Ministack: Free, open-source local AWS emulator - 60+ services, Terraform compatible, real databases. Free forever. MIT licensed.
Python awsaws-emulatoraws-localaws-sdkdevtoolsdockerdynamodbec2emulatorlambdalocalstacklocalstack-alternativeministackmock-awsopen-sourcepythons3sqsterraform
release v1.4.21 (#1457)
Nahuel Nucera529dfd7827af
diff --git a/CHANGELOG.md b/CHANGELOG.mdindex 7d2e0483..c44d3ed9 100644--- a/CHANGELOG.md+++ b/CHANGELOG.md@@ -7,17 +7,22 @@ Versioning follows [Semantic Versioning](https://semver.org/).## [Unreleased]-### Fixed-- **Step Functions — an unimplemented optimized service integration now fails the execution instead of silently succeeding** — a Task using an `arn:aws:states:::<service>:<action>` integration MiniStack does not implement fell through to echoing its input back as a `SUCCEEDED` result, so a state machine whose side effect never ran still reported success. Such a task now fails with `States.Runtime` naming the unimplemented resource; implemented integrations and the generic `aws-sdk:` form are unaffected. Reported by @iwasakar.-- **CloudFormation — an update that would replace a custom-named resource is refused instead of silently executing and destroying data** — a stack update requiring replacement of a resource with an explicit physical name (for example changing a DynamoDB key attribute's type on a table that sets `TableName`) executed the replacement, reported `UPDATE_COMPLETE`, and dropped the table's rows. Such an update now fails and rolls back to `UPDATE_ROLLBACK_COMPLETE` with the message real CloudFormation uses (`CloudFormation cannot update a stack when a custom-named resource requires replacing. Rename <name> and update the stack again.`), leaving the resource and its data untouched; renaming the resource is still the way to force the replacement. Reported by @iot-rocket.-- **CloudFormation — change sets report a valid status, fail when their execution fails, and do not outlive their stack** — `ExecuteChangeSet` wrote `EXECUTE_COMPLETE` into the change set's `Status` (a ChangeSetStatus, whose values are only `CREATE_*`/`DELETE_*`/`FAILED`) and did so at schedule time, so the CDK rejected the change set as not ready (it gates on `Status == CREATE_COMPLETE`) and a failed execution never surfaced. `Status` now stays `CREATE_COMPLETE`, and `ExecutionStatus` moves to `EXECUTE_COMPLETE` or `EXECUTE_FAILED` based on the deployment outcome. Deleting a stack now removes its change sets, and a duplicate change-set name is refused with `AlreadyExistsException`, so a leftover executed change set no longer shadows a same-named one on a re-created stack. A change set with no changes now ends `FAILED` with the reason real CloudFormation gives (`The submitted information didn't contain changes. Submit different information to create a change set.`) instead of reporting `CREATE_COMPLETE`, and executing a change set deletes the stack's other now-invalid change sets. A missing change set now returns the AWS `ChangeSetNotFound` error (404) instead of a non-AWS `ChangeSetNotFoundException` (400), and a direct `UpdateStack` marks the stack's pending change sets `OBSOLETE`. Reported by @iot-rocket.-- **RDS — Aurora engine versions are validated on non-create writes and global inheritance** — `ModifyDBInstance`, `ModifyDBCluster`, `CreateGlobalCluster`, and `ModifyGlobalCluster` stored arbitrary Aurora engine versions that the shared engine catalog did not advertise, and `CreateDBCluster` could inherit such a version from legacy global-cluster state even though supplying it explicitly was rejected. All paths now use the shared catalog validator and reject unknown versions before mutating any state: create paths keep the create-time `Cannot find version {version} for {engine}`, while the modify paths report the upgrade-target shape real AWS uses (`InvalidParameterCombination` / `Cannot find upgrade target from {current} with requested version {requested}.`, as seen in real `ModifyDBCluster` transcripts). `ModifyGlobalCluster` now also propagates an accepted version to every member cluster the way real AWS applies global-database version changes, and `ModifyDBCluster` refuses to move a member cluster to a different major than its global cluster — the combination `CreateDBCluster` already refuses to produce. Contributed by @kiran01bm.