Package org.apache.kafka.server.log.remote.storage

Interface RemoteLogMetadataManager

All Superinterfaces:

AutoCloseable, Closeable, Configurable

@Evolving
public interface RemoteLogMetadataManager
extends Configurable, Closeable

This interface provides storing and fetching remote log segment metadata with strongly consistent semantics.

This class can be plugged in to Kafka cluster by adding the implementation class as remote.log.metadata.manager.class.name property value. There is an inbuilt implementation backed by topic storage in the local cluster. This is used as the default implementation if remote.log.metadata.manager.class.name is not configured.

remote.log.metadata.manager.class.path property is about the class path of the RemoteLogStorageManager implementation. If specified, the RemoteLogStorageManager implementation and its dependent libraries will be loaded by a dedicated classloader which searches this class path before the Kafka broker class path. The syntax of this parameter is same with the standard Java class path string.

remote.log.metadata.manager.listener.name property is about listener name of the local broker to which it should get connected if needed by RemoteLogMetadataManager implementation. When this is configured all other required properties can be passed as properties with prefix of 'remote.log.metadata.manager.listener.

"cluster.id", "broker.id" and all other properties prefixed with "remote.log.metadata." are passed when Configurable.configure(Map) is invoked on this instance.

Method Summary

Modifier and Type	Method	Description
void	addRemoteLogSegmentMetadata(RemoteLogSegmentMetadata remoteLogSegmentMetadata)	Adds RemoteLogSegmentMetadata with the containing RemoteLogSegmentId into RemoteLogMetadataManager.
Optional<Long>	highestOffsetForEpoch(TopicIdPartition topicIdPartition, int leaderEpoch)	Returns the highest log offset of topic partition for the given leader epoch in remote storage.
Iterator<RemoteLogSegmentMetadata>	listRemoteLogSegments(TopicIdPartition topicIdPartition)	Returns all the remote log segment metadata of the given topicIdPartition.
Iterator<RemoteLogSegmentMetadata>	listRemoteLogSegments(TopicIdPartition topicIdPartition, int leaderEpoch)	Returns iterator of remote log segment metadata, sorted by RemoteLogSegmentMetadata.startOffset() in ascending order which contains the given leader epoch.
void	onPartitionLeadershipChanges(Set<TopicIdPartition> leaderPartitions, Set<TopicIdPartition> followerPartitions)	This method is invoked only when there are changes in leadership of the topic partitions that this broker is responsible for.
void	onStopPartitions(Set<TopicIdPartition> partitions)	This method is invoked only when the topic partitions are stopped on this broker.
void	putRemotePartitionDeleteMetadata(RemotePartitionDeleteMetadata remotePartitionDeleteMetadata)	This method is used to update the metadata about remote partition delete event.
Optional<RemoteLogSegmentMetadata>	remoteLogSegmentMetadata(TopicIdPartition topicIdPartition, int epochForOffset, long offset)	Returns RemoteLogSegmentMetadata if it exists for the given topic partition containing the offset with the given leader-epoch for the offset, else returns Optional.empty().
void	updateRemoteLogSegmentMetadata(RemoteLogSegmentMetadataUpdate remoteLogSegmentMetadataUpdate)	This method is used to update the RemoteLogSegmentMetadata.

Methods inherited from interface java.io.Closeable

close

Methods inherited from interface org.apache.kafka.common.Configurable

configure

Method Details

addRemoteLogSegmentMetadata

void addRemoteLogSegmentMetadata(RemoteLogSegmentMetadata remoteLogSegmentMetadata)
                          throws RemoteStorageException

Adds RemoteLogSegmentMetadata with the containing RemoteLogSegmentId into RemoteLogMetadataManager.

RemoteLogSegmentMetadata is identified by RemoteLogSegmentId and it should have the initial state which is RemoteLogSegmentState.COPY_SEGMENT_STARTED.

updateRemoteLogSegmentMetadata(RemoteLogSegmentMetadataUpdate) should be used to update an existing RemoteLogSegmentMetadata.

Parameters:

remoteLogSegmentMetadata - metadata about the remote log segment.

Throws:

RemoteStorageException - if there are any storage related errors occurred.

IllegalArgumentException - if the given metadata instance does not have the state as RemoteLogSegmentState.COPY_SEGMENT_STARTED

updateRemoteLogSegmentMetadata

void updateRemoteLogSegmentMetadata(RemoteLogSegmentMetadataUpdate remoteLogSegmentMetadataUpdate)
                             throws RemoteStorageException

This method is used to update the RemoteLogSegmentMetadata. Currently, it allows to update with the new state based on the life cycle of the segment. It can go through the below state transitions.

 +---------------------+            +----------------------+
 |COPY_SEGMENT_STARTED |----------->|COPY_SEGMENT_FINISHED |
 +-------------------+-+            +--+-------------------+
                     |                 |
                     |                 |
                     v                 v
                  +--+-----------------+-+
                  |DELETE_SEGMENT_STARTED|
                  +-----------+----------+
                              |
                              |
                              v
                  +-----------+-----------+
                  |DELETE_SEGMENT_FINISHED|
                  +-----------------------+
 

RemoteLogSegmentState.COPY_SEGMENT_STARTED - This state indicates that the segment copying to remote storage is started but not yet finished. RemoteLogSegmentState.COPY_SEGMENT_FINISHED - This state indicates that the segment copying to remote storage is finished.
The leader broker copies the log segments to the remote storage and puts the remote log segment metadata with the state as “COPY_SEGMENT_STARTED” and updates the state as “COPY_SEGMENT_FINISHED” once the copy is successful.

