All Issues
24,993 verified issues
Some sites break because of req.url being absolute
According to the HTTP standard, a request's URI (the second field in the first line of an HTTP request, after the verb, and which gets into req.url) may be relative or absolute. A request to a proxy must always have an absolute URI, and indeed when we receive a request, req.url is always absolute. We just forward the request in proxyRequest and copy the URI into the new request. The receiving end of our new request will in turn get an absolute URI. This behaviour, however standard, is breaking some sites which naively expect the path to be relative to the hostname (starting with a "/"). They are wrong, but still node-http-proxy could be "fixed" so as to not break this assumption of theirs. By placing "req.url = req.url.replace(/._?:\/\/._?\//, '/')" (an ugly regexp, granted) before calling proxyRequest I was able to conform to the naiveté of Wordpress and other stuff which is out there being used and does not implement the standard. Thanks for building this nice, flexible proxy.
Windows AWSCLI64PY3 MSI not shipping aws.exe anymore breaks scripts
The new CLI version on Windows does not install the 'aws.exe' file anymore, instead it's aws.cmd. This breaks bash scripts running in Windows because bash does not resolve .cmd files as executable. Would you consider shipping an .exe wrapper? It's nice being able to execute .sh scripts on Linux and Windows unchanged, but now this forces us to change our calls to 'aws.cmd' rather than just 'aws'
ci-kubernetes-e2e-gci-gke-reboot-release-1.5: broken test run
https://k8s-gubernator.appspot.com/build/kubernetes-jenkins/logs/ci-kubernetes-e2e-gci-gke-reboot-release-1.5/348/ Multiple broken tests: Failed: [k8s.io] Reboot [Disruptive] [Feature:Reboot] each node by triggering kernel panic and ensure they function upon restart {Kubernetes e2e suite} [code block] Issues about this test specifically: #34123 #35398 Failed: [k8s.io] Reboot [Disruptive] [Feature:Reboot] each node by switching off the network interface and ensure they function upon switch on {Kubernetes e2e suite} [code block] Issues about this test specifically: #33407 #33623 Failed: [k8s.io] Reboot [Disruptive] [Feature:Reboot] each node by ordering clean reboot and ensure they function upon restart {Kubernetes e2e suite} [code block] Issues about this test specifically: #33874 Failed: [k8s.io] Reboot [Disruptive] [Feature:Reboot] each node by dropping all inbound packets for a while and ensure they function afterwards {Kubernetes e2e suite} [code block] Issues about this test specifically: #33405 Failed: [k8s.io] Reboot [Disruptive] [Feature:Reboot] each node by ordering unclean reboot and ensure they function upon restart {Kubernetes e2e suite} [code block] Issues about this test specifically: #33882 #35316 Failed: [k8s.io] Reboot [Disruptive] [Feature:Reboot] each node by dropping all outbound packets for a while and ensure they function afterwards {Kubernetes e2e suite} [code block] Issues about this test specifically: #33703 #36230 Previous issues for thi
Returning "Network Error" on unauthorized error (401)
Describe the issue Axios request below is returning "Network Error", on API's unauthorized return. The API is in Asp Net Core with Identity. Therefore, it is recognized that my API actually returns a 401, but my request callback is not allowing me to get this return code, so that I can handle this exception directly. I understand that, of course, my axios request tried to connect to the server, however, the identity by default doesn't even establish this connection in cases of unauthorized, it just returns the code 401, so I imagine that then axios returns this condition that there really was a "Network Error", since that's exactly what happened. I would like an idea of how we can solve this. This is the code for the request in question [code block] Environment - Axios Version [0.27.2] - Adapter [e.g. XHR/HTTP] - Browser [e.g. Chrome, Safari] - Browser Version [e.g. 22] - Node.js Version [16.16.0] - OS: [Win 10] - Additional Library Versions [React 18.1.0, Next 12.1.6]
Unusable in Cloudflare workers due to use of node specific APIs/packages
Describe the bug This seems to be a broad regression, given that Cloudflare workers is supposed to be supported. The root of the issue is usage of node specific APIs and globals. - webhooks uses Buffer - not supported in workers, one workaround would be to wrap it in a try/catch and fall back to assuming the value is a string (or I guess just check if it exists and fall back) - utils requires `crypto` which is required early in stripe.ts - it's plausible that import could be replaced/polyfilled with Web Crypto - there's a few references to `http`/`https` in `createNodeHttpClient`- the compiler doesn't seem to be smart enough to skip compiling NodeHttpClient (I think since it's imported by default? It's possible that dynamically requiring it could help?) - somewhere there's `child_process` usage I was able to get things mostly working by using the package.json `browser` field to stub out the `crypto`(swapped for `crypto-js`)/`http`/`https`/`child_process`(swapped the others for an empty dummy.ts file) imports, but webhooks unchecked Buffer usage fully blocked me. To Reproduce I'll see if I can get a repro repo up but: - create a new cloudflare function - install stripe-node - observe many errors Expected behavior Everything should work! Code snippets _No response_ OS Ubuntu Node version 19 Library version 10.15.0 API version 22-08-01 Additional context I'm @cal on slack if chatting synchronously helps!
