This is an automated email from the ASF dual-hosted git repository. github-merge-queue[bot] pushed a commit to branch gh-readonly-queue/main/pr-3273-c9a25b23b7645f3af0e9dfd7e85d8cf646901c55 in repository https://gitbox.apache.org/repos/asf/iceberg-rust.git
commit 119fad92c891c95a7819d87e4195407b3f7c0578 Author: Oleks V <[email protected]> AuthorDate: Thu Sep 24 18:51:51 2026 +0000 ci: replace MinIO with RustFS for integration tests (#3273) * ci: replace MinIO with RustFS for integration tests `quay.io/minio/minio` no longer allows anonymous pulls, so `make docker-up` fails and the integration tests cannot start. DataFusion hit the same failure and switched to RustFS in apache/datafusion#25706. PyIceberg moved to RustFS earlier in apache/iceberg-python#3928. Run `rustfs/rustfs:1.0.0` as the shared S3 service instead. It keeps the `admin`/`password` credentials, the 9000/9001 ports, and virtual-hosted-style access through `RUSTFS_SERVER_DOMAINS` and the bucket aliases. The healthcheck uses `/health/ready`, which waits for storage and IAM. The `mc` bucket setup becomes `curl --aws-sigv4` run from the RustFS image, and the public bucket policy is dropped because no test reads anonymously. Point the REST fixture, HMS, and Spark at `http://rustfs:9000`, and rename the MinIO-specific test helpers and the `ICEBERG_TEST_MINIO_ENDPOINT` override to RustFS. Closes #3272 * ci: use vendor-neutral object-store naming for S3 test service Rename the RustFS service, hostname, bucket aliases, test helpers and env var to a generic object-store name (matching PyIceberg), so future backend swaps only need an image change. Co-authored-by: Copilot App <[email protected]> * ci: fail fast and gate readiness on test bucket creation Treat any non-200/409 response as a failure instead of ignoring it, and add a healthcheck so `docker compose up --wait` blocks until the buckets exist. Co-authored-by: Copilot App <[email protected]> * ci: remove orphan containers on docker-up Renaming the minio service leaves stale containers holding port 9000 for anyone with a previously started stack; clean them up automatically. Co-authored-by: Copilot App <[email protected]> * ci: make create-buckets a one-shot job RustFS returns 200 when re-creating an existing bucket, so plain chained curl -f calls are idempotent. Replace the status-code parsing, marker-file healthcheck and tail with a one-shot container, and gate spark-iceberg on its successful completion. Co-authored-by: Copilot App <[email protected]> --------- Co-authored-by: ovoievodin <[email protected]> Co-authored-by: Kevin Liu <[email protected]> Co-authored-by: Copilot App <[email protected]> --- Makefile | 2 +- crates/catalog/hms/tests/hms_catalog_test.rs | 6 +-- crates/catalog/loader/tests/common/mod.rs | 11 ++-- crates/integration_tests/src/lib.rs | 6 +-- crates/storage/opendal/tests/file_io_s3_test.rs | 20 ++++---- .../opendal/tests/resolving_storage_test.rs | 16 +++--- crates/test_utils/src/lib.rs | 14 +++--- dev/docker-compose.yaml | 58 +++++++++------------- dev/hms/core-site.xml | 2 +- dev/spark/spark-defaults.conf | 2 +- website/src/reference/container-runtimes.md | 2 +- 11 files changed, 65 insertions(+), 74 deletions(-) diff --git a/Makefile b/Makefile index d27893390..5371c4228 100644 --- a/Makefile +++ b/Makefile @@ -112,7 +112,7 @@ site: install-mdbook # Docker targets for integration