Error: P1001: Can't reach database server at MY-CONNECTION-STRING (vercel | neon | supabase | railway)

[Naeem-gg]

Bug description

Hi there! I think there's a problem with prisma with serverless postgresql, I tried this on vercel postgres, supabase, railway and Neon and found this issue with all of mentioned services where it shows the same error as you can see in uploaded image. It happens very less with vercel postgres compared to others. How it gets resolved? well you need to sign in to the service dashboard and do something with databse either view some records or edit it doesn't matter you just need to interact with it and then come back to terminal and run any prisma command it will work fine. I never encountered the same with drizzle. So that's clearly something wrong with prisma and cold start of serverless postgres

How to reproduce

There's no perticular way to reproduce it but It happens with Neon the most. So create a db with Neon and continue working on it (while all the browsers where neon is logged in are closed to make sure its not running in bg) and you'll get the exact same error when you fire prisma commands db push/pull or even migrate dev

Expected behavior

No response

Prisma information

// Add your schema.prisma
generator client {
  provider = "prisma-client-js"
}

datasource db {
  provider = "postgresql"
  url      = env("DATABASE_URL")
}

model User {
  id        String     @id @default(cuid())
  email     String     @unique
  name      String
  password  String
  unicode   String
  createdAt DateTime   @default(now())
  updatedAt DateTime   @updatedAt
}

// Add your code using Prisma Client
import { PrismaClient } from "@prisma/client";

declare global {
  namespace NodeJS {
    interface Global {}
  }
}
interface CustomNodeJsGlobal extends NodeJS.Global {
  prisma: PrismaClient;
}
declare const global: CustomNodeJsGlobal;

const prisma = global.prisma || new PrismaClient();

if (process.env.NODE_ENV === "development") global.prisma = prisma;

export default prisma;

Environment & setup

prisma : 5.11.0
@prisma/client : 5.11.0
Computed binaryTarget : windows
Operating System : win32
Architecture : x64
Node.js : v20.9.0
Studio : 0.499.0

Prisma Version

Environment variables loaded from .env
prisma                  : 5.11.0
@prisma/client          : 5.11.0
Computed binaryTarget   : windows
Operating System        : win32
Architecture            : x64
Node.js                 : v20.9.0
Studio                  : 0.499.0

[Naeem-gg]

I'm using Nextjs v14.1.4 + vercel postgres with mentioned versions of prisma and client (if it helps)

[kameyaw14]

For me I just changed my wifi connection and it worked

[ludralph]

Hi @Naeem-gg 👋

The error you're encountering, P1001: Can't reach database server, is a common issue when using Prisma with serverless databases like Neon. This error typically occurs when the connection created by Prisma Client times out before the serverless compute is activated.

Serverless databases like Neon have two main states: Active and Idle. If there is no query activity for a certain period (like 5 minutes for Neon), the database goes into an idle state. When you try to connect to an idle compute from Prisma ORM, Neon automatically activates it. However, this activation can take a few seconds and added latency can result in a connection timeout.

To address this issue, you can adjust your connection string by adding a connect_timeout parameter. This parameter defines the maximum number of seconds to wait for a new connection to be opened. The default value is 5 seconds. A higher setting should provide the time required to avoid connection timeout issues. For example:

DATABASE_URL=postgres://daniel:<password>@ep-mute-rain-952417.us-east-2.aws.neon.tech/neondb?connect_timeout=10

A connect_timeout setting of 0 means no timeout.

Another possible cause of connection timeouts is Prisma ORM's connection pool, which has a default timeout of 10 seconds. If you are still experiencing connection timeouts, you can try increasing this limit (in addition to the connect_timeout setting described above) by setting the pool_timeout parameter to a higher value.

See this section of the documentation on resolving connection timeout with Neon.

If this answers your question, it would be great if you could mark this Discussion as answered to indicate that it has been resolved.

Otherwise please let us know how else we can help you further or close the Discussion if it was resolved in some other way 🙏

[Naeem-gg]

Hi @Naeem-gg 👋

The error you're encountering, P1001: Can't reach database server, is a common issue when using Prisma with serverless databases like Neon. This error typically occurs when the connection created by Prisma Client times out before the serverless compute is activated.

Serverless databases like Neon have two main states: Active and Idle. If there is no query activity for a certain period (like 5 minutes for Neon), the database goes into an idle state. When you try to connect to an idle compute from Prisma ORM, Neon automatically activates it. However, this activation can take a few seconds and added latency can result in a connection timeout.

To address this issue, you can adjust your connection string by adding a connect_timeout parameter. This parameter defines the maximum number of seconds to wait for a new connection to be opened. The default value is 5 seconds. A higher setting should provide the time required to avoid connection timeout issues. For example:

DATABASE_URL=postgres://daniel:<password>@ep-mute-rain-952417.us-east-2.aws.neon.tech/neondb?connect_timeout=10

A connect_timeout setting of 0 means no timeout.

