That's right, I meant it's not built into R2 at the moment

i have some challenge & fun ideas i wanna try with r2 & workers on my free time. I always appreaciated the fact that CF share pretty powerful tools and also premium no-brainer stuff. But you can always (there are still the exceptions but..) challenge a bit and obtain premium functionality on free plans. r2 & workers will be on paid plan 🙂

hope you will add it in future. S3 have really strong management.

@Skye btw, have a question. For R2 you can specify a geo zone where you will store the files. When files in europe are requested from someone out of europe (lets say USA), people from USA will get them from USA's endpoints or from european endpoints?

the data & the traffic still stays in europe?

Unless you're using the enterprise data localisation suite, traffic will generally enter cloudflare's network at the closest place to the user (in this case in the US), and travel to the destination within the network where it reaches the exit (in this case your bucket in Europe)

But it would still be stored in Europe of course

yep thats why i was asking (i know about this enterprise-only feature)

so no way lock traffic in europe (r2 & workers) w/o enterprise 😐 thanks!

Are you looking for an EU / FedRAMP jurisdiction that forces your objects to stay in those regions? If so, this is coming pretty soon

Object lifecycles are also close!

How exactly do I do that, can you provide some instructions?..

with eu rules it's highly recommend that traffic stays in eu also. personal data - yep in eu. non-personal data - noone cares. and cause CF will route traffic out of europe and use non-european gateways, some traces of that data could leave europe.

lifecycles are verry verry welcome!

Right yeah, the upcoming jurisdictions "feature" will take care of all this. It is a lot stricter than the location hints we have right now!

watch it be enterprise only

Haha no the EU jurisdiction will be available for everyone!

probably you'll move that service feature out of enterprise plan^^

Hello everyone 👋

I was just wondering what's the best way to do lifecycle management for objects in R2.

As far as I can tell from previous discussions, that's planned but not officially supported yet?

I want all objects older than a day to be automatically deleted, so I set up a worker with a cron trigger that lists all objects in the buckets, checks the upload time, compares it to the current time and deletes any objects that are too old.

Is that the way to go?

As noted, object lifecycles are an upcoming feature that isn't out yet

For now, that sounds like a good way to go about it

Am I likely to run into any limits there? For example if there are 100k objects in the bucket.

Theres a max amount you can list with one call if I remember correctly

if you can catch that and list more, it shouldnt be a problem

Got it. I think there's also a limit of 1000 objects per .delete

Thanks a lot for the help 🙏