-
-
Notifications
You must be signed in to change notification settings - Fork 1.7k
meta(changelog): Update Changelog for 8.9.0 #12451
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
This PR introduce a new continuous profiling mode. This mode is exclusive from the current mode which considers starting and stopping profiles on a per span basis. I've picked the interval duration of 5s as somewhat arbitrarily. The idea is that we dont want profiles to grow too large, because that might become a performance issue in the event that we have a lot of deep stack samples to process. Since profiling mode is exclusive, we will require users to add a profilerMode (subject to change) as the SDK option (this is subject to change as we align the APIs cross sdks). In terms of convenience, we are likely also going to add a Sentry.profiler.start/stop methods so that users can have access as to when they can start and stop the profiler (not implemented as we havent standardized on the approach yet) - currently this relies on getIntegrationByName("ProfilingIntegration").profiler.stop Since the UI does not support this mode yet, I will hide the profilerMode hidden and only allow the current automated instrumentation
Instead of #12358, this is a simpler change which ensures we pick the transaction from the scope instead. I also added tests for the various different scenarios, to ensure we see how they behave: 1. INP is emitted _during_ pageload (span is active) 2. INP is emitted _after_ pageload a. Pageload is parametrized (route) b. Pageload is unparametrized (URL) When the pageload is unparametrized (default browser SDK), the transaction is not added to the DSC envelope header (which is correct and also what we do in other places). it is _always_ added to the span attributes now, though. If no span is active, it will use transactionName from the last active pageload/navigation span. There may be edge cases where this is not 100% correct (e.g. if the INP span is only emitted once the pageload is done but another navigation already started) but IMHO these are more edge cases and this change is probably fine for now..? (While at it, I also added an origin to the INP spans)
[Gitflow] Merge master into develop
…lambda-auto` (#12392) Guards the `tryPatcHandler` call by checking for the availability of `require` (which is undefined in ESM). In ESM mode, this call fails because `require` is not available. So let's not call it in this case.
add a test to our ESM Express e2e test app to test that http instrumentation is working correctly by ensuring a `http.client` span is captured. --------- Co-authored-by: Francesco Novy <[email protected]>
## Description As stated [here](https://www.typescriptlang.org/docs/handbook/release-notes/typescript-4-7.html#packagejson-exports-imports-and-self-referencing) in the typescript docs, `types` exports should always come first, so that they are resolved correctly. This also applies to the nested `types` within and `exports` condition. This corrects an issue where `eslint-import-resolver-typescript` was not correctly resolving the full types for Sentry.
Ensure the `types` export is always first, as documented here: https://www.typescriptlang.org/docs/handbook/release-notes/typescript-4-7.html#packagejson-exports-imports-and-self-referencing Follow up to #12355
Remove some unneeded deps, and align playwright version everywhere.
…12398) Fix an issue where the correct transactionName could not be obtained when using nested routes in VueRouter.
Adds CLS, FID, and INP breadcrumbs. Updates all web vital breadcrumbs to include rating Closes #11639 --------- Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: Luca Forstner <[email protected]> Co-authored-by: Andrei <[email protected]> Co-authored-by: Francesco Novy <[email protected]> Co-authored-by: Lukas Stracke <[email protected]> Co-authored-by: Yamagishi Kazutoshi <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Forgot to shoutout @dohooo, @mohd-akram and @ykzts for their help fixing bugs with the SDK this release - will also update the main release notes!
We don't actually use these, so no need to install them.
This updates contributing docs to mention that we use volta, + also to set up PNPM support.
…idn't wrap handler (#12407) Fix span collision by checking if the AWS lambda handler was already wrapped by Otel instrumentation and only if not starting our own root span. The rest of our handler is still being used (i.e. the flushing logic, error capturing, etc) regardless of otel wrapping. Also Adjusted E2E tests to: - more closely resemble the AWS environment - enable auto patching of the handler with the Sentry SDK handler - better check for Otel and manual spans Ref #12409
…span (#12430) Add the `sentry.op` attribute/`op` property to the created root span/transaction of a lambda function.
To automatically capture exceptions from inside a component tree and render a fallback component, wrap the native Solid JS `ErrorBoundary` component with `Sentry.withSentryErrorBoundary`. ```js import * as Sentry from '@sentry/solidjs'; import { ErrorBoundary } from 'solid-js'; Sentry.init({ dsn: '__PUBLIC_DSN__', tracesSampleRate: 1.0, // Capture 100% of the transactions }); const SentryErrorBoundary = Sentry.withSentryErrorBoundary(ErrorBoundary); render( () => ( <SentryErrorBoundary fallback={err => <div>Error: {err.message}</div>}> <ProblematicComponent /> </SentryErrorBoundary> ), document.getElementById('root'), ); ``` **Note**: When using an `ErrorBoundary` in conjunction with Solid Router, the fallback component renders twice, see [here](solidjs/solid-router#440).
Waiting with merging this until the package is ready. --------- Co-authored-by: Lukas Stracke <[email protected]>
Follow-up as cache logic for `ioredis` was already added.
CHANGELOG.md
Outdated
|
||
### Important changes | ||
|
||
- **feat(solid): Add solid sdk** |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
totally optional:
- **feat(solid): Add solid sdk** | |
- **feat(solid): Add solid SDK** |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Updated, thanks :)
9b22918
to
1c36f9b
Compare
CHANGELOG.md
Outdated
- **feat(solid): Add Solid SDK** | ||
|
||
This release adds a dedicated SDK for [Solid JS](https://www.solidjs.com/) in alpha state with instrumentation for | ||
[Solid Router](https://docs.solidjs.com/solid-router) and custom `ErrorBoundary`. See the |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
[Solid Router](https://docs.solidjs.com/solid-router) and custom `ErrorBoundary`. See the | |
[Solid Router](https://docs.solidjs.com/solid-router) and a custom `ErrorBoundary`. See the |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Updated :)
CHANGELOG.md
Outdated
|
||
### Other changes | ||
|
||
- build: Cleanup dependencies (#12405) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
- build: Cleanup dependencies (#12405) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Removed :)
1c36f9b
to
7d29a99
Compare
size-limit report 📦
|
Drafted, waiting for #12453 to make it in. |
Fix a server-side re-export problem of `@sentry/node` exports in the Astro SDK. It seems that the `export * from '@sentry/node'` "overruled" the explicit exports. So this patch changes our export statements to: - only export explicit, named exports in the JS server side entry point - continue exporting all types via the `*` export in the types entry point fixes #12410
7d29a99
to
332a7ab
Compare
No description provided.