ferret Posts - Answer Overflow

ferret

Explore posts from servers

DIAdiscord.js - Imagine an app RRailway

Posts Comments

RRailway

•Created by ferret on 3/23/2024 in #✋｜help

Railway SvelteKit deploy doesn't match local configuration

37 replies

RRailway

•Created by ferret on 12/22/2023 in #✋｜help

issue with prisma connecting to planetscale on railway

9 replies

DIAdiscord.js - Imagine an app

•Created by ferret on 11/8/2023 in #djs-questions

images in an embed don't load

17 replies

RRailway

•Created by ferret on 9/20/2023 in #✋｜help

websocket: bad handshake with gorilla websockets

i have a websocket server running on railway with a custom domain, and when going to the websocket route in the browser, i get the expected error:

Bad Request
websocket: the client is not using the websocket protocol: 'upgrade' token not found in 'Connection' header

Bad Request
websocket: the client is not using the websocket protocol: 'upgrade' token not found in 'Connection' header

so i think the server is exposed correctly. however, when connecting using my cli app with the gorilla websockets package, i get this error:

2023/09/19 18:30:04 websocket: bad handshake
exit status 1

2023/09/19 18:30:04 websocket: bad handshake
exit status 1

i'm not entirely sure why. code for connecting:

package client

import (
    "fmt"
    "net/url"
    "os"
    "os/signal"

    "github.com/gorilla/websocket"
)

type StartRequest struct {
    MetaApiKey  string  `json:"META_API_KEY"`
    RiskPercent float64 `json:"RISK_PERCENT"`
    AccountId   string  `json:"ACCOUNT_ID"`
    BrokerSuffix string `json:"BROKER_SUFFIX"`
}

func Connect() (*websocket.Conn, chan struct{}, error) {
    interrupt := make(chan os.Signal, 1)

    signal.Notify(interrupt, os.Interrupt)

    u := url.URL{Scheme: "ws", Host: os.Getenv("SERVER_HOST"), Path: "/websocket"}

    fmt.Println(u.String())

    c, _, err := websocket.DefaultDialer.Dial(u.String(), nil)

    if err != nil {
        return &websocket.Conn{}, make(chan struct{}, 0), err
    }

    done := make(chan struct{}, 1)

    go func() {
        <-done

        c.Close()
    }()

    return c, done, nil
}

package client

import (
    "fmt"
    "net/url"
    "os"
    "os/signal"

    "github.com/gorilla/websocket"
)

type StartRequest struct {
    MetaApiKey  string  `json:"META_API_KEY"`
    RiskPercent float64 `json:"RISK_PERCENT"`
    AccountId   string  `json:"ACCOUNT_ID"`
    BrokerSuffix string `json:"BROKER_SUFFIX"`
}

func Connect() (*websocket.Conn, chan struct{}, error) {
    interrupt := make(chan os.Signal, 1)

    signal.Notify(interrupt, os.Interrupt)

    u := url.URL{Scheme: "ws", Host: os.Getenv("SERVER_HOST"), Path: "/websocket"}

    fmt.Println(u.String())

    c, _, err := websocket.DefaultDialer.Dial(u.String(), nil)

    if err != nil {
        return &websocket.Conn{}, make(chan struct{}, 0), err
    }

    done := make(chan struct{}, 1)

    go func() {
        <-done

        c.Close()
    }()

    return c, done, nil
}

19 replies

RRailway

•Created by ferret on 8/4/2023 in #✋｜help

trouble exposing a go app

33 replies

RRailway

•Created by ferret on 7/30/2023 in #✋｜help

build stuck at initializing

10 replies

RRailway

•Created by ferret on 7/17/2023 in #✋｜help

flask css error

18 replies

RRailway

•Created by ferret on 7/17/2023 in #✋｜help

custom domain not working

13 replies

RRailway

•Created by ferret on 1/11/2023 in #✋｜help

Question about how logs are gathered

I have a question about how logs are gathered on railway. I have a k8s cluster where user pods are dynamically configured in a new namespace, and I'd like to run a sidecar pod that intercepts logs from another container in the same pod. So far, I haven't found a way to grab the logs from another running container. I'd like to know if railway does something similar

4 replies

RRailway

•Created by ferret on 9/21/2022 in #✋｜help

Question About Railway & Nixpacks

Hey all, I think nixpacks is super interesting and I'm looking to integrate it in my own project, but I'm confused about how I want to implement it. I'm working on my own hosting platform for fun. I'm running a Kubernetes cluster that has all the microservices for the platform. I have a deployment microservice where I'd like to build a Docker image using nixpacks and run it. I was curious about how Railway integrated nixpacks into the deployment pipeline and was wondering if anyone could share some insight. Thanks in advance 🙂

18 replies

Gaming

Programming