Anthropic's dependent library "Anyio" is incompatible with Gunicorn worker class "Eventlet" or "gevent"
I have a python application running using eventlet for serving requests. I also need to install anthropic, but it never settles with the application because of this async library "Anyio". Can't anthropic work without "Anyio" ? We can externally provide the asynchronous feature right ?
[k8s.io] ConfigMap should be consumable in multiple volumes in the same pod {Kubernetes e2e suite}
https://k8s-gubernator.appspot.com/build/kubernetes-jenkins/logs/kubernetes-e2e-gke-staging-parallel/6307/ Failed: [k8s.io] ConfigMap should be consumable in multiple volumes in the same pod {Kubernetes e2e suite} [code block]
Test suite fails with pytest 8
On openSUSE Tumbleweed when we try and run the test suite, we are seeing failures with the following errors. [code block]
Suddenly stopped working: Got an unexpected keyword argument 'proxies'
I have a a script that uses multiple LLMs such as gemini, gpt and claude. I am actually not using claude but the code is still there in case I want. Now, while updating the google genai sdk with pip install google-genai, Claude suddenly stopped working. I haven't touched the part of Claude. The only thing I did was the pip install for google-genai. So, not sure if that is at all related but besides that I have not done any changes. Using version 0.39.0 of anthropic [code block] Inside llm_configs I have this: [code block] If I uncomment CLAUDE_CLIENT it works.
Placeholder issue for feedback on `@defer` & `@stream` proposal
For each release of GraphQL.js, we are now publishing an accompanying release containing an experimental implementation of the `@defer` and `@stream` directives. We are hoping to get community feedback on these releases before the proposal is accepted into the GraphQL specification. You can use this experimental release of GraphQL.js by adding the following to your project's package.json file. [code block] A similar release of `express-graphql` will be coming soon. I am creating this issue as a placeholder for anyone who has tried out these experimental packages to leave feedback. Please let us know if everything is working as expected or if you run into any issues. Any suggestions for improvements are also welcome. References: Demo repo Conference talk about defer/stream defer/stream RFC Spec edits graphql-js branch express-graphql PR graphql-over-http RFC fetch-multipart-graphql (client library) meros (client library) Feedback received so far Call `return` on iterable when connection closes Raised by @danielrearden `graphql-js` Fixed: https://github.com/graphql/graphql-js/pull/2843 Released: TBD `express-graphql` Fixed: https://github.com/graphql/express-graphql/pull/583/commits/ce8429e5c15172b394e65d5a27491611b5fb354e Released: `express-graphql@0.12.0-experimental-stream-defer.1` `content-length` in payload should not be required, parsers should use boundary instead Raised by @maraisr `graphql-over-http` spec Fixed: https://g
Preview feature feedback: Node-API support (`nApi`)
Please share your feedback about the `nApi` functionality released in v2.20.0 in this issue. - If you encounter a bug, please open a bug report in this repo. - If the feature is working well for you, please share this in a comment below or leave a 👍 on this issue. If you have any questions, don't hesitate to ask them in the `#prisma-client` channel in the Prisma Slack. Known Limitations - [x] (Resolved) You can't have more than one instance of the Prisma Client yet, so this will not work properly: [code block] - [x] (Resolved) Currently, logging queries does not work with Node-API and Jest (Issue). For example, the following will not output any query logs. (_Note_: This seems to only be the case when using jest) [code block] - [x] (Resolved) Node-API libraries do not work with process mocking of Jest (issue) - [x] Node-API does not work on M1 ARM Macs, Apple Silicon (issue) - [ ] Node-API does not work on 32bit Node (issue)
"@tailwindcss/vite": "4.0.8" and "@tailwindcss/vite": "4.0.9" have massively increased load times for the CSS in Vite.
