https://discord.cloudflare.com logo
Join Discord
Powered by
# r2
  • e
    if you use 
    caches.default
    , youre using the CDN cache and it can be purged in the same manner
  • d
    it actually works when using --metadata then it also copies custom metadata, the problem is the mtime is incorrect and I currently rely on that
  • d
    but I assume that's impossible to 'backdate' an object in R2 and it will always be the upload time
  • d
    You have to enable public url or custom domain to access an image on R2 directly (without invoking a worker) right?
  • d
    And how would you lock it down to prevent external access to said image, but to allow it from a cloudflare pages site?
  • d
    Would a WAF on Referer be reliable enough?
  • e
    block on invalid Referers will prevent it being embedded on other sites, especially if you add in normal security headers (w/ Transform Rules)
  • e
    but it wont prevent someone outside a browser from requesting it by just adding the header
  • m
    Hello, i am trying to import my s3 buckets, but i do read about sip by sip migration (the one i want to do), but i not see any option related. Is this feature of super slorper still not available? any ETA?
  • m
    I swear I saw that somewhere and I'm looking for it too. don't see it here anymore https://developers.cloudflare.com/r2/r2-migrator/
  • m
    appears to still be in private beta
  • d
    How does R2 handle non-ascii characters in customMetadata? Does it (automatically) encode the data like S3 described here?
  • d
    Sharing this here as well - still looking for examples of what people are doing with Workers and R2. https://discord.com/channels/595317990191398933/779390076219686943/1086327992113963018
  • c
    does R2 have webhook onCreate?
  • k
    Nope
  • c
    my android client upload raw audio to r2, I need it to notify my processing server to pull the raw file and process it when it's ready šŸ˜
  • m
    is there anyway to debug bucket lifecycle?
  • c
    do we have ticket on that? šŸ¤”
  • w
    Is it wrong to connect domains to r2 buckets through undocumented API endpoints?
  • k
    More like ā€œnot documented means not stable, so you wonā€™t get any forward notice if we change itā€
  • w
    oh ok, but it shouldn't eg. put the domain in some kind of limbo state where I can't remove it?
  • a
    You can pretty safely use any API endpoint called by the dashboard.
  • e
    You can use it, and chances are itll work today if youre doing it the same way the dashboard does. the difference is that if we change it, we wont give any warning. so one day your integration might break
  • e
    other difference is that since its unsupported, if you have any problems we most likely won't help you. so youre on your own
  • g
    We have a lot of stability issues with R2 uploads. We often get 502 or 503's reported from our users when they upload large files (500-1000MB) from their client with a pre-signed URL. Are there any known issues here and what can we do to prevent this? We used MinIO before and never had any of those issues here.
  • l
    Anyone ever get an Error Code 10001 when trying to manage Custom Domains on R2 Buckets? I've been getting that error for a few hours now, and it worked fine beforeā€¦ šŸ¤·ā€ā™‚ļø
  • e
  • l
    Used today to migrate ~300GB from a couple of bucketsā€¦ Seems to have worked well, but will report if I find any bugs later šŸ™‚
  • l
    this must be it!!
  • l
    Chat support asked me for the HAR of my HTTP requests, but I was already convinced it was a temporary error server-side
1...955956957...1050Latest