fix(replay): Ignore older performance entries when starting manually #13969
Conversation
This file contains 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
size-limit report 📦
|
c298lee
approved these changes
Oct 15, 2024
billyvg
reviewed
Oct 15, 2024
Comment on lines
1061
to
1062
| // We leave 1s wiggle room to accomodate timing differences for "immedidate" manual starts | ||
| const initialTimestampInSeconds = this._context.initialTimestamp / 1000 - 1; |
There was a problem hiding this comment.
Do we want the wiggle room? Since it's manual start, I think it's ok to only capture events after start. It's easier to explain to users as well.
There was a problem hiding this comment.
the problem is that if you have code like this, I guess:
Sentry.init({
replaysErrorSampleRate: 0,
replaysSessionSampleRate: 0,
integrations: [replay]
});
if (isImportantUser) {
replay.start();
}Like this kind of pattern, where users start manually but "immediately", we would otherwise probably miss many of the page load performance entries, as they are often correctly pre-dated to Date.now() 🤔 but no strong feelings, it is a bit "random". happy to generally remove the wiggle room!
There was a problem hiding this comment.
Ah didn't consider that... I think we keep it simple and remove the wiggle room, but don't feel strongly about this.
billyvg
approved these changes
Oct 17, 2024
billyvg
reviewed
Oct 17, 2024
mydea
force-pushed
the
fn/replay-performance-buffer
branch
from
5ee3f10 to
e861f46
Compare
|
I removed the wiggle room, I left in the test that shows the current behavior for immediate start as well :) |
alexandresoro
pushed a commit
to alexandresoro/ouca
that referenced
this pull request
Oct 21, 2024
This PR contains the following updates: | Package | Type | Update | Change | |---|---|---|---| | [@sentry/node](https://github.com/getsentry/sentry-javascript/tree/master/packages/node) ([source](https://github.com/getsentry/sentry-javascript)) | dependencies | minor | [`8.34.0` -> `8.35.0`](https://renovatebot.com/diffs/npm/@sentry%2fnode/8.34.0/8.35.0) | | [@sentry/react](https://github.com/getsentry/sentry-javascript/tree/master/packages/react) ([source](https://github.com/getsentry/sentry-javascript)) | dependencies | minor | [`8.34.0` -> `8.35.0`](https://renovatebot.com/diffs/npm/@sentry%2freact/8.34.0/8.35.0) | --- ### Release Notes <details> <summary>getsentry/sentry-javascript (@​sentry/node)</summary> ### [`v8.35.0`](https://github.com/getsentry/sentry-javascript/blob/HEAD/CHANGELOG.md#8350) [Compare Source](getsentry/sentry-javascript@8.34.0...8.35.0) ##### Beta release of the official Nuxt Sentry SDK This release marks the beta release of the `@sentry/nuxt` Sentry SDK. For details on how to use it, check out the [Sentry Nuxt SDK README](https://github.com/getsentry/sentry-javascript/tree/develop/packages/nuxt). Please reach out on [GitHub](https://github.com/getsentry/sentry-javascript/issues/new/choose) if you have any feedback or concerns. - **feat(nuxt): Make dynamic import() wrapping default ([#​13958](getsentry/sentry-javascript#13958 (BREAKING) - **feat(nuxt): Add Rollup plugin to wrap server entry with `import()` ([#​13945](getsentry/sentry-javascript#13945 **It is no longer required to add a Node `--import` flag. Please update your start command to avoid initializing Sentry twice (BREAKING CHANGE).** The SDK will now apply modifications during the build of your application to allow for patching of libraries during runtime. If run into issues with this change, you can disable this behavior in your `nuxt.config.ts` and use the `--import` flag instead: ```js sentry: { dynamicImportForServerEntry: false; } ``` - **feat(nuxt): Respect user-provided source map generation settings ([#​14020](getsentry/sentry-javascript#14020 We now require you to explicitly enable sourcemaps for the clientside so that Sentry can un-minify your errors. We made this change so source maps aren't accidentally leaked to the public. Enable source maps on the client as follows: ```js export default defineNuxtConfig({ sourcemap: { client: true, }, }); ``` - feat(nuxt): Log server instrumentation might not work in dev ([#​14021](getsentry/sentry-javascript#14021)) - feat(nuxt): Add Http `responseHook` with `waitUntil` ([#​13986](getsentry/sentry-javascript#13986)) ##### Important Changes - **feat(vue): Add Pinia plugin ([#​13841](getsentry/sentry-javascript#13841 Support for [Pinia](https://pinia.vuejs.org/) is added in this release for `@sentry/vue`. To capture Pinia state data, add `createSentryPiniaPlugin()` to your Pinia store: ```javascript import { createPinia } from 'pinia'; import { createSentryPiniaPlugin } from '@​sentry/vue'; const pinia = createPinia(); pinia.use(createSentryPiniaPlugin()); ``` - **feat(node): Implement Sentry-specific http instrumentation ([#​13763](getsentry/sentry-javascript#13763 This change introduces a new `SentryHttpInstrumentation` to handle non-span related HTTP instrumentation, allowing it to run side-by-side with OTel's `HttpInstrumentation`. This improves support for custom OTel setups and avoids conflicts with Sentry's instrumentation. Additionally, the `spans: false` option is reintroduced for `httpIntegration` to disable span emission while still allowing custom `HttpInstrumentation` instances (`httpIntegration({ spans: false })`). - **feat(core): Make stream instrumentation opt-in ([#​13951](getsentry/sentry-javascript#13951 This change adds a new option `trackFetchStreamPerformance` to the browser tracing integration. Only when set to `true`, Sentry will instrument streams via fetch. ##### Other Changes - feat(node): Expose `suppressTracing` API ([#​13875](getsentry/sentry-javascript#13875)) - feat(replay): Do not log "timeout while trying to read resp body" as exception ([#​13965](getsentry/sentry-javascript#13965)) - chore(node): Bump `@opentelemetry/instrumentation-express` to `0.43.0` ([#​13948](getsentry/sentry-javascript#13948)) - chore(node): Bump `@opentelemetry/instrumentation-fastify` to `0.40.0` ([#​13983](getsentry/sentry-javascript#13983)) - fix: Ensure type for `init` is correct in meta frameworks ([#​13938](getsentry/sentry-javascript#13938)) - fix(core): `.set` the `sentry-trace` header instead of `.append`ing in fetch instrumentation ([#​13907](getsentry/sentry-javascript#13907)) - fix(module): keep version for node ESM package ([#​13922](getsentry/sentry-javascript#13922)) - fix(node): Ensure `ignoreOutgoingRequests` of `httpIntegration` applies to breadcrumbs ([#​13970](getsentry/sentry-javascript#13970)) - fix(replay): Fix onError sampling when loading an expired buffered session ([#​13962](getsentry/sentry-javascript#13962)) - fix(replay): Ignore older performance entries when starting manually ([#​13969](getsentry/sentry-javascript#13969)) - perf(node): Truncate breadcrumb messages created by console integration ([#​14006](getsentry/sentry-javascript#14006)) Work in this release was contributed by [@​ZakrepaShe](https://github.com/ZakrepaShe) and [@​zhiyan114](https://github.com/zhiyan114). Thank you for your contributions! </details> --- ### Configuration 📅 **Schedule**: Branch creation - At any time (no schedule defined), Automerge - At any time (no schedule defined). 🚦 **Automerge**: Disabled by config. Please merge this manually once you are satisfied. ♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox. 🔕 **Ignore**: Close this PR and you won't be reminded about these updates again. --- - [ ] <!-- rebase-check -->If you want to rebase/retry this PR, check this box --- This PR has been generated by [Renovate Bot](https://github.com/renovatebot/renovate). <!--renovate-debug:eyJjcmVhdGVkSW5WZXIiOiIzOC4xMjguNiIsInVwZGF0ZWRJblZlciI6IjM4LjEyOC42IiwidGFyZ2V0QnJhbmNoIjoibWFpbiIsImxhYmVscyI6WyJkZXBlbmRlbmNpZXMiXX0=--> Reviewed-on: https://git.tristess.app/alexandresoro/ouca/pulls/251 Reviewed-by: Alexandre Soro <code@soro.dev> Co-authored-by: renovate <renovate@git.tristess.app> Co-committed-by: renovate <renovate@git.tristess.app>
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.
For replay, performance entries are captured separately, and added to the replay before we send it.
Generally, this works just fine, because when we buffer, we clear the captured performance events whenever the buffer is cleared.
However, when we manually start the replay, it can happen that we capture performane entries from way before we started the replay manually, which will in turn extend the timeline of the replay potentially drastically.
To fix this, we now drop any performance events, when starting manually, that happened more than a second before we manually started the replay.
The 1s is an arbitrary number to accomodate the fact that some entries can be emitted slightly before - when we immediately run
replay.start()we generally want to try to keep those too.