Another possible cause of connection timeouts is Prisma ORM's connection pool, which has a default timeout of 10 seconds. If you are still experiencing connection timeouts, you can try increasing this limit (in addition to the connect_timeout setting described above) by setting the pool_timeout parameter to a higher value.

See this section of the documentation on resolving connection timeout with Neon.

If this answers your question, it would be great if you could mark this Discussion as answered to indicate that it has been resolved.

Otherwise please let us know how else we can help you further or close the Discussion if it was resolved in some other way 🙏

Hey @ludralph 👋 thanks for your response, please let me check these parameters that you suggested. I just have one small question here, is it safe to override the default timeouts?

[Naeem-gg]

Hey @ludralph its still the same with connect_timeout=20 as well

[ludralph]

Can you try increasing both connect_timeout and pool_timeout. According to the documentation on Neon, you can try adjusting both the connect_timeout and pool_timeout parameters:

DATABASE_URL=postgres://daniel:<password>@ep-mute-rain-952417.us-east-2.aws.neon.tech/neondb?connect_timeout=15&pool_timeout=15

[mdkaifansari04]

Thanks @ludralph, it worked for me, I just increase the connect_timeout=20 and it worked. Guys we can defenitly modify the connection_timeout as it was mentioned in the prisma-neon docs, so this is safe.

[cerastvs]

@ludralph thanks bro it worked for me!

[98hub]

I am facing the same issue rn.
Is there anyone who has solved it?

[Lukas-Blom]

Hi I also switched to my mobile internet wifi and it worked. Other wifi might have firewalls that blocks this type of connection. Hope this helps.

[Blake-Larson]

I was on my work's VPN, disconnected and it worked.

[jv18creator]

@Divyanshu1020 how did you solve it?

[Divyanshu1020]

Just change dns setting, use Google dns server
In my opinion
By default computer don't have any dns setting so by default network use router or modem dns setting that's why it's is not working with wifi network

[salman-node]

same issue RESOLVED just my adding ?connect_timeout=15&pool_timeout=15 in url

[Reinhartsamuel]

I tried to connect prisma and neon db for ever but never succeeded, so I moved to some other managed postgresql service:

• koyeb : same problem
• aiven : success after 3 attempts

So I just use aiven instead of neon

[ludralph]

Hi there,

To keep our discussions organized and focused on the most relevant topics, we’re reviewing and tidying up our backlog. As part of this process, we’re closing discussions that haven’t had any recent activity and appear to be outdated.

If this discussion is still important to you or unresolved, we’d love to hear from you! Feel free to reopen it or start a new one with updated details.

For more details about our priorities and vision for the future of Prisma ORM, check out our latest blog post: https://www.prisma.io/blog/prisma-orm-manifesto.

Thank you for your understanding and being part of the community!;

[abdullahilashik]

Removing the sslmode=require resolved the issue for me.

[Akhil7K]

I don't believe it is a right thing to do.

[theogenejr]

The issue you are facing has to deal with the formatting of your DATABASE_URL value in your .env file. What you need to do is add a ? and a timeout command after your url

update the value in your .env file like this

DATABASE_URL = 'url + ?sslaccept=strict&connect_timeout=500'

Forexample

..........us-east-2.aws.neon.tech/neondb?sslmode=require?sslaccept=strict&connect_timeout=500

[bryanpascuadizon]

This worked for me. thanks!

[Tiru-99]

I’ve run into this issue multiple times, and here’s what usually helps:

✅ 1️⃣ Increase the connection timeout — sometimes the default is too low, and bumping it up solves the problem. You can do it by adding sslaccept=strict&connect_timeout=500 at the end of your connection string

✅ 2️⃣ (This works ~70% of the time!) — it’s a network issue.
Try switching your internet connection or disabling your VPN — trust me, it works like magic! ✨

[davistran86]

thanks, connect_timeout solved my current problem 👍

[kadhirvelm]

On the off chance this helps someone else, I was running into the same issue for weeks on an EC2. I tried increasing the connect_timeout and it didn't work, but interestingly I could get it to work from my local machine. So I tried an nslookup on my EC2 and on my local machine to see if it was resolving the addresses differently. Which as it turns out, my EC2 had some suspicious looking IPv6 addresses, which my local machine did not.

So I temporarily disabled the IPv6 stuff on my EC2 via

sudo sysctl -w net.ipv6.conf.all.disable_ipv6=1
sudo sysctl -w net.ipv6.conf.default.disable_ipv6=1

And tried again with connect_timeout=30 and it worked great after that

[yihui-zhou]

not working for me adding timeout

[nestcub]

increasing connection timeout doesnt work for me.
resetting password didnt work for me.
disabling 5432 (ipv6) on the remote server didnt work for me.

what else i can do?

[Shubhofficial1]

Sometimes the wifi network is somehoe blocking the request. Try switching to some other network just to double check if it is realted to firewall issue. Otherwise use the connection_timeout solution

[Bala-Prakash-M]

For me it worked after VPN

[ichavesbr]

Same here. I use Proton VPN, and it always gives me issues with Prisma + Neon. I closed the VPN, and it worked.