We don't have a set timeline for it. We have a fix that has been developed, but deployment and test

We don't have a set timeline for it. We have a fix that has been developed, but deployment and testing is complex and ongoing. The underlying issue, intermittently poor connectivity in the APAC region, is not one that is easily solved.

When we finish rolling out our improvement I'll ping here so folks can test.

AAJR We don't have a set timeline for it. We have a fix that has been developed, but...

yevgen•11/11/24, 12:30 PM

We actually have a workaround for customers with origin databases in APAC region, if this is a case for @chimame we can place theirs specific hyperdrive config in APAC today.

Yyevgen We actually have a workaround for customers with origin databases in APAC region...

okku•11/11/24, 12:33 PM

I am also the one struggling this APAC networking issue, I would appreciate too if you can go for it

AJROP•11/11/24, 12:50 PM

Eng team can with a manual process. If you have configs that need to be placed into apac, please DM me the IDs and we'll see to it.

@chimame I thought we'd worked with you on this over the summer.

AJROP•11/11/24, 12:54 PM

Ah. Yes, the process is to place an individual config. New ones created generally won't have this applied.

I recall we might have an account level option where all your configs would go to the same locations. Let me double check

AJROP•11/11/24, 12:55 PM

Regardless, manual process, workaround is indeed available. The systematic fix so you don't need an engineer to help with it is in progress.

Ben-xD•11/18/24, 11:04 PM

Before hyperdrive was added, there was an article for connecting to postgres over TCP: https://web.archive.org/web/20230519161124/https://developers.cloudflare.com/workers/databases/connect-to-postgres/. This URL now redirects to hyperdrive docs. Does anyone know why? Was the useful feature paywalled as "hyperdrive", since that also uses TCP sockets? Or is hyperdrive something better, and the old approach still works?

Connect to PostgreSQL · Cloudflare Workers docs

Cloudflare Workers supports direct connectivity to PostgreSQL instances over the Transmission Control Protocol (TCP) using the Socket API, and via …

BBen-xD Before hyperdrive was added, there was an article for connecting to postgres ove...

AJROP•11/18/24, 11:09 PM

The second one. TCP direct connect still works, if it works for your use case.

Hyperdrive will work much better for almost everyone though.

AJROP•11/18/24, 11:11 PM

Also, as an interesting side note, Hyperdrive doesn't actually use TCP sockets for almost anything. I think only local dev might be an exception?

Ben-xD•11/18/24, 11:14 PM

Thanks!

thomasgauvin•11/19/24, 4:01 PM

Hyperdrive update:

2024-11-19: Hyperdrive now supports clear-text password authentication. When your database requires your password in clear-text, Hyperdrive will send the required credentials securely over TLS during the connection setup.

https://developers.cloudflare.com/hyperdrive/platform/changelog/#2024-11-19

alex (he/him)•11/20/24, 3:11 PM

Is it possible to use

localConnectionString

localConnectionString

with a remote connectionString, we use Neon for development, so every developer would need to set it's own connectionString, we where thinking of using

localConnectionString

localConnectionString

for this, however it always timesout

Agung Vivekananda•11/20/24, 3:15 PM

Why sometimes i got this error
message:

"Database error: PostgresError: Protocol Error: Unable to close prepared statement: hesvlatshn72",

(I use hyperdrive -> Supabase and CF worker backend)

Aalex (he/him)Is it possible to use `localConnectionString` with a remote connectionString, we...

AJROP•11/20/24, 3:18 PM

It's meant to be possible, yes, as of a couple weeks ago. I'm getting those timeouts too. I'll open a ticket to take a look when we can. Thanks!

AAgung Vivekananda Why sometimes i got this error message: "Database error: PostgresError: Proto...