tests docker-up: - docker compose -f dev/docker-compose.yaml up -d --build --wait + docker compose -f dev/docker-compose.yaml up -d --build --wait --remove-orphans docker-down: docker compose -f dev/docker-compose.yaml down -v --remove-orphans --timeout 0 diff --git a/crates/catalog/hms/tests/hms_catalog_test.rs b/crates/catalog/hms/tests/hms_catalog_test.rs index d0e6486ad..b64184ce7 100644 --- a/crates/catalog/hms/tests/hms_catalog_test.rs +++ b/crates/catalog/hms/tests/hms_catalog_test.rs @@ -33,7 +33,7 @@ use iceberg_catalog_hms::{ HmsCatalog, HmsCatalogBuilder, THRIFT_TRANSPORT_BUFFERED, }; use iceberg_storage_opendal::OpenDalStorageFactory; -use iceberg_test_utils::{get_hms_endpoint, get_minio_endpoint, set_up}; +use iceberg_test_utils::{get_hms_endpoint, get_object_store_endpoint, set_up}; use tokio::time::sleep; use tracing::info; @@ -43,7 +43,7 @@ async fn get_catalog() -> HmsCatalog { set_up(); let hms_endpoint = get_hms_endpoint(); - let minio_endpoint = get_minio_endpoint(); + let object_store_endpoint = get_object_store_endpoint(); let props = HashMap::from([ (HMS_CATALOG_PROP_URI.to_string(), hms_endpoint), @@ -55,7 +55,7 @@ async fn get_catalog() -> HmsCatalog { HMS_CATALOG_PROP_WAREHOUSE.to_string(), "s3a://warehouse/hive".to_string(), ), - (S3_ENDPOINT.to_string(), minio_endpoint), + (S3_ENDPOINT.to_string(), object_store_endpoint), (S3_ACCESS_KEY_ID.to_string(), "admin".to_string()), (S3_SECRET_ACCESS_KEY.to_string(), "password".to_string()), (S3_REGION.to_string(), "us-east-1".to_string()), diff --git a/crates/catalog/loader/tests/common/mod.rs b/crates/catalog/loader/tests/common/mod.rs index 2b5033e6f..0b7e22cea 100644 --- a/crates/catalog/loader/tests/common/mod.rs +++ b/crates/catalog/loader/tests/common/mod.rs @@ -50,7 +50,8 @@ use iceberg_catalog_sql::{ }; use iceberg_storage_opendal::OpenDalStorageFactory; use iceberg_test_utils::{ - get_glue_endpoint, get_hms_endpoint, get_minio_endpoint, get_rest_catalog_endpoint, set_up, + get_glue_endpoint, get_hms_endpoint, get_object_store_endpoint, get_rest_catalog_endpoint, + set_up, }; use sqlx::migrate::MigrateDatabase; use tempfile::TempDir; @@ -230,7 +231,7 @@ async fn rest_catalog(kms_client_factory: Arc<dyn KmsClientFactory>) -> RestCata async fn glue_catalog(kms_client_factory: Arc<dyn KmsClientFactory>) -> GlueCatalog { let glue_endpoint = get_glue_endpoint(); - let minio_endpoint = get_minio_endpoint(); + let object_store_endpoint = get_object_store_endpoint(); let props = HashMap::from([ (AWS_ACCESS_KEY_ID.to_string(), "my_access_id".to_string()), @@ -239,7 +240,7 @@ async fn glue_catalog(kms_client_factory: Arc<dyn KmsClientFactory>) -> GlueCata "my_secret_key".to_string(), ), (AWS_REGION_NAME.to_string(), "us-east-1".to_string()), - (S3_ENDPOINT.to_string(), minio_endpoint), + (S3_ENDPOINT.to_string(), object_store_endpoint), (S3_ACCESS_KEY_ID.to_string(), "admin".to_string()), (S3_SECRET_ACCESS_KEY.to_string(), "password".to_string()), (S3_REGION.to_string(), "us-east-1".to_string()), @@ -279,7 +280,7 @@ async fn glue_catalog(kms_client_factory: Arc<dyn KmsClientFactory>) -> GlueCata async fn hms_catalog(kms_client_factory: Arc<dyn KmsClientFactory>) -> HmsCatalog { let hms_endpoint = get_hms_endpoint(); - let minio_endpoint = get_minio_endpoint(); + let object_store_endpoint = get_object_store_endpoint(); let props = HashMap::from([ (HMS_CATALOG_PROP_URI.to_string(), hms_endpoint), @@ -291,7 +292,7 @@ async fn hms_catalog(kms_client_factory: Arc<dyn KmsClientFactory>) -> HmsCatalo HMS_CATALOG_PROP_WAREHOUSE.to_string(), "s3a://warehouse/hive".to_string(), ), - (S3_ENDPOINT.to_string(), minio_endpoint), + (S3_ENDPOINT.to_string(), object_store_endpoint), (S3_ACCESS_KEY_ID.to_string(), "admin".to_string()), (S3_SECRET_ACCESS_KEY.to_string(), "password".to_string()), (S3_REGION.to_string(), "us-east-1".to_string()), diff --git a/crates/integration_tests/src/lib.rs b/crates/integration_tests/src/lib.rs index feafa3ae9..c137b46f8 100644 --- a/crates/integration_tests/src/lib.rs +++ b/crates/integration_tests/src/lib.rs @@ -22,7 +22,7 @@ use iceberg::io::{ S3_ACCESS_KEY_ID, S3_ENDPOINT, S3_PATH_STYLE_ACCESS, S3_REGION, S3_SECRET_ACCESS_KEY, }; use iceberg_catalog_rest::REST_CATALOG_PROP_URI; -use iceberg_test_utils::{get_minio_endpoint, get_rest_catalog_endpoint, set_up}; +use iceberg_test_utils::{get_object_store_endpoint, get_rest_catalog_endpoint, set_up}; /// Global test fixture that uses environment-based configuration. /// This assumes Docker containers are started externally (e.g., via `make docker-up`). @@ -39,11 +39,11 @@ impl GlobalTestFixture { set_up(); let rest_endpoint = get_rest_catalog_endpoint(); - let minio_endpoint = get_minio_endpoint(); + let object_store_endpoint = get_object_store_endpoint(); let catalog_config = HashMap::from([ (REST_CATALOG_PROP_URI.to_string(), rest_endpoint), - (S3_ENDPOINT.to_string(), minio_endpoint), + (S3_ENDPOINT.to_string(), object_store_endpoint), (S3_ACCESS_KEY_ID.to_string(), "admin".to_string()), (S3_SECRET_ACCESS_KEY.to_string(), "password".to_string()), (S3_REGION.to_string(), "us-east-1".to_string()), diff --git a/crates/storage/opendal/tests/file_io_s3_test.rs b/crates/storage/opendal/tests/file_io_s3_test.rs index 6b57f8660..ac6342ddd 100644 --- a/crates/storage/opendal/tests/file_io_s3_test.rs +++ b/crates/storage/opendal/tests/file_io_s3_test.rs @@ -32,19 +32,19 @@ mod tests { use iceberg_storage_opendal::{ AwsCredential, CustomAwsCredentialLoader, OpenDalStorageFactory, ProvideCredential, }; - use iceberg_test_utils::{get_minio_endpoint, normalize_test_name_with_parts, set_up}; + use iceberg_test_utils::{get_object_store_endpoint, normalize_test_name_with_parts, set_up}; use reqsign_core::Context; async fn get_file_io() -> FileIO { set_up(); - let minio_endpoint = get_minio_endpoint(); + let object_store_endpoint = get_object_store_endpoint(); FileIOBuilder::new(Arc::new(OpenDalStorageFactory::S3 { customized_credential_load: None, })) .with_props(vec![ - (S3_ENDPOINT, minio_endpoint), + (S3_ENDPOINT, object_store_endpoint), (S3_ACCESS_KEY_ID, "admin".to_string()), (S3_SECRET_ACCESS_KEY, "password".to_string()), (S3_REGION, "us-east-1".to_string()), @@ -138,7 +138,7 @@ mod tests { Self { credential } } - fn new_minio() -> Self { + fn new_object_store() -> Self { Self::new(Some(AwsCredential { access_key_id: "admin".to_string(), secret_access_key: "password".to_string(), @@ -162,7 +162,7 @@ mod tests { #[test] fn test_custom_aws_credential_loader_instantiation() { // Test creating CustomAwsCredentialLoader with mock loader - let mock_loader = MockCredentialLoader::new_minio(); + let mock_loader = MockCredentialLoader::new_object_store(); let custom_loader = CustomAwsCredentialLoader::new(mock_loader); // Test