+## [1.4.21] — 2026-08-20+### Added-- **RDS — replicated Aurora PostgreSQL readers survive StopDBCluster / StartDBCluster and warm boot** — stopping a cluster with replicating readers was refused with `InvalidDBClusterStateFault` (the follow-up slice of #1325 the reader feature shipped without), and a MiniStack restart demoted a persisted reader to an alias of the writer's shared container. `StopDBCluster` now stops the writer's container and every reader-owned standby container — a stop that fails on any of them rolls back what it stopped and reports the failure instead of publishing `stopped` — and `StartDBCluster` restarts the writer, waits for it to accept authenticated connections, then revives each reader by re-cloning from the writer at its current address, keeping the member's identity, endpoint, and volume name (reader compute owns no durable state, as on Aurora, so revival recreates the container rather than restarting it). The same revival runs on warm boot. A revival that genuinely fails lands the cluster back on `stopped` for retry; only when compute deliberately cannot come back (control-plane-only restore, or the replication flag withdrawn between runs) does the member fall back to aliasing the writer, releasing its former container and volume, instead of advertising an endpoint nothing serves. Contributed by @kiran01bm.+- **IoT Core — native mTLS MQTT listener on port 8883** — the embedded broker now also accepts MQTT over TLS on 8883 (`IOT_MTLS_ENABLED=0` turns it off, `IOT_MTLS_PORT` moves it), on by default when `cryptography` is present, so AWS IoT Device SDK binaries can connect. The broker certificate comes from the local CA (`GET /_ministack/iot/ca.pem`); a client certificate is optional (none is served under `MINISTACK_ACCOUNT_ID`, like an unsigned WebSocket upgrade) and an unknown or non-`ACTIVE` certificate is refused with a `0x05` CONNACK. Contributed by @iot-rocket.+- **RDS — replicated Aurora PostgreSQL readers survive StopDBCluster / StartDBCluster and warm boot** — under `MINISTACK_RDS_PG_CLUSTER_REPLICATION`, `StopDBCluster` now stops the reader containers alongside the writer and `StartDBCluster` revives each reader by re-cloning from the writer; the same revival runs on warm boot, so a persisted reader comes back as a real hot standby instead of being demoted to a writer alias. Contributed by @kiran01bm.+- **CloudFormation — `AWS::SES::ConfigurationSet` and `AWS::SES::ConfigurationSetEventDestination`** — a template carrying an SES configuration set no longer fails with "Unsupported resource type"; both resource types provision (registered in the classic and v2 SES stores, `Ref` returns the set name, the event destination round-trips), CloudFormation-provisioning fidelity only. Contributed by @ryan-bennett.+### Fixed-- **RDS — global Aurora stop/start preserves topology and MySQL replication** — `StopDBCluster` and `StartDBCluster` allowed individual members of a multi-cluster Aurora global database to transition independently, and deleting a primary's last instance stopped shared compute that secondaries still needed. Starts and stops are now limited to sole-member global databases (`InvalidDBClusterStateFault`, 400), last-instance deletion preserves compute while other global members exist, and a successful recreate resets stale MySQL replica state before reconfiguring or deferring replication after readiness without overwriting a deliberate legacy-volume failure. Contributed by @kiran01bm.-### Added-- **IoT Core — native mTLS MQTT listener on port 8883** — the embedded broker spoke MQTT only over WebSocket, so AWS IoT Device SDK binaries could not connect. The broker now also listens for MQTT over TLS on port 8883 (`IOT_MTLS_ENABLED=0` turns it off, `IOT_MTLS_PORT` moves it), on by default when `cryptography` is available, configured exactly like the Transfer Family SFTP listener. The broker certificate is minted from the local CA (`GET /_ministack/iot/ca.pem`). The client certificate is optional and follows the rule S3 applies to a presigned URL: a client that presents none is served under `MINISTACK_ACCOUNT_ID`, exactly like an unsigned MQTT-over-WebSocket upgrade, while one that presents a certificate has it read against the registry. Registered `ACTIVE` in exactly one account serves that account and region; unknown, not `ACTIVE`, or `ACTIVE` in several accounts is refused with a "not authorized" CONNACK (0x05, reason code 0x87 for MQTT 5 clients), so `UpdateCertificate` to `INACTIVE` cuts a device off the way it does on AWS. Contributed by @iot-rocket.