ECONNREFUSED on localhost

[rtdean93]

Description

I have followed your docs at https://baseai.dev/learn/run-pipe and https://github.com/LangbaseInc/baseai/tree/main/examples/nextjs

I entered my OpenAI key (and even regenerated a new project key) and also added a LangBase key.

I continue to get the following error connecting to :9000 even though the service is running.

APIConnectionError: Connection error.
at Request.send (/Users/bobby/Developer/my-ai-project/node_modules/@baseai/core/src/common/request.ts:55:10)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at Pipe.run (/Users/bobby/Developer/my-ai-project/node_modules/@baseai/core/src/pipes/pipes.ts:192:18)
at main (/Users/bobby/Developer/my-ai-project/index.ts:15:21) {
status: undefined,
headers: undefined,
request_id: undefined,
error: undefined,
code: undefined,
cause: TypeError: fetch failed
at node:internal/deps/undici/undici:12618:11
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at Request.makeRequest (/Users/bobby/Developer/my-ai-project/node_modules/@baseai/core/src/common/request.ts:103:16)
at Request.send (/Users/bobby/Developer/my-ai-project/node_modules/@baseai/core/src/common/request.ts:49:15)
at Pipe.run (/Users/bobby/Developer/my-ai-project/node_modules/@baseai/core/src/pipes/pipes.ts:192:18)
at main (/Users/bobby/Developer/my-ai-project/index.ts:15:21) {
cause: Error: connect ECONNREFUSED ::1:9000
at __node_internal_captureLargerStackTrace (node:internal/errors:496:5)
at __node_internal_exceptionWithHostPort (node:internal/errors:671:12)
at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1555:16) {
errno: -61,
code: 'ECONNREFUSED',
syscall: 'connect',
address: '::1',
port: 9000
}

Code example

No response

Additional context

No response

[msaaddev]

Hi @rtdean93 – Thanks for reaching out! Have you started the BaseAI dev server in your project using

npx baseai@latest dev

If not, please run it and try it again. Here is a quick guide to help you how you can do it: https://baseai.dev/docs/pipe/quickstart#step-6-start-baseai-server

[ahmadawais]

We need to handle this paper cut better. It's on our todo list. @msaaddev @ahmadbilaldev, please prioritize this.

@rtdean93 glad to see you're already trying it out. As @msaaddev mentioned. You are not running your baseai server. Which currently only support 9000 port and can be run by npx baseai@latest dev command in a new terminal.

So you'll be running Next.js dev server in one terminal and BaseAI dev server in another. At the same time.

Closing the issue. Let us know if you still got a problem.

[rtdean93]

