sarama

package module
v1.20.1 Latest Latest
Warning

This package is not in the latest version of its module.

Go to latest
Published: Jan 10, 2019 License: MIT Imports: 32 Imported by: 0

README

sarama

GoDoc Build Status Coverage

Sarama is an MIT-licensed Go client library for Apache Kafka version 0.8 (and later).

Getting started
  • API documentation and examples are available via godoc.
  • Mocks for testing are available in the mocks subpackage.
  • The examples directory contains more elaborate example applications.
  • The tools directory contains command line tools that can be useful for testing, diagnostics, and instrumentation.

You might also want to look at the Frequently Asked Questions.

Compatibility and API stability

Sarama provides a "2 releases + 2 months" compatibility guarantee: we support the two latest stable releases of Kafka and Go, and we provide a two month grace period for older releases. This means we currently officially support Go 1.8 through 1.11, and Kafka 1.0 through 2.0, although older releases are still likely to work.

Sarama follows semantic versioning and provides API stability via the gopkg.in service. You can import a version with a guaranteed stable API via http://gopkg.in/Shopify/sarama.v1. A changelog is available here.

Contributing

Documentation

Overview

Package sarama is a pure Go client library for dealing with Apache Kafka (versions 0.8 and later). It includes a high-level API for easily producing and consuming messages, and a low-level API for controlling bytes on the wire when the high-level API is insufficient. Usage examples for the high-level APIs are provided inline with their full documentation.

To produce messages, use either the AsyncProducer or the SyncProducer. The AsyncProducer accepts messages on a channel and produces them asynchronously in the background as efficiently as possible; it is preferred in most cases. The SyncProducer provides a method which will block until Kafka acknowledges the message as produced. This can be useful but comes with two caveats: it will generally be less efficient, and the actual durability guarantees depend on the configured value of `Producer.RequiredAcks`. There are configurations where a message acknowledged by the SyncProducer can still sometimes be lost.

To consume messages, use the Consumer. Note that Sarama's Consumer implementation does not currently support automatic consumer-group rebalancing and offset tracking. For Zookeeper-based tracking (Kafka 0.8.2 and earlier), the https://github.com/wvanbergen/kafka library builds on Sarama to add this support. For Kafka-based tracking (Kafka 0.9 and later), the https://github.com/bsm/sarama-cluster library builds on Sarama to add this support.

For lower-level needs, the Broker and Request/Response objects permit precise control over each connection and message sent on the wire; the Client provides higher-level metadata management that is shared between the producers and the consumer. The Request/Response objects and properties are mostly undocumented, as they line up exactly with the protocol fields documented by Kafka at https://cwiki.apache.org/confluence/display/KAFKA/A+Guide+To+The+Kafka+Protocol

Metrics are exposed through https://github.com/rcrowley/go-metrics library in a local registry.

Broker related metrics:

+----------------------------------------------+------------+---------------------------------------------------------------+
| Name                                         | Type       | Description                                                   |
+----------------------------------------------+------------+---------------------------------------------------------------+
| incoming-byte-rate                           | meter      | Bytes/second read off all brokers                             |
| incoming-byte-rate-for-broker-<broker-id>    | meter      | Bytes/second read off a given broker                          |
| outgoing-byte-rate                           | meter      | Bytes/second written off all brokers                          |
| outgoing-byte-rate-for-broker-<broker-id>    | meter      | Bytes/second written off a given broker                       |
| request-rate                                 | meter      | Requests/second sent to all brokers                           |
| request-rate-for-broker-<broker-id>          | meter      | Requests/second sent to a given broker                        |
| request-size                                 | histogram  | Distribution of the request size in bytes for all brokers     |
| request-size-for-broker-<broker-id>          | histogram  | Distribution of the request size in bytes for a given broker  |
| request-latency-in-ms                        | histogram  | Distribution of the request latency in ms for all brokers     |
| request-latency-in-ms-for-broker-<broker-id> | histogram  | Distribution of the request latency in ms for a given broker  |
| response-rate                                | meter      | Responses/second received from all brokers                    |
| response-rate-for-broker-<broker-id>         | meter      | Responses/second received from a given broker                 |
| response-size                                | histogram  | Distribution of the response size in bytes for all brokers    |
| response-size-for-broker-<broker-id>         | histogram  | Distribution of the response size in bytes for a given broker |
+----------------------------------------------+------------+---------------------------------------------------------------+

Note that we do not gather specific metrics for seed brokers but they are part of the "all brokers" metrics.

