The previous way of initializing the s3 client in each call was
adding a lot of overhead and would tank performance beyond about
20 simultaneous requests.
Since the backend access is through a single account, we can init
and store this client for use from each api call.
This builds on the previous work that sets up the body streaming
for the put object and put part requests. This adds the auth and
checksum readers to postpone the v4auth checks and the content
checksum until the end of the body stream.
This means that the backend with start reading the data from the
body stream before the request is fully validated and signatures
checked. So the backend must check the error returned from the
body reader for the final auth and content checks. The backend
is expected to discard the data upon error.
This should increase performance and reduce memory utilization
to no longer require caching the entire request body in memory
for put object and put part.
This backend redirects incoming requests to another s3 service.
This will use the incoming credentials to setup the client
requests to the external s3 service. So the IAM accounts (or
root account) must match what the external s3 service expects.
Since objects with trailing "/" are mapped to directories in the
posix filesystem, they must not contain data since there is no
place to store that data.
This checks both PutObject and CreateMultipartUpload for invalid
directory object types containing data.
This moves the internal iam service from the posix backend so
that we can start implementing new iam services right in the auth
module.
The internal iam service has same behavior as before, but now
must be enabled with the --iam-dir cli option.
New single user service is the default when no other iam service
is selected. This just runs the gateway in single user mode with
just the root account.
* fix: Object tag actions cleanup
* fix: Fixes#249, Changed ListObjects default max-keys from -1 to 1000
* fix: Fixes#250, Added support to provide a marker not from the objects list and list the objects after the provided marker in ListObjects(V2) actions
* feat: Closes#256, Addded a check step, to compare object metadatas and allow the copying to itself, if the metadata has been changed
* fix: Simplified range assignment in CopyObject posix function
The scoutfs backend is only supported on 64bit linux. This corrects
the build constraints to only supported linux arch, and prevents the
incompatible import for unspported arch.
We also need to adjust the body limit setting on 32 bit since this
is an int, and our default limit will overfow on 32 bit.
