🐛 Align session-ended WebSocket teardown with the latest view's end clocks

[bdibon]

Motivation

When a RUM session expires, open WebSocket connections are finalized with tracking_end_reason: 'session_end'. The teardown used clocksNow() independently in both trackViews (to end the active view) and webSocketCollection (to close the websocket vital/resource). Because these two calls happened at slightly different times, the websocket's end_time could land after the view's end_time — causing the websocket-closed vital to be dropped during assembly and the resource's end_time to fall outside the lifespan of the view it was attached to.

Changes

• Add a SessionExpiredEvent type ({ endClocks: ClocksState }) carried on the SESSION_EXPIRED lifecycle event, instead of void.
• startRum now captures clocksNow() once when the session expires and notifies SESSION_EXPIRED with those clocks, so every subscriber ends things at the exact same instant.
• trackViews ends the current view using the shared endClocks from SESSION_EXPIRED instead of computing its own.
• webSocketCollection.flushOpenConnections now accepts an optional endClocks param, using the shared clocks passed on SESSION_EXPIRED (falling back to clocksNow() for its own stop() call, where no shared clocks exist).
• Unit tests covering the new SessionExpiredEvent payload and the wired session-end path in startRum, trackViews, and webSocketCollection.
• E2E test (renamed to "collects the websocket-closed vital when the session expires") asserting the websocket-closed vital and resource end_time do not exceed the associated view's end time.

Test instructions

yarn test:unit --spec packages/browser-rum-core/src/boot/startRum.spec.ts
yarn test:unit --spec packages/browser-rum-core/src/domain/view/trackViews.spec.ts
yarn test:unit --spec packages/browser-rum-core/src/domain/resource/webSocketCollection.spec.ts
yarn build:apps && yarn test:e2e -g "collects the websocket-closed vital when the session expires"

Checklist

• Tested locally
• Tested on staging
• Added unit tests for this change.
• Added e2e/integration tests for this change.
• Updated documentation and/or relevant AGENTS.md file

[cit-pr-commenter-54b7da]

Bundles Sizes Evolution

📦 Bundle Name	Base Size	Local Size	𝚫	𝚫%	Status
Rum	180.54 KiB	180.59 KiB	+57 B	+0.03%	✅
Rum Profiler	8.43 KiB	8.43 KiB	0 B	0.00%	✅
Rum Recorder	21.12 KiB	21.12 KiB	0 B	0.00%	✅
Logs	57.05 KiB	57.05 KiB	0 B	0.00%	✅
Rum Salesforce	N/A	138.47 KiB	N/A	N/A	N/A
Rum Slim	138.41 KiB	138.46 KiB	+50 B	+0.04%	✅
Worker	22.96 KiB	22.96 KiB	0 B	0.00%	✅
Rum Shopify	N/A	200.17 KiB	N/A	N/A	N/A
Rum-shopify Profiler	N/A	8.43 KiB	N/A	N/A	N/A
Rum-shopify Recorder	N/A	3.72 KiB	N/A	N/A	N/A

[datadog-datadog-prod-us1]

🎉 All green!

🧪 All tests passed
❄️ No new flaky tests detected

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 77.14% (+0.00%)

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 79de282 | Docs | Datadog PR Page | Give us feedback!}

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 79de282d72

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[bcaudan]

🥜 nitpick: ‏Do we need to check that? I guess ensuring that we are collecting the closing vital should be enough for an e2e scenario

[bdibon]

It doesn't cost us much to have this and it prevents any regression where the vital is emitted with an incoherent timestamp.