Producer related metrics:

+-------------------------------------------+------------+--------------------------------------------------------------------------------------+
| Name                                      | Type       | Description                                                                          |
+-------------------------------------------+------------+--------------------------------------------------------------------------------------+
| batch-size                                | histogram  | Distribution of the number of bytes sent per partition per request for all topics    |
| batch-size-for-topic-<topic>              | histogram  | Distribution of the number of bytes sent per partition per request for a given topic |
| record-send-rate                          | meter      | Records/second sent to all topics                                                    |
| record-send-rate-for-topic-<topic>        | meter      | Records/second sent to a given topic                                                 |
| records-per-request                       | histogram  | Distribution of the number of records sent per request for all topics                |
| records-per-request-for-topic-<topic>     | histogram  | Distribution of the number of records sent per request for a given topic             |
| compression-ratio                         | histogram  | Distribution of the compression ratio times 100 of record batches for all topics     |
| compression-ratio-for-topic-<topic>       | histogram  | Distribution of the compression ratio times 100 of record batches for a given topic  |
+-------------------------------------------+------------+--------------------------------------------------------------------------------------+

Index

Examples

Constants

View Source
const (
	// OffsetNewest stands for the log head offset, i.e. the offset that will be
	// assigned to the next message that will be produced to the partition. You
	// can send this to a client's GetOffset method to get this offset, or when
	// calling ConsumePartition to start consuming new messages.
	OffsetNewest int64 = -1
	// OffsetOldest stands for the oldest offset available on the broker for a
	// partition. You can send this to a client's GetOffset method to get this
	// offset, or when calling ConsumePartition to start consuming from the
	// oldest offset that is still available on the broker.
	OffsetOldest int64 = -2
)
View Source
const CompressionLevelDefault = -1000

CompressionLevelDefault is the constant to use in CompressionLevel to have the default compression level for any codec. The value is picked that we don't use any existing compression levels.

View Source
const GroupGenerationUndefined = -1

GroupGenerationUndefined is a special value for the group generation field of Offset Commit Requests that should be used when a consumer group does not rely on Kafka for partition management.

View Source
const ReceiveTime int64 = -1

ReceiveTime is a special value for the timestamp field of Offset Commit Requests which tells the broker to set the timestamp to the time at which the request was received. The timestamp is only used if message version 1 is used, which requires kafka 0.8.2.

Variables

View Source
var (
	V0_8_2_0  = newKafkaVersion(0, 8, 2, 0)
	V0_8_2_1  = newKafkaVersion(0, 8, 2, 1)
	V0_8_2_2  = newKafkaVersion(0, 8, 2, 2)
	V0_9_0_0  = newKafkaVersion(0, 9, 0, 0)
	V0_9_0_1  = newKafkaVersion(0, 9, 0, 1)
	V0_10_0_0 = newKafkaVersion(0, 10, 0, 0)
	V0_10_0_1 = newKafkaVersion(0, 10, 0, 1)
	V0_10_1_0 = newKafkaVersion(0, 10, 1, 0)
	V0_10_1_1 = newKafkaVersion(0, 10, 1, 1)
	V0_10_2_0 = newKafkaVersion(0, 10, 2, 0)
	V0_10_2_1 = newKafkaVersion(0, 10, 2, 1)
	V0_11_0_0 = newKafkaVersion(0, 11, 0, 0)
	V0_11_0_1 = newKafkaVersion(0, 11, 0, 1)
	V0_11_0_2 = newKafkaVersion(0, 11, 0, 2)
	V1_0_0_0  = newKafkaVersion(1, 0, 0, 0)
	V1_1_0_0  = newKafkaVersion(1, 1, 0, 0)
	V1_1_1_0  = newKafkaVersion(1, 1, 1, 0)
	V2_0_0_0  = newKafkaVersion(2, 0, 0, 0)
	V2_0_1_0  = newKafkaVersion(2, 0, 1, 0)
	V2_1_0_0  = newKafkaVersion(2, 1, 0, 0)

	SupportedVersions = []KafkaVersion{
		V0_8_2_0,
		V0_8_2_1,
		V0_8_2_2,
		V0_9_0_0,
		V0_9_0_1,
		V0_10_0_0,
		V0_10_0_1,
		V0_10_1_0,
		V0_10_1_1,
		V0_10_2_0,
		V0_10_2_1,
		V0_11_0_0,
		V0_11_0_1,
		V0_11_0_2,
		V1_0_0_0,
		V1_1_0_0,
		V1_1_1_0,
		V2_0_0_0,
		V2_0_1_0,
		V2_1_0_0,
	}
	MinVersion = V0_8_2_0
	MaxVersion = V2_1_0_0
)

Effective constants defining the supported kafka versions.

