Skip to content

[Backport Fix to staging-2.2.x]: Prevent Hadoop directory copy failures by correctly setting generation id Preconditions #1466

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
dheerajsngh wants to merge 1 commit into
base: staging-2.2.x
Choose a base branch
from
Draft

[Backport Fix to staging-2.2.x]: Prevent Hadoop directory copy failures by correctly setting generation id Preconditions #1466

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion gcsio/src/main/java/com/google/cloud/hadoop/gcsio/GoogleCloudStorageClientReadChannel.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -554,7 +554,7 @@ private BlobSourceOption[] generateReadOptions(BlobId blobId) {
// To get decoded content
blobReadOptions.add(BlobSourceOption.shouldReturnRawInputStream(false));

if (blobId.getGeneration() != null) {
if (blobId.getGeneration() != null && blobId.getGeneration() > 0) {
blobReadOptions.add(BlobSourceOption.generationMatch(blobId.getGeneration()));
}
if (storageOptions.getEncryptionKey() != null) {
Expand Down
6 changes: 3 additions & 3 deletions gcsio/src/main/java/com/google/cloud/hadoop/gcsio/GoogleCloudStorageReadChannel.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -775,7 +775,7 @@ protected void initMetadata(@Nullable String encoding, long sizeFromMetadata, lo
size = gzipEncoded ? Long.MAX_VALUE : sizeFromMetadata;
checkEncodingAndAccess();

if (resourceId.hasGenerationId()) {
if (resourceId.hasGenerationId() && resourceId.getGenerationId() > 0) {
checkState(
resourceId.getGenerationId() == generation,
"Provided generation (%s) should be equal to fetched generation (%s) for '%s'",
Expand Down Expand Up @@ -1134,7 +1134,7 @@ protected Storage.Objects.Get createDataRequest() throws IOException {
Storage.Objects.Get getData =
storageRequestFactory.objectsGetData(
resourceId.getBucketName(), resourceId.getObjectName());
if (resourceId.hasGenerationId()) {
if (resourceId.hasGenerationId() && resourceId.getGenerationId() > 0) {
Copy link

@gemini-code-assist gemini-code-assist bot Aug 25, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

This condition resourceId.hasGenerationId() && resourceId.getGenerationId() > 0 is also used in initMetadata (line 778) and createMetadataRequest (line 1152). To improve maintainability and reduce duplication, consider extracting this logic into a private helper method, for example shouldUseGenerationId().

getData.setGeneration(resourceId.getGenerationId());
}
return getData;
Expand All @@ -1149,7 +1149,7 @@ protected Storage.Objects.Get createMetadataRequest() throws IOException {
Storage.Objects.Get getMetadata =
storageRequestFactory.objectsGetMetadata(
resourceId.getBucketName(), resourceId.getObjectName());
if (resourceId.hasGenerationId()) {
if (resourceId.hasGenerationId() && resourceId.getGenerationId() > 0) {
getMetadata.setGeneration(resourceId.getGenerationId());
}
return getMetadata;
Expand Down
55 changes: 55 additions & 0 deletions .../src/test/java/com/google/cloud/hadoop/gcsio/GoogleCloudStorageClientReadChannelTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,7 +33,9 @@
import com.google.cloud.hadoop.gcsio.GoogleCloudStorageReadOptions.Fadvise;
import com.google.cloud.hadoop.gcsio.integration.GoogleCloudStorageTestHelper;
import com.google.cloud.hadoop.util.GrpcErrorTypeExtractor;
import com.google.cloud.storage.BlobId;
import com.google.cloud.storage.Storage;
import com.google.cloud.storage.Storage.BlobSourceOption;
import com.google.common.collect.ImmutableList;
import com.google.protobuf.ByteString;
import java.io.EOFException;
Expand Down Expand Up @@ -200,6 +202,59 @@ public void readFullObject() throws IOException {
verifyNoMoreInteractions(fakeReadChannel);
}

@Test
public void read_withPositiveGeneration_usesGenerationMatchPrecondition() throws IOException {
long generation = 12345L;
GoogleCloudStorageItemInfo itemInfo =
GoogleCloudStorageItemInfo.createObject(
new StorageResourceId(V1_BUCKET_NAME, OBJECT_NAME, generation),
/* creationTime= */ 10L,
/* modificationTime= */ 15L,
/* size= */ OBJECT_SIZE,
/* contentType= */ "text/plain",
/* contentEncoding= */ "text",
/* metadata= */ null,
/* contentGeneration= */ generation,
/* metaGeneration= */ 2L,
/* verificationAttributes= */ null);

readChannel = getJavaStorageChannel(itemInfo, DEFAULT_READ_OPTION);
readChannel.read(ByteBuffer.allocate(1));

