Workers Builds! 🙌 can't seem to get Bun working though... even tried setting BUN_VERSION build env
Workers Builds! 🙌 can't seem to get Bun working though... even tried setting BUN_VERSION build env var, tried .tools-versions. Anyone get this working? Also a retry button on failed builds would be very helpful!
16 Replies
reported the same issue here https://discord.com/channels/595317990191398933/1288707773277929492/1288791482257641472
Because it's not widely supported in Workers runtime.
At least, that's what I believe.
any idea when remote dev sessions will be available again? (please ping me if you reply)
adding assets to workers is sick and i've got it working. but
not_found_handling = "single-page-application" doesn't seem to quite work without handling in the worker? Am i right about that? You want to bind it and hand the request back to assets to get the correct asset.Does anyone know how I would go about testing the total response time for a worker from different geographical locations?
The new worker logging is fantastic. The only thing that could make it better would be a way to see a consolidated view across workers.
anyone using prisma? generating the prisma client seems to add 830kb to the size of my worker (30kb without) which seems unreasonable. am I doing something wrong?
What happens if you have multiple parallel worker requests from same colo that make use of the Cache API and there is a Cache Miss, the key/data stored in the Cache API is the same in all requests.
- Do all the worker requests try to hit the origin at once or are things smart enough to so that only one worker request hits the origin and the rest will pull from the Cache?
Also, On a pro account when using the Cache API in a worker, it’s cached in the nearest colo to where the worker is executed from, from what I understand.
My question is there any way to cache in a location closer to using that api on the pro plan?
Also, On a pro account when using the Cache API in a worker, it’s cached in the nearest colo to where the worker is executed from, from what I understand.That's the same for any plan. It's not "the nearest colo to the worker", workers run directly in colos. Cache is all PoP based now though
My question is there any way to cache in a location closer to using that api on the pro plan?no, cache api is only local. What do you mean "closer"? though, closer to what? Maybe you want something like Smart Placement to run the worker closer to your backend? ?crossposting -> https://discord.com/channels/595317990191398933/1289068368116322345/1289068368116322345
Please do not post your question in multiple channels/post it multiple times per the rules at #😃welcome-and-rules. It creates confusion for people trying to help you and doesn't get your issue or question solved any faster.
Very surprised by the worker update; after trying to deploy a worker with assets, it seems that using
_headers files to modify static file request headers is not currently supported, right?
Additionally, trying to add Cache Everything in the Cache Rules does not seem to work either,
The contents of the assets dir always return cache-control: public, max-age=0, must-revalidate
The index document also did not hit the cache and there are no cf-cache-status headersThough note that just like with Pages, assets are cached even if you don’t have the rule
Thank you for your reply. If worekr with assets requires customizing some headers, how should it be handled? For example,
cross-origin-embedder-policy: require-corp