AJROP•11/20/24, 3:19 PM

Which driver and/or ORM are you using?

thomasgauvin•11/20/24, 3:19 PM

@alex (he/him) what version of Wrangler are you using? it works on the latest 3.88.0

AJROP•11/20/24, 3:20 PM

I hit it with Neon just now on 3.88.0

thomasgauvin•11/20/24, 3:20 PM

And @alex (he/him) do you have IP allowlisting for each developers' IP address? Might be responsible for timeout

alex (he/him)•11/20/24, 3:22 PM

@thomasgauvin I'm using latest 3.88.0, and I have allowed connection from

0.0.0.0/0

0.0.0.0/0

, so I really don't think it's the cause here

alex (he/him)•11/20/24, 3:23 PM

It also happens for RDS & Neon, so two different provider.. If there is anything else I can provide, logs? Or you want me to run any command, let me know. I can provide those.

thomasgauvin•11/20/24, 3:24 PM

And presumably, developers can sign in from local machine to the pg database directly in Node or via PgAdmin?

alex (he/him)•11/20/24, 3:24 PM

Yes that's correct! However using the "original" connectionString, I have not tried to use the hyperdrive specific one

postgresql://XXX:XXX@ffca328615404666884948d7d21ccf36.hyperdrive.local:5432/database?sslmode=prefer

postgresql://XXX:XXX@ffca328615404666884948d7d21ccf36.hyperdrive.local:5432/database?sslmode=prefer

AJROP•11/20/24, 3:25 PM

The Hyperdrive one won't work outside a Worker, that's an internally-generated identifier

AJROP•11/20/24, 3:26 PM

If you deploy a random test worker with your configs it behaves as expected?

AJROP•11/20/24, 3:26 PM

Any particular SSL/TLS configuration you've set up on your origin DBs? I assume not, just checking

alex (he/him)•11/20/24, 3:26 PM

Production works as expected. It really is about

localConnectionString

localConnectionString

combined with a db not running on localhost.

alex (he/him)•11/20/24, 3:27 PM

No particular SSL/TLS.. I would need to confirm if

localConnectionString

localConnectionString

works with a db on localhost

alex (he/him)•11/20/24, 3:28 PM

{"DB":{"connectionString":"postgresql://repguide_owner:XXX@ffca328615404666884948d7d21ccf36.hyperdrive.local:5432/repguide?sslmode=prefer","port":5432,"host":"ffca328615404666884948d7d21ccf36.hyperdrive.local","password":"XXX","user":"repguide_owner","database":"repguide"}}
✘ [ERROR] Uncaught (in promise) Error: write CONNECT_TIMEOUT ffca328615404666884948d7d21ccf36.hyperdrive.local:5432

      at connectTimedOut
  (file:///Users/alexanderniebuhr/Developer/Projects/frickegroup/workers/node_modules/.pnpm/postgres@3.4.5/node_modules/postgres/cf/src/connection.js:259:20)
      at done
  (file:///Users/alexanderniebuhr/Developer/Projects/frickegroup/workers/node_modules/.pnpm/postgres@3.4.5/node_modules/postgres/cf/src/connection.js:1035:8)

{"DB":{"connectionString":"postgresql://repguide_owner:XXX@ffca328615404666884948d7d21ccf36.hyperdrive.local:5432/repguide?sslmode=prefer","port":5432,"host":"ffca328615404666884948d7d21ccf36.hyperdrive.local","password":"XXX","user":"repguide_owner","database":"repguide"}}
✘ [ERROR] Uncaught (in promise) Error: write CONNECT_TIMEOUT ffca328615404666884948d7d21ccf36.hyperdrive.local:5432

      at connectTimedOut
  (file:///Users/alexanderniebuhr/Developer/Projects/frickegroup/workers/node_modules/.pnpm/postgres@3.4.5/node_modules/postgres/cf/src/connection.js:259:20)
      at done
  (file:///Users/alexanderniebuhr/Developer/Projects/frickegroup/workers/node_modules/.pnpm/postgres@3.4.5/node_modules/postgres/cf/src/connection.js:1035:8)

AJROP•11/20/24, 3:28 PM

Yep. Ok, I'll circle up with Thomas to dig into this when we have some spare cycles. Making this work with remotes is a new feature, and it's worked for us in our testing, but it seems there's some configuration still to be done on our side.