+- **API Gateway (REST) — routing is on the resource+method pair, not the resource alone** — a request a matched resource does not serve (a methodless intermediate node, a CORS-preflight-only node, or an undeclared verb) now falls through to a `{proxy+}` elsewhere in the tree as it does on AWS; only an exact resource+method match keeps the request, and routing precedes authorization. Supersedes the 1.4.16 change that answered a methodless resource `403`. Contributed by @iot-rocket.+- **Step Functions — an unimplemented optimized service integration fails instead of silently succeeding** — a Task using an `arn:aws:states:::<service>:<action>` integration MiniStack does not implement fell through to echoing its input back as `SUCCEEDED`; it now fails with `States.Runtime` naming the unimplemented resource. Reported by @iwasakar.+- **CloudFormation — a replacement of a custom-named resource is refused instead of destroying data** — an update requiring replacement of a resource with an explicit physical name (for example changing a DynamoDB key attribute's type on a table that sets `TableName`) now fails and rolls back to `UPDATE_ROLLBACK_COMPLETE` with `CloudFormation cannot update a stack when a custom-named resource requires replacing. Rename <name> and update the stack again.`, leaving the resource and its data intact. Reported by @iot-rocket.+- **CloudFormation — change sets report a valid status, fail when execution fails, and do not outlive their stack** — `ExecuteChangeSet` wrote the invalid `EXECUTE_COMPLETE` into `Status`, breaking the CDK's `Status == CREATE_COMPLETE` gate. `Status` now stays `CREATE_COMPLETE` while `ExecutionStatus` moves to `EXECUTE_COMPLETE` or `EXECUTE_FAILED` on the deployment outcome. A no-change set ends `FAILED`, a missing set returns `ChangeSetNotFound` (404), a duplicate name is `AlreadyExistsException`, deleting a stack removes its change sets, executing one deletes the others, and a direct `UpdateStack` marks pending sets `OBSOLETE`. Reported by @iot-rocket.+- **RDS — Aurora engine versions are validated on non-create writes and global inheritance** — `ModifyDBInstance`, `ModifyDBCluster`, `CreateGlobalCluster`, `ModifyGlobalCluster`, and global-inherited `CreateDBCluster` now reject engine versions the catalog does not advertise; modify paths return `InvalidParameterCombination` / `Cannot find upgrade target from {current} with requested version {requested}.`, `ModifyGlobalCluster` propagates an accepted version to every member, and a member moved to a different major than its global is refused. Contributed by @kiran01bm.+- **RDS — global Aurora stop/start preserves topology and MySQL replication** — `StopDBCluster` and `StartDBCluster` are now limited to sole-member global databases (`InvalidDBClusterStateFault`, 400), deleting a primary's last instance preserves compute other global members still need, and a successful recreate resets stale MySQL replica state before re-linking replication. Contributed by @kiran01bm.+- **S3 — a conditional delete of an absent key is answered correctly** — `DeleteObject` with `If-Match` on a key that is not there now returns `204` (deleting an already-gone key is done), and `If-Match: *` is honored as an existence check — it holds against any present object and returns `412 PreconditionFailed` when the key is absent — rather than being compared as a literal ETag. Contributed by @gaul.+- **S3 — multipart uploads carry their parts' checksums through to a composite** — `CreateMultipartUpload` records the checksum algorithm, `UploadPart` validates and echoes each part's checksum, and completion builds the AWS composite (`<digest>-<parts>`, the digest of the parts' digests) read back through `ChecksumMode` as `COMPOSITE`; a mismatch is `BadDigest`. Contributed by @gaul.## [1.4.20] — 2026-08-19diff --git a/pyproject.toml b/pyproject.tomlindex d53e33eb..c0438e59 100644--- a/pyproject.toml+++ b/pyproject.toml@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"[project]name = "ministack"-version = "1.4.20"+version = "1.4.21"description = "Free, open-source local AWS emulator — drop-in LocalStack replacement"readme = "README.md"license = { file = "LICENSE" }