can s3 egress be configured using a role instead of an access key?

Thanks for reaching out, raj. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.

can this be done with the python sdk as well?

Absolutely! You can set up S3 egress using an IAM role with the LiveKit Python SDK. The S3Upload configuration includes the

assume_role_arn

and

assume_role_external_id

fields, which let the SDK smoothly make an AssumeRole API call to take on the role you specify. If you need a hand with this, just let me know! Sources: protobufs/livekit_egress.proto | README.md

the protobuf is updated but is the python-sdk up to date?

Hey there! Just a quick heads-up about the LiveKit Python SDK — it includes the latest protobuf updates for S3 egress configuration, like support for the

assume_role_arn

and

assume_role_external_id

fields. That said, there’s a small hiccup: the type hints file (

.pyi

) doesn’t yet show these fields in the

S3Upload

class definition. Don’t worry though, the functionality should still work just fine! Here are a few tips to keep things smooth: • Double-check that your type hints are up to date by regenerating the

.pyi

file if you need to. • Give the AssumeRole fields a quick test to make sure you can access them without any issues. • Keep an eye out for any SDK updates — they might fix this little inconsistency soon. Sources: livekit/livekit_egress.pb.go | livekit-protocol/livekit/protocol/egress.pyi

since

S3Upload

is a protobuf message, is there any way to send those fields by setting them dynamically on the message?

Absolutely! You can easily set the

assume_role_arn

and

assume_role_external_id

fields dynamically on the

S3Upload

protobuf message in the LiveKit Python SDK. Here’s a quick example to help you out:

from livekit import S3Upload

# Create an S3Upload instance
s3_config = S3Upload(
    access_key="YOUR_ACCESS_KEY",
    secret="YOUR_SECRET",
    assume_role_arn="ARN_OF_THE_ROLE",  # dynamic value
    assume_role_external_id="EXTERNAL_ID"  # dynamic value
)

# Update existing S3Upload instance
s3_config.assume_role_arn = "NEW_ARN"
s3_config.assume_role_external_id = "NEW_EXTERNAL_ID"

These fields are part of the S3Upload message structure, so you can tweak them anytime you need. Hope this helps! Sources: .changeset/big-sloths-accept.md

how do you need to set it to ensure it gets set in the right protobuf field number?

Hey there! To make sure fields like

assume_role_arn

and

assume_role_external_id

have the right protobuf field numbers in the LiveKit Python SDK, here’s a quick guide for you: 1. These fields are set in the protobuf schema with specific numbers:

assume_role_arn

is field 12, and

assume_role_external_id

is field 13. 2. When you’re creating or updating an

S3Upload

instance, just assign values to these fields directly. For example:

from livekit import S3Upload

s3_config = S3Upload(
    assume_role_arn="arn:aws:iam::123456789:role/yourrole",
    assume_role_external_id="your_external_id"
)

1. The protobuf message will take care of mapping these fields to the right numbers when it’s serialized. 2. If you want, you can double-check the serialized output to make sure everything’s assigned correctly. If you have any questions or want me to walk you through anything else, just let me know—I’m here to help!