Thank you Ahmad, Yes my local server was running… I even tested with a curl command..  I was following the docs you provided — but to be sure - I started again with a fresh install.. and the same thing is occurring. I am attaching the ZIP file of my code (with node_modules and API key removed).  This is the error I am still getting. "(base) ***@***.*** my2-ai-project % npx tsx index.ts pipe.run summararizer RUN pipe.run.baseUrl.endpoint http://localhost:9000/beta/pipes/run pipe.run.options { pipe: Pipe { request: Request { config: { apiKey: undefined, baseUrl: 'http://localhost:9000 <http://localhost:9000/>' } }, pipe: { name: 'summararizer', description: '', status: 'private', meta: { stream: true, json: false, store: true, moderate: true }, model: { name: 'gpt-4o-mini', provider: 'OpenAI', params: { top_p: 1, max_tokens: 1000, temperature: 0.7, presence_penalty: 1, frequency_penalty: 1, stop: [] }, tool_choice: 'auto', parallel_tool_calls: true }, messages: [ { role: 'system', content: 'You are a content summarizer. You will summarize content without loosing context into less wordy to the point version.\n' } ], variables: [], tools: [], functions: [], memorysets: [] }, tools: {}, maxCalls: 100, hasTools: false }, messages: [ { role: 'user', content: '\n' + 'Langbase studio is your playground to build, collaborate, and deploy AI. It allows you to experiment with your pipes in real-time, with real data, store messages, version your prompts, and truly helps you take your idea from building prototypes to deployed in production with LLMOps on usage, cost, and quality.\n' + 'A complete AI developers platform.\n' + '- Collaborate: Invite all team members to collaborate on the pipe. Build AI together.\n' + "- Developers & Stakeholders: All your R&D team, engineering, product, GTM (marketing and sales), literally invlove every stakeholder can collaborate on the same pipe. It's like a powerful version of GitHub x Google Docs for AI. A complete AI developers platform.\n" } ], stream: true } Request.post.options { endpoint: '/beta/pipes/run', body: { pipe: { name: 'summararizer', description: '', status: 'private', meta: { stream: true, json: false, store: true, moderate: true }, model: { name: 'gpt-4o-mini', provider: 'OpenAI', params: { top_p: 1, max_tokens: 1000, temperature: 0.7, presence_penalty: 1, frequency_penalty: 1, stop: [] }, tool_choice: 'auto', parallel_tool_calls: true }, messages: [ { role: 'system', content: 'You are a content summarizer. You will summarize content without loosing context into less wordy to the point version.\n' } ], variables: [], tools: [], functions: [], memorysets: [] }, messages: [ { role: 'user', content: '\n' + 'Langbase studio is your playground to build, collaborate, and deploy AI. It allows you to experiment with your pipes in real-time, with real data, store messages, version your prompts, and truly helps you take your idea from building prototypes to deployed in production with LLMOps on usage, cost, and quality.\n' + 'A complete AI developers platform.\n' + '- Collaborate: Invite all team members to collaborate on the pipe. Build AI together.\n' + "- Developers & Stakeholders: All your R&D team, engineering, product, GTM (marketing and sales), literally invlove every stakeholder can collaborate on the same pipe. It's like a powerful version of GitHub x Google Docs for AI. A complete AI developers platform.\n" } ], stream: true, llmApiKey: 'sk-proj-XXXXXXXXXXXXXXeUsQh27RvM623jDdE6D6hBMFMPhooiaq2XT3BlbkFJHab8F0hRxoMoMwnWyYqKulsOMG6i5Zrbn-sdDuDC9-dvYke5oQFVNcZ_oA' } } ***@***.***/core/src/common/request.ts:55 throw new APIConnectionError({ ^ APIConnectionError: Connection error. at Request.send ***@***.***/core/src/common/request.ts:55:10) at process.processTicksAndRejections (node:internal/process/task_queues:95:5) at Pipe.run ***@***.***/core/src/pipes/pipes.ts:192:18) at main (/Users/bobby/Developer/my2-ai-project/index.ts:15:21) { status: undefined, headers: undefined, request_id: undefined, error: undefined, code: undefined, cause: TypeError: fetch failed at node:internal/deps/undici/undici:12618:11 at process.processTicksAndRejections (node:internal/process/task_queues:95:5) at Request.makeRequest ***@***.***/core/src/common/request.ts:103:16) at Request.send ***@***.***/core/src/common/request.ts:49:15) at Pipe.run ***@***.***/core/src/pipes/pipes.ts:192:18) at main (/Users/bobby/Developer/my2-ai-project/index.ts:15:21) { cause: Error: connect ECONNREFUSED ::1:9000 at __node_internal_captureLargerStackTrace (node:internal/errors:496:5) at __node_internal_exceptionWithHostPort (node:internal/errors:671:12) at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1555:16) { errno: -61, code: 'ECONNREFUSED', syscall: 'connect', address: '::1', port: 9000 } } } Node.js v18.20.4 (base) ***@***.*** my2-a” I will use your web UI and learn more about your framework - and I know you guys have bigger things to sort out than this… but as a developer - I’d want to know if someone was trying to follow my docs and still not getting the app running. I will also trying this on another MacBook. Thanks and Cheers, Bobby Dean

…

On Sep 30, 2024, at 3:49 PM, Ahmad Awais ⌘ ***@***.***> wrote: We need to handle this paper cut better. It's on our todo list. @msaaddev <https://github.com/msaaddev> @ahmadbilaldev <https://github.com/ahmadbilaldev>, please prioritize this. @rtdean93 <https://github.com/rtdean93> glad to see you're already trying it out. As @msaaddev <https://github.com/msaaddev> mentioned. You are not running your baseai server. Which currently only support 9000 port and can be run by npx ***@***.*** dev command in a new terminal. So you'll be running Next.js dev server in one terminal and BaseAI dev server in another. At the same time. Closing the issue. Let us know if you still got a problem. — Reply to this email directly, view it on GitHub <#15 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAHAKDSPLM4VZCJSE5F2SSDZZG2MXAVCNFSM6AAAAABPEEVUUCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGOBUGEYTSMRRGI>. You are receiving this because you were mentioned.

[ahmadawais]

It looks like something else is running on port 9000 like a PHP server?

Have you tried killing the 9000 port and trying again?

Error: connect ECONNREFUSED ::1:9000

From Claude:

To find and terminate a process using port 9000 on a Mac, you can follow these steps:

1. Open Terminal on your Mac.

2. Run the following command to find the process ID (PID) using port 9000:

lsof -i :9000

This will display information about any process using port 9000, including its PID.

3. If a process is found, note its PID from the output.

4. To terminate the process, use the kill command followed by the PID:

kill <PID>

Replace <PID> with the actual process ID you found in step 3.

5. If the process doesn't terminate, you can use a stronger termination signal:

kill -9 <PID>

This forcefully terminates the process.

6. Verify that the port is no longer in use by running the lsof command again:

lsof -i :9000

If no output is displayed, the port is now free.

Then try again.

Let me know how it goes.