RemoteLogSegmentState.DELETE_SEGMENT_STARTED - This state indicates that the segment deletion is started but not yet finished. RemoteLogSegmentState.DELETE_SEGMENT_FINISHED - This state indicates that the segment is deleted successfully.
Leader partitions publish both the above delete segment events when remote log retention is reached for the respective segments. Remote Partition Removers also publish these events when a segment is deleted as part of the remote partition deletion.

Parameters:

remoteLogSegmentMetadataUpdate - update of the remote log segment metadata.

Throws:

RemoteStorageException - if there are any storage related errors occurred.

RemoteResourceNotFoundException - when there are no resources associated with the given remoteLogSegmentMetadataUpdate.

IllegalArgumentException - if the given metadata instance has the state as RemoteLogSegmentState.COPY_SEGMENT_STARTED

remoteLogSegmentMetadata

Optional<RemoteLogSegmentMetadata> remoteLogSegmentMetadata(TopicIdPartition topicIdPartition,
int epochForOffset,
long offset)
                                                     throws RemoteStorageException

Returns RemoteLogSegmentMetadata if it exists for the given topic partition containing the offset with the given leader-epoch for the offset, else returns Optional.empty().

Parameters:

topicIdPartition - topic partition

epochForOffset - leader epoch for the given offset

offset - offset

Returns:

the requested remote log segment metadata if it exists.

Throws:

RemoteStorageException - if there are any storage related errors occurred.

highestOffsetForEpoch

Optional<Long> highestOffsetForEpoch(TopicIdPartition topicIdPartition,
int leaderEpoch)
                              throws RemoteStorageException

Returns the highest log offset of topic partition for the given leader epoch in remote storage. This is used by remote log management subsystem to know up to which offset the segments have been copied to remote storage for a given leader epoch.

Parameters:

topicIdPartition - topic partition

leaderEpoch - leader epoch

Returns:

the requested highest log offset if exists.

Throws:

RemoteStorageException - if there are any storage related errors occurred.

putRemotePartitionDeleteMetadata

void putRemotePartitionDeleteMetadata(RemotePartitionDeleteMetadata remotePartitionDeleteMetadata)
                               throws RemoteStorageException

This method is used to update the metadata about remote partition delete event. Currently, it allows updating the state (RemotePartitionDeleteState) of a topic partition in remote metadata storage. Controller invokes this method with RemotePartitionDeleteMetadata having state as RemotePartitionDeleteState.DELETE_PARTITION_MARKED. So, remote partition removers can act on this event to clean the respective remote log segments of the partition.

In the case of default RLMM implementation, remote partition remover processes RemotePartitionDeleteState.DELETE_PARTITION_MARKED

• sends an event with state as RemotePartitionDeleteState.DELETE_PARTITION_STARTED
• gets all the remote log segments and deletes them.
• sends an event with state as RemotePartitionDeleteState.DELETE_PARTITION_FINISHED once all the remote log segments are deleted.

Parameters:

remotePartitionDeleteMetadata - update on delete state of a partition.

Throws:

RemoteStorageException - if there are any storage related errors occurred.

RemoteResourceNotFoundException - when there are no resources associated with the given remotePartitionDeleteMetadata.

listRemoteLogSegments

Iterator<RemoteLogSegmentMetadata> listRemoteLogSegments(TopicIdPartition topicIdPartition)
                                                  throws RemoteStorageException

Returns all the remote log segment metadata of the given topicIdPartition.

Remote Partition Removers uses this method to fetch all the segments for a given topic partition, so that they can delete them.

Returns:

Iterator of all the remote log segment metadata for the given topic partition.

Throws:

RemoteStorageException

listRemoteLogSegments

Iterator<RemoteLogSegmentMetadata> listRemoteLogSegments(TopicIdPartition topicIdPartition,
int leaderEpoch)
                                                  throws RemoteStorageException

Returns iterator of remote log segment metadata, sorted by RemoteLogSegmentMetadata.startOffset() in ascending order which contains the given leader epoch. This is used by remote log retention management subsystem to fetch the segment metadata for a given leader epoch.

Parameters:

topicIdPartition - topic partition

leaderEpoch - leader epoch

Returns:

Iterator of remote segments, sorted by start offset in ascending order.

Throws:

RemoteStorageException

onPartitionLeadershipChanges

void onPartitionLeadershipChanges(Set<TopicIdPartition> leaderPartitions,
Set<TopicIdPartition> followerPartitions)

This method is invoked only when there are changes in leadership of the topic partitions that this broker is responsible for.

Parameters:

leaderPartitions - partitions that have become leaders on this broker.

followerPartitions - partitions that have become followers on this broker.

onStopPartitions

void onStopPartitions(Set<TopicIdPartition> partitions)

This method is invoked only when the topic partitions are stopped on this broker. This can happen when a partition is emigrated to other broker or a partition is deleted.

Parameters:

partitions - topic partitions that have been stopped.