what the hell is up with every cloudflare service being in circular dependancy so its just crashes e

what the hell is up with every cloudflare service being in circular dependancy so its just crashes everything. super weak architecture tbh

cg•9/9/25, 8:33 PM

Is anyone else still seeing issues?

cg•9/9/25, 8:50 PM

The second DB has a size of 0B, this seems suspect. I just created the top row (a fresh DB) and it has a 8kB size with nothing in it

Mmr.niko.la what the hell is up with every cloudflare service being in circular dependancy s...

Chaika•9/9/25, 8:54 PM

I don't think anything is circular, it's all just dogfooding/using internal products.

Access, Containers, Gateway, Realtime, Stream, Turnstile, Workers Builds, Workflows, D1, etc, all depend on/use Durable Objects SQLite in some way. imo that's nothing but good, as it gives some real incentive to make D1/Sqlite stronger. Alternatively, they would all not use D1 and then just external customers would be impacted, not much better for D1 users.

Ccg Is anyone else still seeing issues?

Chaika•9/9/25, 8:55 PM

no but that size is curious, it's lazily updated tho/not from talking to the actual Sqlite DB tho as far as I know, but could be related. What errors are you getting?

cg•9/9/25, 8:56 PM

I'm not seeing any errors. I just cannot connect at all

cg•9/9/25, 8:56 PM

I cannot even load the D1 database in the dashboard

cg•9/9/25, 8:57 PM

It just hangs in the skeleton loading state

Chaika•9/9/25, 8:58 PM

I'm assuming you're querying it from Workers or external? Do you see any explicit errors from that/maybe sending too many queries?

cg•9/9/25, 9:00 PM

Querying from Workers

cg•9/9/25, 9:00 PM

I turned off all our requests to it for about 20 minutes, I've created a new DB and pushing data to that instead

cg•9/9/25, 9:01 PM

So there's no traffic to the D1 and it's unresponsive

cg•9/9/25, 9:01 PM

We were seening no log messages/errors

cg•9/9/25, 9:04 PM

The D1 id is: 1e8baa37-5523-4a34-a01f-c72d83c2b5561e8baa37-5523-4a34-a01f-c72d83c2b556

CChaika I don't think anything is circular, it's all just dogfooding/using internal prod...

mr.niko.laOP•9/9/25, 9:43 PM

not a good look for cf for d1,r2 to go do down compared to aws equivalent services tbh. it really negates all the cool features of cf , workers.

Mmr.niko.la not a good look for cf for d1,r2 to go do down compared to aws equivalent servic...

Dean•9/9/25, 10:19 PM

tend to agree - means the blast radius when DO,KV & R2 go down is huge

Joey9•9/10/25, 3:20 AM

how is the dx for those that have worked with drizzle and d1? are there too many bugs still being worked on or is it ok?

JJoey9 how is the dx for those that have worked with drizzle and d1? are there too many...

Dean•9/10/25, 3:24 AM

seems fine to me. one thing drizzle doesn't support D1 replication so take that in consideration if you want to use that.

Joey9•9/10/25, 3:27 AM

ok thank you

JJoey9 how is the dx for those that have worked with drizzle and d1? are there too many...

cloudkite•9/10/25, 4:58 AM

If you’re just starting the project would recommend using the drizzle beta which should be stable in 1-2 months. Saves having todo a big refactor later when you decide to upgrade to v1

cloudkite•9/10/25, 5:01 AM

Also worth considering if DO SQLite is a better fit for your project, we recently migrated and it simplified our code while improving reliability and performance

Ccloudkite If you’re just starting the project would recommend using the drizzle beta which...

Joey9•9/10/25, 5:14 AM

do you have a link to track progress of the beta? where you see it will be stable in 1-2 months

JJoey9 do you have a link to track progress of the beta? where you see it will be stabl...

cloudkite•9/10/25, 5:32 AM

Their discord server has info on beta and timelines

maj•9/10/25, 2:34 PM

QQ. on the Queries overview for D1: are the "Rows Read/Written" the total for the "count" number of requests? Or average values? I'm trying to understand how those number can be mapped to the numbers given at the top of that page. Because there I have 30M rows read. And I wonder how come I would reach that number, given the top four requests are only in the thousands. And there's not that huge tail with millions of smaller queries.

Mmaj QQ. on the Queries overview for D1: are the "Rows Read/Written" the total for th...

lambrospetrou•9/10/25, 5:03 PM

Let me clarify with the team and get back to you.

OOr We're starting to see an increase in internal errors again, is D1 having known i...

lambrospetrou•9/10/25, 5:43 PM

Are you getting these in some specific location? There isn't any service wide issue at the moment. There was a blip of network issues for 1-2 minutes in eeur for a tiny % of requests, from what I see.

Llambrospetrou Let me clarify with the team and get back to you.

lambrospetrou•9/10/25, 5:46 PM

Currently the column Rows Read is using averages, so 45.6K*662=30M45.6K*662=30M, which matches what you said. We should probably make that sum as well, but it might confuse folks with the count in that case.

