Issue encountered: the MCP server request was unexpectedly redirected to
an HTML page. Because the client didn’t validate the response
Content-Type, it still entered handleSSE, which then kept treating the
HTML as an SSE stream and repeatedly re-requesting the same
page—creating an infinite request loop.
When POST requests in the streamableClientConn return a transient error,
return this error to the caller rather than permanently breaking the
connection.
This is achieved by using the special sentinel ErrRejected error to the
jsonrpc2 layer. In doing so, the change revealed a pre-existing bug:
ErrRejected had the same code as ErrConnectionClosing, and
jsonrpc2.WireError implements errors.Is, so the two sentinel values
could be conflated. This is fixed by using a new internal code.
There's more to do for #683: we should also retry transient errors in
handleSSE.
For #683
Change CloseSSEStream from func(time.Duration) to
func(CloseSSEStreamArgs) for forward compatibility, allowing new fields
to be added without breaking existing callers.
I encountered these logs, and they were distracting and noisy. We
shouldn't log to stderr unless configured by the user. Thread through
the logger.
Also, use ensureLogger in the streamable client to simplify call sites.
Updates #708
Updates #684
I am seeing a bug where connections with a Streamable HTTP client are
not reused at all. This makes performance terrible, since each request
establishes a new TCP connection.
I was able to root cause this down to behavior of the `processStream`
function. This sets up a bufio.NewScanner, and waits to get data. Once
we get a message, we abort and *close the connection*.
At this point, there are two paths:
* The bufio reader read the *full* response body to EOF, and then we
closed. This case is fine; we can re-use the idle connection.
* The bufio reader *did not* real the full response body. It may have
read the entire actual data that *would be sent*, but since its
streaming happened to not yet be delivered. In this case, we call
`body.Close()` before an EOF, which causes the HTTP client to not
attempt to re-use the connection.
In my case, the server I am using reliable sends the response over 2 TCP
packets (one with the event, and the other with the closure of the
`chunked` encoding (`0`)). This causes the body to never be fully read.
This PR fixes that case by unconditionally reading the full body before
we close the connection. With this change, I am able to reliably re-use
connections for future requests.
Replace closure-based deliver and closeLocked fields with explicit
state:
- pendingJSONMessages: signals JSON mode and collects messages
- w: indicates stream is claimed by HTTP request
- done: signals stream completion (non-nil implies w is non-nil)
- lastIdx, supportsPrimeClose, logger: moved from closures to fields
Add stream methods to consolidate logic:
- close: sends close event and closes done channel
- release: releases stream from HTTP request for resumption
- deliverLocked: writes/stores messages, handles request accounting
Factor out hangResponse to document HTTP response lifecycle.
Inline writeEvent to eliminate confusing pointer-based index
management.
After a deep dive with the team, document areas of the codebase
identified as unclear. For now, this is a documentation-only commit (apart
from one switch of jsonrpc2.Request.ID.IsValid->Request.IsCall). Special
TODO(cleanup) comments are added for additional code changes in
follow-up commits.
Also, include high level doc comments for server and client transports,
in new streamable_server.go and streamable_client.go. Eventually, the
server and client transport logic should be migrated to these files, as
streamable.go has gotten too large. However, we can do that at a future
date, to minimize diff.
The prime and close SSE events (SEP-1699) were added in protocol version
2025-11-25. Only send these events when the client negotiates that
version or later.
- In servePOST, extract protocol version from InitializeParams for
initialize requests, otherwise use the Mcp-Protocol-Version header
- In serveGET/acquireStream, read protocol version from header
- Only set closeLocked callback when prime/close is supported
- Add unexported protocolVersion field to ClientSessionOptions for
testing
Fixes#686
Add a UserID field to auth.TokenInfo that TokenVerifiers can populate
from JWT "sub" claims or token introspection. The streamable HTTP
transport uses this to bind sessions to users, rejecting requests where
the user ID doesn't match the session's original user.
Fixes#589
After walking through our handling of streamable client context
cancellation (due to encountering a shutdown deadlock), I think I've
settled on a more coherent strategy for handling call cancellation:
- In our call handler, retire the request if the call exits due to
cancellation: the caller will never see the actual result anyway.
- In connectSSE, use the actual request context (the same context used
in Write) for the client request, so that it terminates when the context
is cancelled. Thread through the initialization context for the
standalone SSE request.
Also, a couple minor improvements:
- Use a detached context for the background context of the client
connection. We want to preserve context values (see #513), but it is not
right to cancel the connection after Connect has already returned, if
the context times out.
- Don't use Last-Event-ID != "" as the signal for whether the connectSSE
call is initial: if the standalone SSE stream disconnects without an
event ID, we'll still reconnect it, and don't want to do so without a
delay.
+ tests, updating the streamable client connection test harness to
accomodate the new aspects being exercised.
Fixes#662
This CL implements support for SEP-1699, adding the following new API:
- Event.Retry corresponds to the 'retry' field of a server sent event.
This is used to signal client retry behavior.
