All Issues
24,993 verified issues
The library doesn't work on Deno.
Describe the bug It was fixed in https://github.com/stripe/stripe-node/issues/997, but as I've pointed out in https://github.com/stripe/stripe-node/issues/997#issuecomment-1115941961, it doesn't work anymore. Maybe it's a regression. To Reproduce See https://github.com/stripe-samples/stripe-node-deno-samples/issues/1 for a description on how to reproduce. Expected behavior No TypeScript errors. Code snippets _No response_ OS macOS Node version Deno 1.21.3 Library version Latest API version 2020-08-27 Additional context Deno compatibility should be verified in CI, to prevent regressions.
how expire key with callback
[code block] like this? [code block]
Getting a lot of ERR_CONTENT_LENGTH_MISMATCH errors for HTTPS
When proxying HTTPS to HTTP, I am getting a lot of ERR_CONTENT_LENGTH_MISMATCH errors in Chrome. For example, if I request an image from the web server through the HTTPS proxy, I get a content length of 50KB but the downloaded image is 30KB (and the image appears to be half-downloaded). This is true for _any_ HTTPS request, not just images. This is also true for _any_ web browser. I am testing in a Windows environment with a fairly slow Internet connection. Perhaps the connection is ending prematurely? Any thoughts on this?
node-http-proxy leaks file descriptors
Run this: [code block] - `ps -ef | grep node` to figure out the PID that's running it. - `watch 'lsof -p 5966 | wc -l'` (replace 5966 with your pid) - Point a browser at http://localhost:9000 (you'll get a 404 from example.com). I was using Chrome and FF. - Hit CTRL-R over and over. - Watch your open file count climb. Sockets don't seem to close for 2 minutes. On OS/X, heavy traffic will run out of file descriptors very quickly (my product's Selenium test case kills node-http-proxy about 1/10th of the way through.) Using node v0.10.25 and node-http-proxy 1.0.2.
Keep Alive (NTLM)
I am trying to proxy content and the target server is running NTLM auth. The NTLM spec requires a certain handshake where the first 401 is sent, and the client responds. http://www.innovation.ch/personal/ronald/ntlm.html [code block] I've run wireshark and fiddler and I can see that the connection is NOT being re-used even though the req and res object include the required connection = 'keep-alive'. I've attached screenshots of wireshark showing different ports being used during the multiple stages of NTLM authentication. Doing some research I got conflicting stories on if keep-alive actually does work, and if it does / does not work in node-http-proxy. so far, running through squid-proxy, fiddler, and other proxies wireshark reports the correct re-using of the connection. The only one that does not show this is node-http-proxy (and it's also the only one that isn't able to auth the user correctly). In addition it seems like the headers for www-authenticate are being mangled when there are multiple ones so I had to put in a patch... Without this Firefox, Chrome, IE were not prompting for NTLM authentication because there was only ONE www-authenticate header being returned rather than two separate ones. [code block] Bad (through node-http-proxy) As shown above you can see for each request we grab a new socket (which destroys NTLM auth and proves that it is not doing keep-alive). Good (through fiddler, squid, etc) The above capture shows that with a proxy running (fidd
HNSW testing
_Originally posted by @alanwli in https://github.com/pgvector/pgvector/issues/181#issuecomment-1693821662_ @ankane, not sure if you've run into this in your testing. When I ran it with 86c29b3bf038de50bb2aec21b6d896823ff1fbbe on an usecases, I was hitting what looks like some kind of race condition with the index. I simplified it down to the following python script - where there is a table with 1k vectors, inserts+deletes workload that always keeps the table at 1k vectors. But when doing an index scan with ef_search set to 1k, I see that the index will not have all 1k vectors - quite a number of occasions where it's missing 10%+ of what should be there, the worst I saw was ~24% missing. Note that I don't see this with ivfflat (uncomment the ivfflat line). Is this expected? [code block] Example output: [code block]
Unhandled error event: ClusterAllFailedError: Failed to refresh slots cache
