fix(browser-utils): Ensure web vital client hooks unsubscribe correctly#17272
Merged
fix(browser-utils): Ensure web vital client hooks unsubscribe correctly#17272
Conversation
Contributor
size-limit report 📦
|
AbhiPrasad
approved these changes
Jul 31, 2025
Contributor
AbhiPrasad
left a comment
There was a problem hiding this comment.
Honestly feels like we shouldn't be using arrays at all, but approving to unblock
Member
Author
|
Yeah you're right, we should think of a more robust way of storing and removing the subscribers. I'll merge this for now to unblock but will think about a safer approach to solve this more generally. Opened #17276 to track. |
edwardgou-sentry
approved these changes
Jul 31, 2025
This was referenced Sep 19, 2025
Lms24
added a commit
that referenced
this pull request
Sep 19, 2025
This PR fixes two edge cases around our client hook subscriber management: 1. Registering the same callback instance twice on the same hook, resulted in the callback only being invoked once. Fixed by wrapping the passed callback in a function to "unique-ify" it. 2. Unregistering one callback synchronously within the callback, caused other callbacks to not be invoked due to in-place, sync array mutation. Fixed by converting the hooks data structure from `Array<Function>` to `Set<Function>` which is resilient to sync, in-place mutation. This also lets us remove the workaround introduced in #17272 where we initially discovered this bug. Added regression tests for both cases that failed beforehand. closes #17276
This was referenced Sep 29, 2025
This was referenced Oct 7, 2025
This was referenced Oct 15, 2025
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
So this was a fun one to track down 😅
In our standalone span web vitals code, we register multiple client hooks to listen e.g. for
afterStartPageloadSpanevents. This hook will only fire once (by design), so we want to unsbscribe from it afterwards.Howerver, we register two callbacks (one for LCP, once for CLS). Because we used to unsubscribe synchronously from within the client hook callback, we synchronously removed the callback from the client's hooks array. This synchronous array mutation (shrinking) caused the second callback to no longer be executed.
This surfaced by the LCP span being sent but the CLS span not being sent, due to the CLS span's hook callback no longer being called.
This PR fixes this incorrect unsubscription by deferring the unsubscription calls to the next tick. This way, the array mutation no longer happens synchronously and all remaining callback hooks are invoked correctly.
If you're confused by this, rest assured, I was too 😅 Happy to explain better/in-person on request :D
closes https://linear.app/getsentry/issue/JS-811/investigate-missing-standalone-cls-spans-in-latest-sdk-versions