Oh! Is there a way to generate them on demand / generate service bindings for that

if that question isn't answered by https://developers.cloudflare.com/workers/runtime-apis/durable-objects/#object-stubs let us know what would make it clearer

Hey guys! May can anyone help me with this error: In order to use Durable Objects, you must agree to pricing at https://dash.cloudflare.com/$ID/workers/overview?enable-durable-objects [code: 10084] I got enterprise subscription and cannot find anything related to DO enabling in UI.

let's switch to #779390076219686943

DM me your account id. We're hopefully going to remove that check altogether today.

Same error :\

Hi! We're looking to build some type of push system to send messages to our app cliënt in as real-time as possible, considering websockets are priced at wall-time, would the same be the case for server-sent events?

Does anyone has any other recommendations on how to do this on the cheap, without paying +-5 dollars per hour for an connection to be open to a durable object?

Its $4.22/mo for each DO (if active all month) not $5 per hour. But the team has been working on a hibernation API that will sleep the durable object if it isnt actively handling a message from a websocket. No clue when that will actually come out but its been in talks the last few months.

Ahhh okay got my pricing all off then, that's actually pretty decent.

Are there any similar libraries like DOG that came out? https://github.com/cloudflare/dog

Not that I am aware of

Do we know the minimum storage of a DO, or rather, how many bytes a DO takes even when no data is stored ?

we happen to know it! 0 bytes unless you explicitly use the storage API

The key value data is stored as is right, with no compression

it is serialized with v8's serializer

but no further compression afaik

It might be compressed, but you are billed for what you send

weird situation happened - I am pretty sure I had a DO get "stuck" until I opened

wrangler tail

which finally restarted it. we have a task queue that is triggered by alarms. there was a task in the queue that was stuck for a few hours. in the alarm, it finds stuck tasks and then immediately updates storage to increment a value and then continues on with the work that value was never set and even though I was triggering the alarm it wasn't picking up that the task was "stuck" (which is defined by a field not being updated in 30 minutes but it had been 3 hours) anyways, as soon as I opened

wrangler tail

(which restarts DOs) within a minute it "un-stuck" the item in the queue and updated the storage as expected

That code happens to be open source so you can see how it is counted https://github.com/cloudflare/workerd/blob/main/src/workerd/api/actor-state.c%2B%2B

Got it thanks !

if you have a DO id (not the name you passed to idFromName, the 64 character hex id) for that DO instance, DM it to me

I only have the name... one sec let me see what I can do

given a name, OBJECT_NAMESPACE.idFromName is deterministic for a given namespace so you'd just need the result of calling .toString on that return value

working on this, a bit confusing with itty-durable

I am new to DO's. I am trying to schedule a mail channels email using DO alarms... I have this code: https://codeshare.io/N36zBd Is this the best way to do it? And what would be DO cost of this be per email scheduled and then triggered? Is there a way to calculate that?

DMd 🙂

Couple of things I noticed - scheduler.fetch needs to be passed a Request object or a valid url (even though the domain doesn't matter). storage.setAlarm takes a single argument (Date or integer millis) not a pair of arguments. Making a new DO for every email you send is an interesting choice, do you know how frequently you're going to be sending them?

Speaking of creating new DOs... > Error: Your account is generating too much load on Durable Objects. Please back off and try again later. > There is a limit on how quickly you can create new objects or lookup different existing objects Is there any specific limit we can actually judge if we will be creating/looking up too many DOs?

How have I not hit that yet lol?