Error description Hello, we are running ioredis 4.14.0 against an AWS ElastiCache replication group with cluster mode on (12 nodes, 3 shard) and using the cluster configuration address to connect to the cluster. this._redisClient = new Redis.Cluster([`//${host}:${port}`], { maxRedirections: process.env.MAX_REDIRECTIONS, redisOptions: { reconnectOnError: function (err) { const targetError = 'READONLY'; if (err.message.slice(0, targetError.length) === targetError) { // Only reconnect when the error starts with "READONLY" return 2; } return false; }, }, }); }` The log of error that we are getting reads by: `@message 2019-10-30T00:00:06.045Z 2970a6b4-2f41-468a-96f7-96649d690fbc [ioredis] Unhandled error event: ClusterAllFailedError: Failed to refresh slots cache. at tryNode (/opt/nodejs/node_modules/ioredis/built/cluster/index.js:359:31) at /opt/nodejs/node_modules/ioredis/built/cluster/index.js:376:21 at Timeout.duplicatedConnection.cluster.utils_2.timeout (/opt/nodejs/node_modules/ioredis/built/cluster/index.js:623:24) at Timeout.run (/opt/nodejs/node_modules/ioredis/built/utils/index.js:156:22) at Timeout.<anonymous> (/opt/nodejs/node_modules/async-listener/glue.js:188:31) at ontimeout (timers.js:486:15) at tryOnTimeout (timers.js:317:5) at Timer.listOnTimeout (timers.js:277:5)` Error scenario We have an AWS Lambda function that connects to said cluster and is instantiated frequently. Sometimes the request is attende
OpenAI not working on AWS Lambda
Confirm this is a Node library issue and not an underlying OpenAI API issue - [X] This is an issue with the Node library Describe the bug I am trying to use OpenAI on AWS Lambda but currently the code just hangs. I have also tried adding OpenAI as a lambda layer but that is not working as well To Reproduce I have the following package.json [code block] The following is the openai as a Layer package layer-package.zip Code snippets [code block] OS AWS Lambda Node version 18 Library version openai v4.19.0
v5.9.0 (kindof) breaks bull library
Hi, Thanks for maintaing this excellent library. We're trying to upgrade from v5.8.2 to v5.9.0 and we're facing problems with the bull library. This library manages queues of jobs and is backed by Redis. It uses a blocking `brpoplpush` call to wait for jobs to be added to queues, and start processing them as soon as possible. The blocking call uses a timeout of `5` seconds and thus, triggers the new "blocking calls timeout" code you just introduced in 5.9.0 (see https://github.com/redis/ioredis/pull/2050) I didn't manage to reproduce using a small project, the problems appear only in our CI pipelines, which are quite huge, but I tracked the issue down to `Command.setBlockingTimeout`. If I simply return from this command, the build passes. I understand that this may be a difficult race condition problem, but apparently there are side effects when ioredis installs client side timeouts. I'm stil trying to figure out the exact issue and provide a reproduction, but I'm affraid this will be a bit difficult. Currently I'm hacking `blockingTimeoutGrace` with a negative value, so that the new code is not triggered, and this works fine. I'm kindly asking if you would accept a PR providing a way for users of ioredis to disable the blolcking call timeout feature completely, for instance by passing `-1` to the `blockingTimeout` option, or by accepting a new option. Thanks in advance, Cheers, David
Question about client-side use of stripe
Pardon the potential newbie question here, but, I'm wanting to create a token based off card data, without using the `stripe.js` code, and do it 100% in JS + React. So in one of my components, I do this: [code block] And I get this error: [code block] I'm assuming this is because I'm in a strictly client-side only environment here. But surely there has to be a way where I can take credit card information filled into a form, and generate a token based off that, in a PCI-compliant way? Full code: [code block]
Logical replication of Vector column
Hi all, Are the Vector column types supported by logical replication? I am trying to replicate a Vector column on Postgres 15.5, but it is failing with: [code block] As workaround, I am able to replicate the vector as text and cast it back to vector. Thank you!
How to get Twilio class type?
Issue Summary With v3.84, we were able to pull in the type of the Twilio client and reference it in our application: [code block] However, now with v4.7, the same code not longer works, with the following message: [code block] However, `typeof Twilio` does not work and is not what we're looking for. We are looking to reference an instance of the Twilio class. Any advice on how to do this with v4?
