is there a way to set `content-type` of already uploaded objects in a bucket? I'd like to avoid re-u

is there a way to set

content-type

content-type

content-type

content-type of already uploaded objects in a bucket? I'd like to avoid re-uploading the files as i have lots of them and their size is large...

AAnu is there a way to set `content-type` of already uploaded objects in a bucket? I'...

Doyin•2/19/24, 8:20 PM

You can manipulate the content type with a worker or any other interfacing code before delivering the object

AAnu is there a way to set `content-type` of already uploaded objects in a bucket? I'...

Erisa•2/19/24, 8:59 PM

From amazon's docs,

After you upload the object, you cannot modify object metadata. The only way to modify object metadata is to make a copy of the object and set the metadata. In the copy operation, set the same object as the source and target.

I suspect the same advice applies to R2

Erisa•2/19/24, 8:59 PM

Basically means CopyObject with the source and target the same should allow you to edit metadata

Adar•2/20/24, 5:24 AM

Praying for event notifications or lifecycle webhooks

. Can't trust clients to give me data and using pre-signed URL to upload files.

mrngm•2/20/24, 1:58 PM

We're looking into some intermittent latency issues on R2 with custom domain (Pro plan). We don't have exact figures (the CF dashboard doesn't show at least), but a colleague captured two requests that took about 15s to complete according to their browser's network inspector. I have 2 Ray-ID's at hand, but cannot really find a way to verify whether this was a browser issue (e.g. too many open connections) or whether this is on Cloudflare's side. Both requests report cache MISS, one was Last-Modified 19 February 2024 (size: ~19KB), one was Last-Modified 6 February 2024 (size: ~9KB).

Mmrngm We're looking into some intermittent latency issues on R2 with custom domain (Pr...

Erisa•2/20/24, 2:04 PM

Can you share the ray IDs with me? They're safe to put in public

EErisa Can you share the ray IDs with me? They're safe to put in public

mrngm•2/20/24, 2:06 PM

of course, 858715f22c0b0e48-AMS858715f22c0b0e48-AMS and 858715f22c0f0e48-AMS858715f22c0f0e48-AMS. If it helps: location hint on the bucket at creation time was Western Europe, and our requests should've come from NL

Erisa•2/20/24, 2:06 PM

Thanks

mrngm•2/20/24, 2:14 PM

I do see a CF incident on R2 just now (13:48Z), both requests were done some 20 minutes before (13:27Z). I also have a few requests around that same timestamp that completed <200ms, ray ID's available ifneedbe.

Mmrngm I do see a CF incident on R2 just now (13:48Z), both requests were done some 20 ...

Erisa•2/20/24, 2:19 PM

From what I can understand it likely would be best to wait until the incident is resolved and if its still happening at that point then I can dig further

Erisa•2/20/24, 2:20 PM

The statuspage is updated manually, there is often times where impact is present or ramping up for a time before we noticed and went through the motions of updating it

EErisa The statuspage is updated manually, there is often times where impact is present...

mrngm•2/20/24, 2:21 PM

Yeah, I figured that as well.

EErisa From what I can understand it likely would be best to wait until the incident is...

mrngm•2/20/24, 2:22 PM

Thanks, I'll keep that in mind and collect Ray ID's when it happens again.

Mmrngm We're looking into some intermittent latency issues on R2 with custom domain (Pr...

geomaster•2/20/24, 3:04 PM

Same issue, we've been facing it for a good 2-3hrs before the incident was acknowledged and are still experiencing the problem

geomaster•2/20/24, 3:05 PM

Total of 5 hours since it started

geomaster•2/20/24, 3:11 PM

No idea. A full 2hours before they acknowledged the incident we opened a support ticket providing 10 ray IDs with high R2 latency and even narrowed down the problem to the 2 EU regions. A large % of our requests faced this, unless we were hit unusually hard compared to others, it's difficult to imagine the impact being so low during that time that it doesn't warrant a status page update

Zora•2/20/24, 3:12 PM

I got a huge spike of HTTP 500 errors here

Zora•2/20/24, 3:12 PM

WEUR bucket

geomaster•2/20/24, 3:13 PM

Interestingly, we didn't get a single 500, only requests always taking a little over 15 seconds. As if an internal service is timing out after that time, but somehow not causing an error?

geomaster•2/20/24, 3:14 PM

Honestly 500s would be better, this way our users are waiting for a loading spinner for 30 seconds because we do 2 R2 writes on the backend... not fun

Ggeomaster No idea. A full 2hours before they acknowledged the incident we opened a support...

