Add an RPC cancellation handler #2090
Merged
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
Motivation: As a service author it's useful to know if the RPC has been cancelled (because it's timed out, the remote peer closed it, the connection dropped etc). For cases where the stream has already closed this can be surfaced by a read or write failing. However, for cases like server-streaming RPCs where there are no reads and writes can be infrequent it's useful to have a more explicit signal. Modifications: - Add a `ServerCancellationManager`, this is internal per-stream storage for registering cancellation handlers and storing whether the RPC has been cancelled. - Add the `RPCCancellationHandle` nested within the `ServerContext`. This holds an instance of the manager and provides higher level APIs allowing users to check if the RPC has been cancellation and to wait until the RPC has been cancelled. - Add a top-level `withRPCCancellationHandler` which registers a callback with the manager. - Add a top-level `withServerContextRPCCancellationHandle` for creating and binding the task local manager. This is intended for use by transport implementations rather than users. - Update the in-process transport to cancel RPCs when shutting down gracefully. - Update the server executor to cancel RPCs when the timeout fires. Result: Users can watch for cancellation using `withRPCCancellationHandler`.
glbrntt
force-pushed
the
v2/rpc-is-cancelled-lifecycle
branch
from
9dad9df to
6fa20bd
Compare
gjcairo
reviewed
Oct 9, 2024
| public typealias Inbound = RPCAsyncSequence<RPCRequestPart, any Error> | ||
| public typealias Outbound = RPCWriter<RPCResponsePart>.Closable | ||
|
|
||
| private let newStreams: AsyncStream<RPCStream<Inbound, Outbound>> | ||
| private let newStreamsContinuation: AsyncStream<RPCStream<Inbound, Outbound>>.Continuation | ||
|
|
||
| private struct State: Sendable { | ||
| private var id: UInt64 |
There was a problem hiding this comment.
Can we maybe call this nextHandleID or something like that?
Comment on lines
96
to
97
| /// the duration of the RPC. This function is intended for use when implementing | ||
| /// a ``ServerTransport``. |
There was a problem hiding this comment.
Can we maybe put the second sentence inside an Important block? I think it could be misused.
gjcairo
approved these changes
Oct 9, 2024
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.
Motivation:
As a service author it's useful to know if the RPC has been cancelled (because it's timed out, the remote peer closed it, the connection dropped etc).
For cases where the stream has already closed this can be surfaced by a read or write failing. However, for cases like server-streaming RPCs where there are no reads and writes can be infrequent it's useful to have a more explicit signal.
Modifications:
ServerCancellationManager, this is internal per-stream storage for registering cancellation handlers and storing whether the RPC has been cancelled.RPCCancellationHandlenested within theServerContext. This holds an instance of the manager and provides higher level APIs allowing users to check if the RPC has been cancellation and to wait until the RPC has been cancelled.withRPCCancellationHandlerwhich registers a callback with the manager.withServerContextRPCCancellationHandlefor creating and binding the task local manager. This is intended for use by transport implementations rather than users.Result:
Users can watch for cancellation using
withRPCCancellationHandler.