So I have noticed drastically increased load times when using Vite and Tailwind when updating to "@tailwindcss/vite": "4.0.8" and "@tailwindcss/vite": "4.0.9". "@tailwindcss/vite": "4.0.8" and "@tailwindcss/vite": "4.0.9" <img width="1331" alt="Image" src="https://github.com/user-attachments/assets/9b284158-be5f-4a4b-994e-ed3b4624befa" /> "@tailwindcss/vite": "4.0.7" and lower. <img width="1421" alt="Image" src="https://github.com/user-attachments/assets/b76206e7-effb-4d97-9583-77a5be53937e" /> Not sure what is causing this. Happy to share more information about my setup or debug further with instructions. Thanks
LUA eval not recognizing key slot
I am using the cluster access capabilities of ioredis. Running scripts that SET data are working, but when trying to run a simple GET: "return redis.call('GET', 'foo')" the client is not able to recognize where the key "foo" is stored in the cluster. Is there a way to do this besides parsing CLUSTER SLOTS to find the keyslot for foo manually?
[FEATURE]: @libsql/client/web Cloudflare Pages
Describe what you want When running in a JavaScript environment without Node.js APIs (for example, Cloudflare Workers or browsers) we need `import { createClient } from "@libsql/client/web";` intead `import { createClient } from "@libsql/client";` I would very like to use Drizzle with Turso and CloudFlare pages. Thanks so much for the amazing job.
Memory Leak
We've been experiencing a memory leak in our node servers. Over the course of a day or so the memory in our servers started to fill until we got an alert and had to restart the app. We added new relic to check the node VM memory: Upon debugging and testing we've come to the conclusion that the leak could be in express. We did some test on a clean express server, printing the heap before and after a lot of request and we see the memory increasing but never going back to the initial value. We tested on various versions of Node and two versions of express (4.16.2 and 4.14.1) and we had the same results. I'll add the last test setup so you can reproduce it. Tested on OSX 10.12.6 Node 7.6.0 Express 4.16.1 Code [code block] run with `node --expose-gc server.js` autocannon command `autocannon -a100000 -c5 http://localhost:9000/` Results: <img width="745" alt="screen shot 2018-02-05 at 17 54 44" src="https://user-images.githubusercontent.com/1228656/35828372-01003654-0a9e-11e8-95ca-899450cf9f20.png"> This is just one of the many test we made, all with similar results
ReplyError: ERR SELECT is not allowed in cluster mode
Seeing SELECT not allowed in cluster mode every time we connect to Redis. Exact Error Seen You have triggered an unhandledRejection, you may have forgotten to catch a Promise rejection: ReplyError: ERR SELECT is not allowed in cluster mode at JavascriptRedisParser.replyParser.Parser.returnError (/opt/app/node_modules/ioredis/lib/redis/parser.js:25:25) at JavascriptRedisParser.execute (/opt/app/node_modules/ioredis/node_modules/redis-parser/lib/parser.js:572:12) at Socket.<anonymous> (/opt/app/node_modules/ioredis/lib/redis/event_handler.js:107:22) at Socket.emit (events.js:107:17) at readableAddChunk (_stream_readable.js:163:16) at Socket.Readable.push (_stream_readable.js:126:10) at TCP.onread (net.js:540:20)
[FEATURE]: Method to convert all fields to camelCase when using db.execute
Describe what you want EDIT: The casing can be fetched from the global config, as described in #2834, instead of having some arbitrary method for it. When using `db.execute`, the fields remain the same case as they are in the DB; this is totally valid and acceptable. However, it would be neat to have a method on it (`$toCamelCase`) that converts all the fields to camelCase. For example, running this method: [code block] currently outputs `{ foo_bar: 'some-value' }`, but with something like `$toCamelCase`, [code block] it would output `{ fooBar: 'some-value' }`. Currently I do this manually; I wonder if the team would consider this since it's kind of subjective. Feel free to close if this is not planned.
Unable to do paging from the built-in library file - /lib/base/page.js or /lib/rest/api/v2010/account/message.js
Hi, am unable to do paging concepts using the built-in file like page.js or message.js. My requirement is - want to fetch 1000 records at a time and get all the messages from each/list with help of nextpageuri. I tried to make connection of message.js with the below code snippet and it throws error. [code block] Please provide the sample code to get a list of messages(ex:1000) at a time and then recursively using nextpageuri concept get all the remaining messages until the nextpageuri is NULL.
How to do 301 Redirects
I'm trying to remove the www. requests to point them to the canonical non www. urls. This is the make sure copy pasted urls point to one domain and thus not divide up my Google Page Rank juice across the two domains.
Expose GITHUB_JOB_ID as an environment variable
Describe the enhancement Expose `GITHUB_JOB_ID` as an environment variable to the running steps. We need this in order for a job to be able to self-identify itself and potentially send it's own ID to a different service that would then use the GitHub API to perform actions (like check logs for instance). Code Snippet It looks like we need to add `job_id` to the whitelist here --> https://github.com/actions/runner/blob/master/src/Runner.Worker/GitHubContext.cs Not sure though, hence raising this issue