Erisa•2/20/24, 3:25 PM

What is the ticket num?

Ggeomaster Interestingly, we didn't get a single 500, only requests always taking a little ...

Erisa•2/20/24, 3:25 PM

It was probably retried internally but 15 seconds is way too long to be acceptable even in error

EErisa What is the ticket num?

geomaster•2/20/24, 3:38 PM

3153662

Erisa•2/20/24, 3:39 PM

Thanks

Ohyzd•2/20/24, 3:53 PM

Yup, still seeing some requests take a long time. Not as many as 2 hours ago though.

Erisa•2/20/24, 3:57 PM

I looked into these further and my understanding is that they are related to the incident, I'm not sure why some saw impact start much earlier however

Scruffy•2/20/24, 4:20 PM

Getting 500 a lot on trying to upload with a signed URL, i'm having to retry my requests up to 10 times

Scruffy•2/20/24, 4:21 PM

We encountered an internal error. Please try again

We encountered an internal error. Please try again

Scruffy•2/20/24, 4:23 PM

Eastern Europe (EEUR)

SScruffy Eastern Europe (EEUR)

Chaika•2/20/24, 4:37 PM

if you haven't already seen: https://www.cloudflarestatus.com/incidents/69rmr1tydn49

R2 degraded availability in EU Region

CChaika if you haven't already seen: https://www.cloudflarestatus.com/incidents/69rmr1ty...

Scruffy•2/20/24, 4:38 PM

well that would explain it

Unsmart•2/20/24, 4:39 PM

seems its probably resolved to me :NODDER:

Unsmart•2/20/24, 4:39 PM

minimal issues in the last hour

Unsmart•2/20/24, 4:40 PM

guess theres still the occasional high latency request but not really any errors

UUnsmart seems its probably resolved to me :NODDER:

Erisa•2/20/24, 4:47 PM

I've added

We have deployed some further fixes and are seeing improvements. We are continuing to monitor for any further issues.

Lone Ranger•2/20/24, 4:50 PM

how to protect videos url from other site?

Lone Ranger•2/20/24, 5:00 PM

how can i get free?

Lone Ranger•2/20/24, 5:00 PM

i don't know coding

Unsmart•2/20/24, 5:01 PM

You cant protect videos from another site realistically there will always be ways around it

Unsmart•2/20/24, 5:02 PM

You dont need to know how to code to setup CORs though you just set what the url to your site is pretty much

EErisa I've added > We have deployed some further fixes and are seeing improvements. W...

Erisa•2/20/24, 5:03 PM

Now marked resolved

UUnsmart You dont need to know how to code to setup CORs though you just set what the url...

Lone Ranger•2/20/24, 5:03 PM

how can you guide me?

LLone Ranger how can you guide me?

Unsmart•2/20/24, 5:04 PM

The documentation for CORS is here: https://developers.cloudflare.com/r2/buckets/cors/

Erisa•2/20/24, 5:04 PM

@mrngm @geomaster @Ohyzd @Scruffy are you still seeing issues? We've closed out the incident as resolved

UUnsmart You cant protect videos from another site realistically there will always be way...

Lone Ranger•2/20/24, 5:05 PM

protect mean off the urls hotlinking to other site

EErisa @mrngm @geomaster @Ohyzd @Scruffy are you still seeing issues? We've closed out ...

Scruffy•2/20/24, 5:06 PM

yes

SScruffy yes

Erisa•2/20/24, 5:07 PM

May be a different issue then - could you share your public account ID?

EErisa May be a different issue then - could you share your public account ID?

Scruffy•2/20/24, 5:08 PM

225af28a5aa74bd199db63641e9b1efd

EErisa @mrngm @geomaster @Ohyzd @Scruffy are you still seeing issues? We've closed out ...

geomaster•2/20/24, 5:08 PM

Still seeing high latencies, also errors this time

Ggeomaster Still seeing high latencies, also errors this time

Erisa•2/20/24, 5:10 PM

Can you reply to the ticket with more information so I can take a look?

is there a way to set `content-type` of already uploaded objects in a bucket? I'd like to avoid re-u

Similar Threads

Similar Threads

Similar Threads