OOr Not as far as I can see, our main DBs are all located in ENAM, we turned read r...

lambrospetrou•9/10/25, 5:48 PM

Replication shouldn't cause this. Maybe your errors were from the replica in eeur during that blip of errors?
If you continue seeing consistent errors let us know (account+DB IDs, and some rough timerange).

OOr We're seeing again lots of internal errors, right after I enabled read replicati...

lambrospetrou•9/11/25, 7:22 AM

Yeah, please do if you are still seeing these.

Mukesh•9/11/25, 7:28 AM

I have a use case where I need to connect my D1 database from outside worker, is there possibility to do that?
I searched on google but no luck.

lambrospetrou•9/11/25, 2:15 PM

Hey folks, for those that don't read changelogs, we documented the automatic retries we do for read-only queries. https://developers.cloudflare.com/changelog/2025-09-11-d1-automatic-read-retries/

This has been asked many times over the past few months, so we rolled out our first iteration in the past few weeks.
Read only queries are now retried automatically when the errors are retryable, so your app code should mostly retry idempotent write queries that you deem necessary, or if you want more than two retries that we are doing.

Read the changelog for the full details.

Cloudflare Docs

D1 automatically retries read-only queries

D1 now detects read-only queries and automatically attempts up to two retries to execute those queries in the event of failures with retryable errors.

Silvan•9/11/25, 2:26 PM

damn i just implemented auto retry it all my d1 projects

but nice change

Jun•9/11/25, 2:35 PM

Application-level retries are still recommended for idempotent queries (https://developers.cloudflare.com/d1/best-practices/retry-queries/)

Silvan•9/11/25, 4:43 PM

oh is that page new? never saw it before but good to know

Ehbraheem•9/11/25, 10:45 PM

@Jun I'm curious are the retried queris billable?

What I mean is, if I send a single read, and CF did a retry. Does that count as 2 reads in my bill

EEhbraheem @Jun I'm curious are the retried queris billable? What I mean is, if I send a ...

Larry•9/12/25, 12:42 AM

If it happens often enough to be detectable in your bill, you have bigger things to worry about.

Llambrospetrou Hey folks, for those that don't read changelogs, we documented the automatic ret...

James•9/12/25, 1:33 AM

This is an awesome addition to D1 :MeowHeartCloudflare:

EEhbraheem @Jun I'm curious are the retried queris billable? What I mean is, if I send a ...

lambrospetrou•9/12/25, 6:57 AM

Fair question we did think about. Short answer is that it should not happen in practice, although in theory it is possible.
There is a very rare case that attempt 1 executed on the underlying database, published stats about rows read/written, and then the response failed to be returned back to our D1 worker (what your code is calling through the D1 worker binding), so the worker will retry the query.
In our experience with the errors we see is that the vast majority of them is literally not even reaching the database in the first place (or failing before start execution), and the retries are meant to solve that problem mostly in case of transient errors. Also, if the response failed to come back to us, probably the retry will fail too since it is probably a network issue.

I haven't notice this being an issue in a few accounts I have been spot checking over the past few weeks, but please, if you do notice in your case some billing issue, let us know. We are still monitoring the retry success ratios and considering more improvements to do over time, so any feedback is welcome.

Remember, D1 is not billing on requests, but on actual rows read and written successfully.

Original message was deleted

lambrospetrou•9/12/25, 7:42 AM

I will ping the team about this. Is the beacon_minorbeacon_minor column newly added through the Studio or did it exist before you made changes through the Studio UI?

Original message was deleted

Visal .In•9/12/25, 7:55 AM

Could you provide the schema of your table?

Visal .In•9/12/25, 8:06 AM

I see. Trigger editor support is also on our roadmap for this month. Glad you figured it out! We’ll continue improving the studio experience, and please feel free to share here if you encounter any issues or have suggestions for specific areas of improvement.

Llambrospetrou Fair question we did think about. Short answer is that it should not happen in p...

Ehbraheem•9/12/25, 10:23 AM

Awesome, thank you @lambrospetrou for the detailed response

Perry•9/12/25, 1:29 PM

i m currently using cloudflare workers and Kv , and some other DB just for the sake of GUI, i would love to use Cloudflare D1 is there a way to have some GUI on D1??

PPerry i m currently using cloudflare workers and Kv , and some other DB just for the ...

Brayden•9/12/25, 1:34 PM

There is a Studio (SQL browser) interface inside of Cloudflare for D1. Click the blue "Explore data" button when inside a D1 instance, or the "..." menu next to an item in the resource list page when you first click D1 from the sidebar.

dev.itsarun•9/12/25, 9:28 PM

https://github.com/cloudflare/workers-sdk/issues/10638

Guys Please have a View and Share your Reactions ?

GitHub

Support Exporting D1 Data in JSON & CSV Formats with Encoding Optio...

Describe the solution Currently, wrangler d1 export supports exporting schema and data in SQL format only. It would be very helpful if wrangler could support alternative export formats such as JSON...

Original message was deleted

Praefe•9/12/25, 9:36 PM

Surely someones created a worker template for this by now

what the hell is up with every cloudflare service being in circular dependancy so its just crashes e

Similar Threads

Similar Threads

Similar Threads