- RequestExtra.CloseStream is set by the streamable transport to allow
closing a stream.
The streamable server transport is updated to set CloseStream for calls,
allowing server operations or middleware to close the underlying stream.
If the configured 'reconnectAfter' delay is set, a 'retry' message is
sent prior to terminating the stream.
The streamable client transport is updated to read the configured delay.
Fixes#630
Fixes#636
## Problem
The StreamableClientTransport tries to JSON-decode ALL SSE events,
including ping events sent by servers for keep-alive purposes. This
causes errors like 'invalid character 'p' looking for beginning of
value' when encountering non-JSON event data.
## Solution
Modified processStream to check evt.Name and skip events that are not
'message' events (or unnamed events, which default to 'message' per SSE
spec).
## Changes
- Added event name check in processStream before attempting JSON decode
- Added test case for ping event filtering
- Added test for scanning multiple events including ping events
## Testing
- All existing tests pass
- New tests verify ping events are properly filtered
- Tested with DeepWiki MCP server which sends ping events
Some MCP servers do not support SSE, but when receiving a request, they
return 400 Bad Request or 404 Not Found instead of the standard 405
Method Not Allowed. The TypeScript and Python SDK ignore this error,
while the Go SDK terminates the session. This patch addresses the issue
in the non-strict mode.
Fixes#610
---------
Signed-off-by: Xie Zhihao <zhihao.xie@bytedance.com>
If a stateful streamable session fails to initialize, it is unusable.
Avoid allocating resources by closing the session immediately.
Also:
- Fix a bug where InitializeParams is not guarded with its mutex.
- Fix a resource leak where sessions are persisted with session id "",
even though they are unaddressable.
- Add a relevant benchmark, and update tests.
Fixes#578
Add a timeout option for the streamable HTTP handler that automatically
cleans up idle sessions.
Also, fix a bug in the streamable client, where we hang on a request
even though the client can never get a response (because the HTTP
request terminated without a response or Last-Event-Id).
Fixes#499
Address longstanding TODOs to clean up stream metadata after streams are
complete:
- Interpret a missing stream as a complete stream.
- Remove the request->stream mapping when the request completes. Any
writes after that point will fail as there can be no recipient.
Tested using examples/client/loadtest, and updating
TestStreamableTransports to check in-band and out-of-band behavior.
This CL implements two changes that must be logically connected.
The first is to provide a mechanism for disabling stream replay.
Previously, our documentation stated that if
StreamableServerTransport.EventStore was nil, a default in-memory event
store would be used. In that case, there would be no way to completely
disable stream replay, and as described in #576, there are many cases
where replay is undesirable.
But of course changing the behavior of the transport zero value
implicitly affects its default behavior, and so this CL also implements
the proposal #580: stream replay is disabled by default.
Implementing this change required a significant refactoring, as
previously we were relying on the event store for serialized message
delivery from the JSON-RPC layer to the MCP layer: the connection would
write to the event store, and independently the stream (be it an
incoming POST or replay GET) would iterate and serve messages in the
stream.
In order to achieve the goals of this CL, it was necessary to decouple
storage from delivery. The 'stream' abstraction now tracks a delivery
callback that writes straight to the HTTP response. It would have been
convenient to store the ongoing http.ResponseWriter directly in the
stream (this is how typescript does it), but due to the design of our
EventStore API, only the HTTP handler knows the next event index, so a
'deliver' abstraction was an unfortunate requirement (suggestions for
how to further simplify this are welcome).
More simplification is possible: in particular, as a result of this
refactoring it should be entirely possible to clean up streams once
we've received all responses. Any replay would only need access to the
EventStore, if at all. This is left to a follow-up CL, to limit this
already significant change.
Furthermore, a nice consequence of this refactoring is that, when not
using event storage, servers can get synchronous feedback that message
delivery failed, which should avoid unnecessary work. We can
additionally cancel ongoing requests on early client termination, but
that is also left to a follow-up CL.
Throughout, the terminology 'standalone SSE stream' replaced 'hanging
GET' when referring to the non-replay GET request issued by the client.
This is consistent with other SDKs.
Fixes#576
Updates #580
PR 501 added logging to the streamable server, but didn't pass a non-nil
logger to the streamable transport (an oversight). However, tests still
passed due to net/http's panic recovery.
Update test servers to enforce that handlers do not panic, and fix the
panic by:
1. Passing along the configured logger to the session transport.
2. Using ensureLogger in Connect, since we technically allow
StreamableServerTransports to be constructed by the user.
Fixes#556
mcp/server.go: implement server‑side logging
- Thread a *slog.Logger through server paths to improve debuggability
- Introduce no‑op discard handler + ensureLogger so loggers are never
nil
- Use ensured loggers in Server
- address todo comments related to adding logging support
All tests are passing and no functionality should change with this PR
Fixes https://github.com/modelcontextprotocol/go-sdk/issues/170
In practice, the streamable client transport was having trouble
connecting to various backends, because they are nonconformant in
various ways.