AJROP•11/20/24, 3:29 PM

Thanks for letting us know

alex (he/him)•11/20/24, 3:29 PM

What did you use in your testing? I'm using RDS & Neon, so two different providers which are quite common.

AAJR Which driver and/or ORM are you using?

Agung Vivekananda•11/20/24, 3:30 PM

I'm using node-postgres 8.13.1

AAgung Vivekananda I'm using node-postgres 8.13.1

AJROP•11/20/24, 3:31 PM

Are you explicitly using CLOSE statements?

alex (he/him)•11/20/24, 3:34 PM

I would need to confirm if localConnectionString works with a db on localhost

@AJR @thomasgauvin I actually get timeouts too for a db running on localhost

let me go back one version of wrangler

Aalex (he/him)> I would need to confirm if localConnectionString works with a db on localhost ...

AJROP•11/20/24, 3:35 PM

Well that's not good. That worked fine before.

Aalex (he/him)> I would need to confirm if localConnectionString works with a db on localhost ...

Slylencer•11/20/24, 3:35 PM

This is occurring for me as well

alex (he/him)•11/20/24, 3:36 PM

3.87.0 also seems to be broken for me.. maybe it's something on my system, but I'm still doubting this

Slylencer•11/20/24, 3:37 PM

I got the same error both on 3.87 and 3.88 locally as well

AJROP•11/20/24, 3:40 PM

Ok, this is going to take some time to unpack. I'll let the team know.

AJROP•11/20/24, 3:40 PM

Sorry for the inconvenience, all

AAJR Are you explicitly using CLOSE statements?

Agung Vivekananda•11/20/24, 3:40 PM

I'm not using CLOSE statement on my query

alex (he/him)•11/20/24, 3:40 PM

3.86.0 also broken for me

alex (he/him)•11/20/24, 3:42 PM

3.84.1 works for me.. 3.85.0 doesn't work at all, cause it breaks with service bindings

Aalex (he/him)3.86.0 also broken for me

AJROP•11/20/24, 3:42 PM

3.83.0 is the last version before we made the only tweak I'm aware of to local dev

AJROP•11/20/24, 3:42 PM

Ah, 3.84.1 works locally? Does it cooperate with local dev against a remote connection string?

Big Tay•11/23/24, 12:07 AM

I put together a very simple example repo showing a Worker + Durable object + Hyperdrive (Postgres) integration
https://github.com/shadrach-tayo/cloudflare-worker-example

I just recently migrated our Automerge CRDT sync server architecture to use Partykit + Cloudflare workers Durable Object + Hyperdrive (Postgres) with tunnel deployed in our kubernetes cluster.

Happy hacking weekend guys

GitHub

GitHub - shadrach-tayo/cloudflare-worker-example: An example reposi...

An example repository showcasing how to implement a Cloudflare worker + DurableObjects binding + Hyperdrive (Postgres) integration - shadrach-tayo/cloudflare-worker-example

Jjared I'm not able to create new Hyperdrive configs or update the existing ones via th...

Huxwell | Reown•11/26/24, 9:18 AM

Hey @AJR I'm facing the same issue.

{"DB":{"connectionString":"postgresql://repguide_owner:XXX@ffca328615404666884948d7d21ccf36.hyperdrive.local:5432/repguide?sslmode=prefer","port":5432,"host":"ffca328615404666884948d7d21ccf36.hyperdrive.local","password":"XXX","user":"repguide_owner","database":"repguide"}} ✘ [ERROR] Uncaught (in promise) Error: write CONNECT_TIMEOUT ffca328615404666884948d7d21ccf36.hyperdrive.local:5432 at connectTimedOut (file:///Users/alexanderniebuhr/Developer/Projects/frickegroup/workers/node_modules/.pnpm/postgres@3.4.5/node_modules/postgres/cf/src/connection.js:259:20) at done (file:///Users/alexanderniebuhr/Developer/Projects/frickegroup/workers/node_modules/.pnpm/postgres@3.4.5/node_modules/postgres/cf/src/connection.js:1035:8)

We don't have a set timeline for it. We have a fix that has been developed, but deployment and test

Similar Threads

Similar Threads

Similar Threads