Performance degrades when using node-http-proxy to proxy http calls to EC2 instance
My current setup is as follows: I have a Node.js server that is serving http requests (let's call that the main server) and another Node.js server that is proxying requests to the main server (running node-http-proxy v1.0.3). If the main server and proxy server are both running on my machine, everything works great and I don't see any drop in performance when going through the proxy. However, as soon as the main server is running on an EC2 instance (and the proxy still on my local machine), I see the average response time go from 50-100ms to about 800-3000ms per request when I make requests through the proxy vs. directly calling the main server. I'm kind of at a loss because it seems like as far as EC2 (or the network/internet) is concerned, the actual request leaving my machine and coming back to my machine should be the same in both scenarios, correct? If that is the case, then if feels like the issue is somewhere in node/node-http-proxy. However, then why isn't the performance impacted when the main server is running on my machine? Honestly I'm not even sure about the right way to go about debugging the issue and if anyone can provide some guidance I'd be more than happy to run some more tests to help point people in the right direction. Also, I saw issues #334, #305, and #491, but I didn't feel like any of those could have the same root cause as the issues I'm seeing; feel free to correct me though if that assumption happens to be wrong.
Will enableOfflineQueue:false conflict when it's in both cluster config and redisOptions?
There are two "enableOfflineQueue: false" in both cluster config and redisOptions. [code block] after running the code above, there's an error: Stream isn't writeable and enableOfflineQueue options is false. Is there conflict with these config? Thank you very much!
test suite fails w/ ECONNREFUSED
I'm running into issues that seem like a problem with just my system but I was hoping to find some assistance. [code block] When I run [code block] all tests fail that are not expecting a 404 or 500. The errors look like this: [code block] Anyone seen this before? Seems like I'm missing something important. There is very little information on ECONNREFUSED. I do know that the error happens on making the request from proxy to target using the 'request' module (which is expected given the error) but I can't determine why. I'm hoping there's something simple I missed.
Please add a note to README warning of limitations on Windows
On Windows, `maintenance_work_mem` cannot be bigger than 2GB, something that is poorly documented. Because of this, building indices is prohibitively slow for larger vector stores. In order for others not to have to find this out the hard way like I did, I propose adding a small note to the README that warns people of this. Maybe something like this: [code block] Documenting my experience: I created 37 million vectors (PubMed) of 384 dimensions. Loading them in pgvector using bulk loading was quick (< 2 hours). But when I run [code block] After running for 65 hours, progress is at 74.5%, which is 1% higher than it was 24 hours ago.
Slow Responses
Hi, I am having some issues with node-http-proxy. I am running node 0.8.10 on Linux (Debian Squeeze). Without routing through node-http-proxy I see response times of 16 - 30ms. With I get 200+ms. What am I doing wrong? The code I have tried is below: [code block] Regards, Zac
Capabilities List filters for AvailablePhoneNumbers resource don't work AT ALL
Note: These issues are for bugs and feature requests for the helper libraries. If you need help or support, please email help@twilio.com and one of our experts will assist you! 3.17.0 Code Snippet [code block] [code block] [code block] Exception/Log [code block] Steps to Reproduce 1. Request a list of available phone numbers for purchase using either the voiceEnabled, smsEnabled, or mmsEnabled list filters. 2. Observe that the list of phone numbers returns phone numbers with capabilities.voice, capabilities.SMS, or capabilities.MMS return true when they should return false depending on which code snippet you run.
An error occurred with our connection to Stripe. Request was retried 2 times.
Describe the bug I'm running a nextJS app project on Cloudflare workers. Using the latest Stripe node sdk version 20. On the production server when I try to create a checkout session from my api route, I see this error. `An error occurred with our connection to Stripe. Request was retried 2 times.` On localhost the exact same code is working perfectly fine. I double checked my env vars to make sure they're properly passed and correct in production, yet i'm still getting the error. ----- For now what I did is omit the SDK and just use a raw fetch request and it works in production. ----- To Reproduce 1. `npm create cloudflare@latest -- my-next-app --framework=next` https://developers.cloudflare.com/workers/framework-guides/web-apps/nextjs/ 2. `npm i stripe --save` 2. Create an api route like "/api/stripe-checkout" that uses the sdk like this `const session = await stripe.checkout.sessions.create({ ... })` This will cause the error 3. Deploy to Cloudflare workers 4. Call the endpoint Expected behavior The Stripe Checkout page should open the same way it does when I'm doing it from localhost. Code snippets [code block] OS macos Node version 21 Library version 20.0.0 API version 2025-11-17.clover Additional context _No response_
createChatCompletion method returning undefined in the choices array
Describe the bug I'm using node.js with the official opean ai lib and when I tried to send a request, I got nothing, the choices array is empty. It's working fine with davinci model, just this new one `model: "gpt-3.5-turbo",` isn't working. To Reproduce Code snippets _No response_ OS windows Node version v10.19.0 Library version 3.2.1