While it would be nice if all servers conformed to the spec, in practice
there are certain spec violations that are recoverable, and we can and
should recover them. Specifically:
- tolerate 404 instead of 405 for the hanging GET (#393)
- tolerate (=ignore) spurious response body for notifications and
responses, since we know none are expected
Additionally, fix a bug that we weren't parsing Content-Type correctly.
Fixes#521
## Problem
Context propagation was broken in `StreamableClientTransport` because:
1. `Connect()` used `context.Background()` instead of the parent context
2. `Close()` created a race condition where DELETE requests were
cancelled before completion
## Solution
- Use parent context when creating the connection context in `Connect()`
- Reorder `Close()` operations to perform cleanup DELETE before
cancelling context
## Impact
- Request-scoped values (auth tokens, trace IDs) now propagate correctly
to background HTTP operations
- Eliminates race condition in cleanup operations
- Maintains proper Go context semantics
Fixes#513
Add a framework for feature documentation, and start populating it with
our SDK documentation.
This framework is as follows:
- internal/docs/**.src.md is the markdown source for the docs/
directory.
- The x/example/internal/cmd/weave tool is used to compile these docs to
the top-level docs, supporting both linked code samples and generated
tables of contents.
- The readme-check workflow is updated to check these docs as well.
- The structure of these docs follows the MCP spec.
- Wherever possible, example code is linked from actual Go documentation
examples, so that it is testable.
Some minor modifications to the weave tool were made to support this
framework.
Additionally, partially fill out this documentation with content on
base protocol and client features, as well as troubleshooting help.
Along the way, a bug was encountered that our LoggingTransport was not
concurrency safe. This is fixed with a mutex.
Fixes https://github.com/modelcontextprotocol/go-sdk/issues/466
Fixes https://github.com/modelcontextprotocol/go-sdk/issues/409
Updates https://github.com/modelcontextprotocol/go-sdk/issues/442
Previously, the error message received from ClientSession.Wait would
only report the closeErr, which would often be nil even if the client
transport was broken. Wait should return the reason the session
terminated, if abnormal.
I'm not sure of the exact semantics of this, but surely returning nil is
less useful than returning a meaningful non-nil error. We can refine our
handling of errors once we have more feedback.
Also add a test for client termination on HTTP server shutdown,
described in #265. This should work as long as (1) the session is
stateful (with a hanging GET), or (2) the session is stateless but the
client has a keepalive ping.
Also: don't send DELETE if the session was terminated with 404; +test.
Fixes#265
In #272, we made transports open structs and deprecated their
constructors. However, we left the constructors in place to allow go:fix
directives to facilitate migration.
Now let's remove the constructors before cutting the release.
Fixes#305
The streamable client connection can break for a variety of reasons,
asynchronously to the client's request. Decorate these failures with
additional context to clarify why they occurred.
Add a test for the failure message of #393.
Fixes#393
Use a fake streamable server to facilitate testing client behavior.
For this commit, just update the existing test (moved to a new file for
isolation). Subsequent CLs will add more tests.
Improve one client error message that occurred while debuging tests.
For #393
During initialization, the default logging level is set to info allowing
servers to emit logs immediately without waiting for a client to set a
log level.
The updated example shows how to send a notification in the context of
the request.
Fixes#387
The condition sessionID == "" was not quite right for disallowing GET
requests. Since we decided to differentiate "stateless" vs "sessionless"
servers, we need to disallow GET requests for both.
For #393
Exporting StreamableHTTPOptions.JSONResponse is helpful for some users behind proxies that don't work well with text/event-stream.
For #211
Co-authored-by: ccpro10 <co1797249167@gmail.com>
This CL utilizes the event store to write outgoing messages and removes
the unbounded outgoing data structure.
It also adds a new interface [EventStore.Open]
For #190
An onClose function is passed to the ServerSession and ClientSession to
help cleanup resources from the caller.
The onClose function will be executed as part of the ServerSession and
ClientSession closure.
Fixes#258
If there is a TokenInfo in the request context of a
StreamableServerTransport, then propagate it through to the
ServerRequest that is passed to server methods like callTool.
Fixes#317.
A few problems with reconnection cropped up in the review of PR #307.
We should allow for the hanging GET to fail with StatusMethodNotAllowed.
This simply means that the server does not support sending notifications
or requests over the GET, which is allowed in the spec.
Also, we should fix the initial delay of the hanging GET request: it
should start with 0 delay. Fix the math for this and subsequent
attempts. Incidentally, this makes the tests take 3s on my machine, down
from 9s.
Also address some comments from #307.
Several improvements for the stateless streamable mode, plus support for
a 'distributed' (or rather, distributable) version of the stateless
server.
- Add a 'Stateless' option to StreamableHTTPOptions and
StreamableServerTransport, which controls stateless behavior.
GetSessionID may still return a non-empty session ID.
- Audit validation of stateless mode to allow requests with a session
id. Propagate this session ID to the temporary connection.
- Peek at requests to allow 'initialize' requests to go through to the
session, so that version negotiation can occur (FIXME: add tests).
Fixes#284
For #148
