fix(graphql-server): attach the notify client's error handler before the LISTEN - #1710
Merged
Merged
Conversation
…the LISTEN A dropped connection between checkout and line 279 emitted 'error' on a Client with no listener, so Node rethrew and the whole GraphQL server exited. Also observes the LISTEN itself so a failed LISTEN reconnects instead of leaving the server silently deaf to schema:update.
Contributor
🤖 Devin AI EngineerI'll be helping with this pull request! Here's what you should know: ✅ I will automatically:
Note: I can only respond to comments from users who have write access to this repository. ⚙️ Control Options:
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Fixes constructive-io/constructive-planning#1605:
graphql-publicexits onError: Connection terminated unexpectedlyemitted on aClientwith noerrorlistener, taking every GraphQL host down until Kubernetes restarts the pod.listenForChangesissuedLISTEN "schema:update"before attaching theerrorhandler, so between pool checkout and thatclient.on('error', …)the client was unguarded — and Node rethrows an unlistenederrorevent. The window is re-entered on every reconnect (the error handler itself callsaddEventListener()), so a flapping Postgres gets repeated chances to kill the process.Teardown moved into
dropListener(client, message, err), which no-ops unlessthis.listenClient === client— so a second failure (or one racing shutdown/removeEventListener) cannot double-release or spawn a second reconnect chain. Reconnect/log/release behavior is otherwise unchanged.Also addresses the second half of the issue: the
LISTENwas fire-and-forget, so a failedLISTENleft the server believing it was watchingschema:updatewhile schema changes silently stopped propagating. It now reconnects through the same path.Test plan
New
graphql/server-test/__tests__/notify-listener.test.ts(no database needed — fakePoolClient) asserts: anerrorlistener is present at the momentquery()is called, a dropped connection releases + reconnects exactly once, repeated failures tear down once, and a rejectedLISTENreconnects.Link to Devin session: https://app.devin.ai/sessions/acb0555695da4f628f32371a9a93dc02
Requested by: @pyramation