Overcast is alpha — behaviour and APIs may change between releases. Pin your version and read the changelog before upgrading.

overcast

downloads

One line installs it. Docker images and binaries are below.

The installer picks the binary for your OS and CPU, checks it against the release's SHA256SUMS, and puts it in a directory of your own. It never asks a question and never uses sudo.

>_ install
curl -fsSL https://overcast.sh/install.sh | sh

--slim installs the headless overcastd, --version pins a release, and irm | iex takes the same choices as environment variables. Every flag is on the install page; the scripts are readable at install.sh and install.ps1.

Docker images

The full image serves the AWS-compatible API on port 4566 and the web console on port 4567. The slim image is the emulator alone, for headless CI and local service tests.

Full image
>_ full image with web UI
docker run --rm \
  -p 4566:4566 \
  -p 4567:4567 \
  ghcr.io/overcast-sh/overcast:latest
Slim image
>_ slim CI image
docker run --rm \
  -p 4566:4566 \
  ghcr.io/overcast-sh/overcast-slim:latest

Native binaries

Older versions

Showing assets from v0.0.1-alpha.44 published .

latest changelog

v0.0.1-alpha.44

Published

View on GitHub

11 added · 26 fixed · 7 breaking

Added Breaking apigateway

enforce Lambda TOKEN/REQUEST authorizers (REST v1) and Lambda REQUEST authorizers (HTTP v2)

2 details
  • requests to a method or route with a Lambda authorizer were let through unchecked; they now reach the integration only when the authorizer allows them

  • AWS_IAM authorizers remain unenforced

Migration

make the authorizer function return an Allow policy (or isAuthorized: true) for the requests your tests make, or remove the authorizer from the local deployment

Added s3tables

Amazon S3 Tables: table buckets, namespaces and Iceberg tables, with all 49 modeled operations answering; AWS::S3Tables::TableBucket, Namespace, Table, TableBucketPolicy and TablePolicy provision through it

2 details
  • Each table gets a real --table-s3 warehouse bucket in S3; CreateTable with a schema writes the first Iceberg metadata.json.

  • UpdateTableMetadataLocation is a compare-and-swap on versionToken; maintenance, expiration and replication are stored, never run.

Added glue

partitions (create, get, update, delete, singly and in batches; GetPartitions filters by Expression), UpdateTable with VersionId concurrency and archived table versions, UpdateDatabase and BatchDeleteTable; AWS::Glue::Partition provisions and AWS::Glue::Table keeps every TableInput property (#2064)

2 details
  • the expression subset is comparisons, AND/OR/NOT, IN, BETWEEN, LIKE and IS NULL; anything else is InvalidInputException.

  • a stale VersionId is ConcurrentModificationException; GetTableVersion(s), DeleteTableVersion and BatchDeleteTableVersion read and prune versions.

Full changelog