that the loader can be used in FileIOBuilder with OpenDalStorageFactory @@ -182,17 +182,17 @@ mod tests { let _file_io = get_file_io().await; // Create a mock credential loader - let mock_loader = MockCredentialLoader::new_minio(); + let mock_loader = MockCredentialLoader::new_object_store(); let custom_loader = CustomAwsCredentialLoader::new(mock_loader); - let minio_endpoint = get_minio_endpoint(); + let object_store_endpoint = get_object_store_endpoint(); // Build FileIO with custom credential loader via OpenDalStorageFactory let file_io_with_custom_creds = FileIOBuilder::new(Arc::new(OpenDalStorageFactory::S3 { customized_credential_load: Some(custom_loader), })) .with_props(vec![ - (S3_ENDPOINT, minio_endpoint), + (S3_ENDPOINT, object_store_endpoint), (S3_REGION, "us-east-1".to_string()), (S3_PATH_STYLE_ACCESS, "true".to_string()), ]) @@ -213,14 +213,14 @@ mod tests { let mock_loader = MockCredentialLoader::new(None); let custom_loader = CustomAwsCredentialLoader::new(mock_loader); - let minio_endpoint = get_minio_endpoint(); + let object_store_endpoint = get_object_store_endpoint(); // Build FileIO with custom credential loader via OpenDalStorageFactory let file_io_with_custom_creds = FileIOBuilder::new(Arc::new(OpenDalStorageFactory::S3 { customized_credential_load: Some(custom_loader), })) .with_props(vec![ - (S3_ENDPOINT, minio_endpoint), + (S3_ENDPOINT, object_store_endpoint), (S3_REGION, "us-east-1".to_string()), (S3_PATH_STYLE_ACCESS, "true".to_string()), ]) diff --git a/crates/storage/opendal/tests/resolving_storage_test.rs b/crates/storage/opendal/tests/resolving_storage_test.rs index ba94b8332..bcc3d5184 100644 --- a/crates/storage/opendal/tests/resolving_storage_test.rs +++ b/crates/storage/opendal/tests/resolving_storage_test.rs @@ -33,16 +33,16 @@ mod tests { S3_SECRET_ACCESS_KEY, }; use iceberg_storage_opendal::OpenDalResolvingStorageFactory; - use iceberg_test_utils::{get_minio_endpoint, normalize_test_name_with_parts, set_up}; + use iceberg_test_utils::{get_object_store_endpoint, normalize_test_name_with_parts, set_up}; fn get_resolving_file_io() -> iceberg::io::FileIO { set_up(); - let minio_endpoint = get_minio_endpoint(); + let object_store_endpoint = get_object_store_endpoint(); FileIOBuilder::new(Arc::new(OpenDalResolvingStorageFactory::new())) .with_props(vec![ - (S3_ENDPOINT, minio_endpoint), + (S3_ENDPOINT, object_store_endpoint), (S3_ACCESS_KEY_ID, "admin".to_string()), (S3_SECRET_ACCESS_KEY, "password".to_string()), (S3_REGION, "us-east-1".to_string()), @@ -268,9 +268,9 @@ mod tests { use reqsign_core::Context; #[derive(Debug)] - struct MinioCredentialLoader; + struct ObjectStoreCredentialLoader; - impl ProvideCredential for MinioCredentialLoader { + impl ProvideCredential for ObjectStoreCredentialLoader { type Credential = AwsCredential; async fn provide_credential( @@ -287,14 +287,14 @@ mod tests { } set_up(); - let minio_endpoint = get_minio_endpoint(); + let object_store_endpoint = get_object_store_endpoint(); let factory = OpenDalResolvingStorageFactory::new() - .with_s3_credential_loader(CustomAwsCredentialLoader::new(MinioCredentialLoader)); + .with_s3_credential_loader(CustomAwsCredentialLoader::new(ObjectStoreCredentialLoader)); let file_io = FileIOBuilder::new(Arc::new(factory)) .with_props(vec![ - (S3_ENDPOINT, minio_endpoint), + (S3_ENDPOINT, object_store_endpoint), (S3_REGION, "us-east-1".to_string()), (S3_PATH_STYLE_ACCESS, "true".to_string()), ]) diff --git a/crates/test_utils/src/lib.rs b/crates/test_utils/src/lib.rs index e44d96c38..b64e46269 100644 --- a/crates/test_utils/src/lib.rs +++ b/crates/test_utils/src/lib.rs @@ -37,24 +37,24 @@ mod common { } // Environment variable names for service endpoints - pub const ENV_MINIO_ENDPOINT: &str = "ICEBERG_TEST_MINIO_ENDPOINT"; + pub const ENV_OBJECT_STORE_ENDPOINT: &str = "ICEBERG_TEST_OBJECT_STORE_ENDPOINT"; pub const ENV_REST_CATALOG_ENDPOINT: &str = "ICEBERG_TEST_REST_ENDPOINT"; pub const ENV_HMS_ENDPOINT: &str = "ICEBERG_TEST_HMS_ENDPOINT"; pub const ENV_GLUE_ENDPOINT: &str = "ICEBERG_TEST_GLUE_ENDPOINT"; pub const ENV_GCS_ENDPOINT: &str = "ICEBERG_TEST_GCS_ENDPOINT"; // Default ports matching dev/docker-compose.yaml - pub const DEFAULT_MINIO_PORT: u16 = 9000; + pub const DEFAULT_OBJECT_STORE_PORT: u16 = 9000; pub const DEFAULT_REST_CATALOG_PORT: u16 = 8181; pub const DEFAULT_HMS_PORT: u16 = 9083; pub const DEFAULT_GLUE_PORT: u16 = 5001; pub const DEFAULT_GCS_PORT: u16 = 4443; - /// Returns the MinIO S3-compatible endpoint. - /// Checks ICEBERG_TEST_MINIO_ENDPOINT env var, otherwise returns localhost default. - pub fn get_minio_endpoint() -> String { - std::env::var(ENV_MINIO_ENDPOINT) - .unwrap_or_else(|_| format!("http://localhost:{DEFAULT_MINIO_PORT}";)) + /// Returns the S3-compatible object store endpoint. + /// Checks ICEBERG_TEST_OBJECT_STORE_ENDPOINT env var, otherwise returns localhost default. + pub fn get_object_store_endpoint() -> String { + std::env::var(ENV_OBJECT_STORE_ENDPOINT) + .unwrap_or_else(|_| format!("http://localhost:{DEFAULT_OBJECT_STORE_PORT}";)) } /// Returns the REST catalog endpoint. diff --git a/dev/docker-compose.yaml b/dev/docker-compose.yaml index 16c999cb9..cb89e3042 100644 --- a/dev/docker-compose.yaml +++ b/dev/docker-compose.yaml @@ -25,53 +25,43 @@ networks: services: # ============================================================================= - # MinIO - S3-compatible storage (shared by all tests) + # Object store - S3-compatible storage backed by RustFS (shared by all tests) # ============================================================================= - minio: - image: quay.io/minio/minio:RELEASE.2025-05-24T17-08-30Z + object-store: + image: rustfs/rustfs:1.0.0 environment: - - MINIO_ROOT_USER=admin - - MINIO_ROOT_PASSWORD=password - - MINIO_DOMAIN=minio - hostname: minio + - RUSTFS_ACCESS_KEY=admin + - RUSTFS_SECRET_KEY=password + - RUSTFS_SERVER_DOMAINS=object-store + hostname: object-store networks: iceberg_test: # Add aliases for virtual-hosted style bucket access aliases: - - icebergdata.minio - - warehouse.minio - - bucket1.minio + - icebergdata.object-store + - warehouse.object-store + - bucket1.object-store ports: - "9000:9000" - "9001:9001" - command: ["server", "/data", "--console-address", ":9001"] healthcheck: - test: ["CMD", "mc", "ready", "local"] + test: ["CMD", "curl", "-f", "http://localhost:9000/health/ready";] interval: 5s timeout: 5s retries: 5 - # MinIO client - creates buckets for tests - mc: + # One-shot job that creates buckets for tests. Re-creating an existing bucket + # returns 200, so this is safe to re-run. + create-buckets: depends_on: - minio: + object-store: condition: service_healthy - image: quay.io/minio/mc:RELEASE.2025-05-21T01-59-54Z - environment: - - AWS_ACCESS_KEY_ID=admin - - AWS_SECRET_ACCESS_KEY=password - - AWS_REGION=us-east-1 + image: rustfs/rustfs:1.0.0 entrypoint: > /bin/sh -c " - /usr/bin/mc alias set minio http://minio:9000 admin password; - /usr/bin/mc mb --ignore-existing minio/icebergdata; - /usr/bin/mc mb --ignore-existing minio/warehouse; - /usr/bin/mc mb --ignore-existing minio/bucket1; - /usr/bin/mc policy set public minio/icebergdata; - /usr/bin/mc policy set public minio/warehouse; - /usr/bin/mc policy set public minio/bucket1; - echo 'Buckets created successfully'; - tail -f /dev/null + curl -fsS --aws-sigv4 aws:amz:us-east-1:s3 -u admin:password -X PUT http://object-store:9000/icebergdata && + curl -fsS --aws-sigv4 aws:amz:us-east-1:s3 -u admin:password -X PUT http://object-store:9000/warehouse && + curl -fsS --aws-sigv4 aws:amz:us-east-1:s3 -u admin:password -X PUT http://object-store:9000/bucket1 " networks: iceberg_test: @@ -89,9 +79,9 @@ services: - CATALOG_URI=jdbc:sqlite:file:/tmp/iceberg_rest.db?journal_mode=WAL - CATALOG_WAREHOUSE=s3://icebergdata/demo - CATALOG_IO__IMPL=org.apache.iceberg.aws.s3.S3FileIO - - CATALOG_S3_ENDPOINT=http://minio:9000 + - CATALOG_S3_ENDPOINT=http://object-store:9000 depends_on: - minio: + object-store: condition: service_healthy networks: iceberg_test: @@ -113,7 +103,7 @@ services: dockerfile: Dockerfile platform: ${DOCKER_DEFAULT_PLATFORM:-linux/amd64} depends_on: - minio: + object-store: condition: service_healthy networks: iceberg_test: @@ -170,8 +160,8 @@ services: depends_on: rest: condition: service_healthy - minio: - condition: service_healthy + create-buckets: + condition: service_completed_successfully environment: - AWS_ACCESS_KEY_ID=admin - AWS_SECRET_ACCESS_KEY=password diff --git a/dev/hms/core-site.xml b/dev/hms/core-site.xml index f23efa888..d88ef9357 100644 --- a/dev/hms/core-site.xml +++ b/dev/hms/core-site.xml @@ -30,7 +30,7 @@ </property> <property> <name>fs.s3a.endpoint</name> - <value>http://minio:9000</value> + <value>http://object-store:9000</value> </property> <property> <name>fs.s3a.access.key</name> diff --git a/dev/spark/spark-defaults.conf b/dev/spark/spark-defaults.conf index b576f624b..a02640bbd 100644 --- a/dev/spark/spark-defaults.conf +++ b/dev/spark/spark-defaults.conf @@ -23,7 +23,7 @@ spark.sql.catalog.rest.type rest spark.sql.catalog.rest.uri http://rest:8181 spark.sql.catalog.rest.io-impl org.apache.iceberg.aws.s3.S3FileIO spark.sql.catalog.rest.warehouse s3://warehouse/rest/ -spark.sql.catalog.rest.s3.endpoint http://minio:9000 +spark.sql.catalog.rest.s3.endpoint http://object-store:9000 spark.sql.catalog.rest.cache-enabled false spark.sql.defaultCatalog rest diff --git a/website/src/reference/container-runtimes.md b/website/src/reference/container-runtimes.md index 7b4ac3851..412aa7005 100644 --- a/website/src/reference/container-runtimes.md +++ b/website/src/reference/container-runtimes.md @@ -19,7 +19,7 @@ # Container Runtimes -Iceberg-rust uses containers for integration tests, where `docker` and `docker compose` start containers for MinIO and various catalogs. You can use any of the following container runtimes. +Iceberg-rust uses containers for integration tests, where `docker` and `docker compose` start containers for an S3-compatible object store (RustFS) and various catalogs. You can use any of the following container runtimes. ## Docker Desktop
