_BUS_SUBSCRIBE_PORT value: "4443" - name: SE_EVENT_BUS_PUBLISH_PORT value: "4442" - name: SE_NODE_SESSION_TIMEOUT value: "1200" - name: SE_SESSION_REQUEST_TIMEOUT value: "600" - name: JAVA_OPTS value: "-Dwebdriver.chrome.whitelistedIps=" - name: HUB_PORT_4444_TCP_ADDR value: "selenium-hub-c-test" - name: HUB_PORT_4444_TCP_PORT value: "4444" - name: SE_NODE_GRID_URL value: https://selenium-dc-c.deepintent.com/wd/hub - name: SE_NODE_PORT value: "5555" - name: SE_START_VNC value: "False" - name: SE_START_XVFB value: "False" - name: SE_DRAIN_AFTER_SESSION_COUNT value: "5" - name: NODE_MAX_SESSION value: "1" # - name: SE_NODE_HOST # value: "0.0.0.0" - name: SCREEN_WIDTH value: "1920" - name: SCREEN_HEIGHT value: "1024" - name: SE_VNC_NO_PASSWORD value: "1" - name: SE_SESSION_RETRY_INTERVAL value: "2" - name: JAVA_OPTS value: "-Djdk.httpclient.websocket.intermediateBufferSize=3000000 -Dwebdriver.chrome.whitelistedIps=" - name: SELENIUM_LOG_LEVEL value: "ERROR" - name: CHROME_OPTS value: "--no-sandbox --headless" - name: SE_SESSION_TIMEOUT value: "1200" - name: SE_OPTS value: "--log-level SEVERE" volumeMounts: - mountPath: /dev/shm name: dshm resources: limits: memory: "3000Mi" cpu: "500m"

Command used to start Selenium Grid with Docker (or Kubernetes)

kubectl -n platform-sre rollout restart deployment/selenium-chrome-node-c-test
kubectl -n platform-sre rollout status deployment/selenium-chrome-node-c-test

Relevant log output

selenium-chrome-node-c-test-6d686c67df-svfmj selenium-chrome-node-c-test 05:52:56.754 INFO [ProxyNodeWebsockets.createWsEndPoint] - Establishing connection to ws://localhost:43797/devtools/browser/ac8483eb-fa03-4784-8926-6a9a26980658
selenium-chrome-node-c-test-6d686c67df-svsks selenium-chrome-node-c-test 05:16:37.394 INFO [LocalNode.newSession] - Session created by the Node. Id: 35dfba80429ae1e5bf829fad07b484dc, Caps: Capabilities {acceptInsecureCerts: false, browserName: chrome, browserVersion: 125.0.6422.76, chrome: {chromedriverVersion: 125.0.6422.76 (67dcf7562b8f..., userDataDir: /tmp/.org.chromium.Chromium...}, fedcm:accounts: true, goog:chromeOptions: {debuggerAddress: localhost:37699}, networkConnectionEnabled: false, pageLoadStrategy: normal, platformName: linux, proxy: Proxy(), se:bidiEnabled: false, se:cdp: wss://selenium-dc-c.deepint..., se:cdpVersion: 125.0.6422.76, setWindowRect: true, strictFileInteractability: false, timeouts: {implicit: 0, pageLoad: 300000, script: 30000}, unhandledPromptBehavior: dismiss and notify, webauthn:extension:credBlob: true, webauthn:extension:largeBlob: true, webauthn:extension:minPinLength: true, webauthn:extension:prf: true, webauthn:virtualAuthenticators: true}
selenium-chrome-node-c-test-6d686c67

Operating System

kubernetes

Docker Selenium version (image tag)

4.23.0-20241008

Selenium Grid chart version (chart version)

No response

Start by reproducing the reported behavior with the Kubernetes node configuration and the SE_NODE_SESSION_TIMEOUT=1200 setting, then inspect the session shown in the Grid UI. Confirm whether a hanging WebDriver command is treated differently from an idle session, and consider the issue done when the configured timeout reliably terminates the session or the expected limitation is documented.