ArgumentCaptor<BlobSourceOption> optionsCaptor =
ArgumentCaptor.forClass(BlobSourceOption.class);
verify(mockedStorage).reader(any(BlobId.class), optionsCaptor.capture());

assertThat(optionsCaptor.getAllValues()).contains(BlobSourceOption.generationMatch(generation));
}

@Test
public void read_withZeroGeneration_doesNotUseGenerationMatchPrecondition() throws IOException {
long generation = 0L;
GoogleCloudStorageItemInfo itemInfo =
GoogleCloudStorageItemInfo.createObject(
new StorageResourceId(V1_BUCKET_NAME, OBJECT_NAME, generation),
/* creationTime= */ 10L,
/* modificationTime= */ 15L,
/* size= */ OBJECT_SIZE,
/* contentType= */ "text/plain",
/* contentEncoding= */ "text",
/* metadata= */ null,
/* contentGeneration= */ generation,
/* metaGeneration= */ 2L,
/* verificationAttributes= */ null);

readChannel = getJavaStorageChannel(itemInfo, DEFAULT_READ_OPTION);
readChannel.read(ByteBuffer.allocate(1));

ArgumentCaptor<BlobSourceOption> optionsCaptor =
ArgumentCaptor.forClass(BlobSourceOption.class);
verify(mockedStorage).reader(any(BlobId.class), optionsCaptor.capture());

assertThat(optionsCaptor.getAllValues())
.doesNotContain(BlobSourceOption.generationMatch(generation));
}
Comment on lines +231 to +256
Copy link

@gemini-code-assist gemini-code-assist bot Aug 25, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The test read_withZeroGeneration_doesNotUseGenerationMatchPrecondition is very similar to read_withPositiveGeneration_usesGenerationMatchPrecondition. To improve readability and maintainability, you could refactor them into a single helper method that is called by two separate tests, or use a parameterized test. The helper method could take the generation and a boolean indicating whether the precondition should be present as parameters.


@Test
public void fadviseAuto_onForwardRead_switchesToRandom() throws IOException {
int chunkSize = FakeReadChannel.CHUNK_SIZE;
Expand Down
49 changes: 49 additions & 0 deletions gcsio/src/test/java/com/google/cloud/hadoop/gcsio/GoogleCloudStorageReadChannelTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -603,6 +603,55 @@ public void read_gzipEncoded_shouldReadAllBytes() throws IOException {
assertThat(readChannel.size()).isEqualTo(testData.length);
}

@Test
public void read_withPositiveGeneration_usesGenerationMatchPrecondition() throws IOException {
long generation = 12345L;
byte[] testData = {0x01, 0x02, 0x03};
MockHttpTransport transport =
mockTransport(
jsonDataResponse(newStorageObject(BUCKET_NAME, OBJECT_NAME).setGeneration(generation)),
dataResponse(testData));
List<HttpRequest> requests = new ArrayList<>();
Storage storage = new Storage(transport, GsonFactory.getDefaultInstance(), requests::add);
GoogleCloudStorageReadChannel readChannel =
createReadChannel(storage, GoogleCloudStorageReadOptions.builder().build(), generation);

readChannel.read(ByteBuffer.allocate(testData.length));

// The first request is for metadata, the second is for data.
// Verify the data request URL contains the generation parameter.
String mediaRequestUrl = getMediaRequestString(BUCKET_NAME, OBJECT_NAME, generation);
assertThat("GET:" + requests.get(1).getUrl().toString()).isEqualTo(mediaRequestUrl);
}

@Test
public void read_withZeroGeneration_doesNotUseGenerationMatchPrecondition() throws IOException {
long requestedGeneration = 0L;
// The actual object has a real, positive generation ID.
long actualGeneration = 54321L;
byte[] testData = {0x04, 0x05, 0x06};
MockHttpTransport transport =
mockTransport(
jsonDataResponse(
newStorageObject(BUCKET_NAME, OBJECT_NAME).setGeneration(actualGeneration)),
dataResponse(testData));
List<HttpRequest> requests = new ArrayList<>();
Storage storage = new Storage(transport, GsonFactory.getDefaultInstance(), requests::add);
GoogleCloudStorageReadChannel readChannel =
createReadChannel(
storage,
GoogleCloudStorageReadOptions.builder().setFastFailOnNotFoundEnabled(false).build(),
requestedGeneration);

readChannel.read(ByteBuffer.allocate(testData.length));

// The data request URL should not contain a generation parameter,
// because the requested generation was 0. The generation parameter in getMediaRequestString
// will be null when the check for generation > 0 fails.
String mediaRequestUrl = getMediaRequestString(BUCKET_NAME, OBJECT_NAME);
assertThat("GET:" + requests.get(0).getUrl().toString()).isEqualTo(mediaRequestUrl);
}

@Test
public void open_gzipContentEncoding_succeeds_whenContentEncodingSupported() throws Exception {
MockHttpTransport transport =
Expand Down