Initialization problem on custom client
Hi, I'm working on custom client and using with stdio.
` server_params = StdioServerParameters( command=self.server_config.command, args=self.server_config.args, env=self.server_config.env, )
stdio_transport = await self.exit_stack.enter_async_context(
stdio_client(server_params)
)
self.stdio, self.write = stdio_transport
session = await self.exit_stack.enter_async_context(
ClientSession(self.stdio, self.write)
)
await session.initialize()
`
On the initialization, it gives log
FireCrawl MCP Server running on stdio
But do not continue after it. It seems hang or stuck.
I'm using latest mcp library for Python.
I encountered the same problem, my pyther mcp version is the latest mcp==1.3.0. Have you solved it?
Hi, I'm working on custom client and using with stdio.嗨,我正在开发定制客户端并使用 stdio。
` server_params = StdioServerParameters(server_params = StdioServerParameters( command=self.server_config.command, args=self.server_config.args, env=self.server_config.env, )
stdio_transport = await self.exit_stack.enter_async_context( stdio_client(server_params) ) self.stdio, self.write = stdio_transport session = await self.exit_stack.enter_async_context( ClientSession(self.stdio, self.write) ) await session.initialize()`
On the initialization, it gives log初始化时,它给出日志
FireCrawl MCP Server running on stdioBut do not continue after it. It seems hang or stuck.但是不要继续之后。它似乎挂起或卡住了。
I'm using latest mcp library for Python.我正在使用最新的 Python mcp 库。
I encountered the same problem, my pyther mcp version is the latest mcp==1.3.0. Have you solved it?
I encountered the same problem, my pyther mcp version is the latest mcp==1.3.0. Have you solved it?
Hi, I'm working on custom client and using with stdio.嗨,我正在开发定制客户端并使用 stdio。 ` server_params = StdioServerParameters(server_params = StdioServerParameters( command=self.server_config.command, args=self.server_config.args, env=self.server_config.env, )
stdio_transport = await self.exit_stack.enter_async_context( stdio_client(server_params) ) self.stdio, self.write = stdio_transport session = await self.exit_stack.enter_async_context( ClientSession(self.stdio, self.write) ) await session.initialize()
On the initialization, it gives log初始化时,它给出日志FireCrawl MCP Server running on stdio` But do not continue after it. It seems hang or stuck.但是不要继续之后。它似乎挂起或卡住了。 I'm using latest mcp library for Python.我正在使用最新的 Python mcp 库。I encountered the same problem, my pyther mcp version is the latest mcp==1.3.0. Have you solved it?
did you solve the issue?
No, I can not. I believe lots of MCP's are not yet production level
someone was able to make it work on python?
Encounter the same problem
"It seems like there's some sort of two-way communication deadlock happening. I managed to get around it by commenting out the success initialization message code in my local index.js file (which, for me, was located at ~/.npm/_npx/xxxxx/node_modules/firecrawl-mcp/dist).
server.sendLoggingMessage({
level: 'info',
data: 'FireCrawl MCP Server initialized successfully',
});
server.sendLoggingMessage({
level: 'info',
data: `Configuration: API URL: ${FIRECRAWL_API_URL || 'default'}`,
});
After doing that, it runs successfully, so this might be a temporary workaround. My guess is that the asynchronous code just above,
await server.connect(transport);
might be the culprit. Perhaps the subsequent success initialization messages are being sent before the connection is fully established? I haven't looked into it deeply.
Commenting out the code mentioned above allows the MCP server to start, but similar hanging problems occur when tools are called. My solution was to comment out all server.sendLoggingMessage statements, and so far, everything seems to be working fine.
comment out all server.sendLoggingMessage statements ??? I met the same issue. when will the issue be fixed?