Documentation

From the CDN docs: https://docs.digitalocean.com/products/spaces/how-to/enable-cdn/ "You can also use presigned URLs with the Spaces CDN. To do so, configure your SDK or S3 tool to use the non-CDN endpoint, generate a presigned URL for a GetObject request, then modify the hostname in the URL to be the CDN hostname (<space-name>.<region>. cdn.digitaloceanspaces.com , unless the bucket uses a custom hostname)." This only works for "Virtual-hosted style file URL" type urls, it does not work for "Path-style file URL". That tripped me up because I had forcePathStyle: true in my s3 client. I can see questions where other people have likely run into the same issue and it has not been explained correctly. See below: https://www.digitalocean.com/community/questions/s3-cdn-endpoint-does-not-seem-to-work-with-presigned-url-generated-from-application-code https://www.digitalocean.com/community/questions/setting-up-pre-signed-urls-with-custom-domain-name-and-cdn-in-php?comment=191224 Then there is this: "Using presigned URLs does not allow transferred files to be cached when using the Spaces CDN. Attempting to do so may result in double the bandwidth charge without the CDN’s performance benefit." This appears contradictory to the first section. If presigned URLS cannot be cached, does that mean there is zero performance benefit in using presigned urls with the spaces CDN? In my mind, a CDN _is_ an edge cache. In this context, I do not understand the difference between the cache and the CDN itself. I see another person has had the same point of confusion and it hasn't really been fully addressed: https://ideas.digitalocean.com/documentation/p/confusing-phrasing-around-presigned-urls-caching-and-cdns Thanks

https://docs.digitalocean.com/products/monitoring/concepts/metrics/#disk-utilization This article needs to be updated to reflect that disk utilization alerts only consider root storage & does not include attached volumes.

doctl compute ssh-key import ... documentation inconsistent with command usage. It says I'm missing required arguments, but no idea from online docs or help option what that would be. From the doc: Usage: doctl compute ssh-key import <key-name> [flags] In reality: $ doctl compute ssh-key import mykey Error: (ssh-key.import.public-key-file) command is missing required arguments [flags] should be optional, so what other arguments are required???

It should be documented that variable names, such when you do ${my-db.DATABASE_URL} are truncated to 32 chars., i.e. if your db name is too long, you have to use the first 32 chars as the variable name.

In order to migrate seamlessly from HTTP/1.1 to HTTP/2 cleartext ( rationale ), I went through the following steps: I kept my HTTP/1.1 server on port 8080 and added a secondary h2c server on port 8081. The deployment was fine. I edited the app spec via the web frontend to change the service's port from 8080 to 8081 and add protocol: HTTP2 as outlined in the documentation . Unfortunately the deployment failed as it looks like the health check was never successful. Here's what I got from the deployment log: > Sep 12 10:26:59 ERROR failed health checks after 12 attempts with error Readiness probe failed: Get "http://<private-ip-address>:8081/v1/health-check": net/http: HTTP/1.x transport connection broken: malformed HTTP response "\x00\x00\f\x04\x00\x00\x00\x00\x00\x00\x03\x00\x00\x00d\x00\x06\x00\x00@\x00\x00\x00\b\a\x00\x00\x00\x00\x00\x7f\xff\xff\xff\x00\x00\x00\x01" Is it possible the health check feature doesn't support h2c yet?

The "How to Configure CORS on DigitalOcean Spaces" ( https://docs.digitalocean.com/products/spaces/how-to/configure-cors/ ) misses a key moment about purging the cache after adding any headers. Please add this moment to the docs. I spent a lot of hours to figure out that was the problem.

I can't see the GetPreSignedURL function in the list of supported S3 API commands in the official documentation ( https://docs.digitalocean.com/products/spaces/reference/s3-compatibility/ Since this is a common operation and there's related configurations in other areas of the Spaces documentation ( https://docs.digitalocean.com/products/spaces/how-to/set-file-permissions/#presigned-url ), I wonder if there is any way to achieve this kind of operation by coding using the AWS S3 SDK

current docs are giving a deprecated example for config a client with go Here is a update example with the new aws sdk go v2 package config import ( "context" "os" "github.com/aws/aws-sdk-go-v2/config" "github.com/aws/aws-sdk-go-v2/credentials" "github.com/aws/aws-sdk-go-v2/service/s3" ) func InitSpaces() (*s3.Client, error) { spacesKey := os.Getenv("SPACES_KEY") spacesSecret := os.Getenv("SPACES_SECRET") creds := credentials.NewStaticCredentialsProvider(spacesKey, spacesSecret, "") cfg, err := config.LoadDefaultConfig(context.TODO(), config.WithCredentialsProvider(creds), config.WithRegion("us-east-1"), config.WithBaseEndpoint("https://[Bucket Name].nyc3.digitaloceanspaces.com"), ) if err != nil { return nil, err } client := s3.NewFromConfig(cfg, func(o *s3.Options) { o.UsePathStyle = false // Another way to set the endpoint is with o.BaseEndpoint // o.BaseEndpoint = aws.String("[Your Digital Ocean origin endpoint]") }) return client, nil }

The current documentation shows an example using the v1/chat/completions endpoint: "For example, https://inference.do-ai.run/v1/chat/completions is used to ..." This language implies that there are other endpoints, but I could not find a comprehensive list of all endpoints. Please provide one.

See https://docs.digitalocean.com/products/networking/ipv6/how-to/enable/#on-existing-droplets that was linked from my droplets network / enable ipv6 page I had to calculate the the network mask /124 on my own :-) And the file on disk has a lot more " than you example.. Adding that directly to the droplet page would be awesome.