View Source
var BalanceStrategyRange = &balanceStrategy{
	name: "range",
	coreFn: func(plan BalanceStrategyPlan, memberIDs []string, topic string, partitions []int32) {
		step := float64(len(partitions)) / float64(len(memberIDs))

		for i, memberID := range memberIDs {
			pos := float64(i)
			min := int(math.Floor(pos*step + 0.5))
			max := int(math.Floor((pos+1)*step + 0.5))
			plan.Add(memberID, topic, partitions[min:max]...)
		}
	},
}

BalanceStrategyRange is the default and assigns partitions as ranges to consumer group members. Example with one topic T with six partitions (0..5) and two members (M1, M2):

M1: {T: [0, 1, 2]}
M2: {T: [3, 4, 5]}
View Source
var BalanceStrategyRoundRobin = &balanceStrategy{
	name: "roundrobin",
	coreFn: func(plan BalanceStrategyPlan, memberIDs []string, topic string, partitions []int32) {
		for i, part := range partitions {
			memberID := memberIDs[i%len(memberIDs)]
			plan.Add(memberID, topic, part)
		}
	},
}

BalanceStrategyRoundRobin assigns partitions to members in alternating order. Example with topic T with six partitions (0..5) and two members (M1, M2):

M1: {T: [0, 2, 4]}
M2: {T: [1, 3, 5]}
View Source
var ErrAlreadyConnected = errors.New("kafka: broker connection already initiated")

ErrAlreadyConnected is the error returned when calling Open() on a Broker that is already connected or connecting.

View Source
var ErrClosedClient = errors.New("kafka: tried to use a client that was closed")

ErrClosedClient is the error returned when a method is called on a client that has been closed.

View Source
var ErrClosedConsumerGroup = errors.New("kafka: tried to use a consumer group that was closed")

ErrClosedConsumerGroup is the error returned when a method is called on a consumer group that has been closed.

View Source
var ErrConsumerOffsetNotAdvanced = errors.New("kafka: consumer offset was not advanced after a RecordBatch")

ErrConsumerOffsetNotAdvanced is returned when a partition consumer didn't advance its offset after parsing a RecordBatch.

View Source
var ErrControllerNotAvailable = errors.New("kafka: controller is not available")

ErrControllerNotAvailable is returned when server didn't give correct controller id. May be kafka server's version is lower than 0.10.0.0.

View Source
var ErrIncompleteResponse = errors.New("kafka: response did not contain all the expected topic/partition blocks")

ErrIncompleteResponse is the error returned when the server returns a syntactically valid response, but it does not contain the expected information.

View Source
var ErrInsufficientData = errors.New("kafka: insufficient data to decode packet, more bytes expected")

ErrInsufficientData is returned when decoding and the packet is truncated. This can be expected when requesting messages, since as an optimization the server is allowed to return a partial message at the end of the message set.

View Source
var ErrInvalidPartition = errors.New("kafka: partitioner returned an invalid partition index")

ErrInvalidPartition is the error returned when a partitioner returns an invalid partition index (meaning one outside of the range [0...numPartitions-1]).

View Source
var ErrMessageTooLarge = errors.New("kafka: message is larger than Consumer.Fetch.Max")

ErrMessageTooLarge is returned when the next message to consume is larger than the configured Consumer.Fetch.Max

View Source
var ErrNoTopicsToUpdateMetadata = errors.New("kafka: no specific topics to update metadata")

ErrNoTopicsToUpdateMetadata is returned when Meta.Full is set to false but no specific topics were found to update the metadata.

View Source
var ErrNotConnected = errors.New("kafka: broker not connected")

ErrNotConnected is the error returned when trying to send or call Close() on a Broker that is not connected.

View Source
var ErrOutOfBrokers = errors.New("kafka: client has run out of available brokers to talk to (Is your cluster reachable?)")

ErrOutOfBrokers is the error returned when the client has run out of brokers to talk to because all of them errored or otherwise failed to respond.

View Source
var ErrShuttingDown = errors.New("kafka: message received by producer in process of shutting down")

ErrShuttingDown is returned when a producer receives a message during shutdown.

View Source
var MaxRequestSize int32 = 100 * 1024 * 1024

MaxRequestSize is the maximum size (in bytes) of any request that Sarama will attempt to send. Trying to send a request larger than this will result in an PacketEncodingError. The default of 100 MiB is aligned with Kafka's default `socket.request.max.bytes`, which is the largest request the broker will attempt to process.

View Source
var MaxResponseSize int32 = 100 * 1024 * 1024

MaxResponseSize is the maximum size (in bytes) of any response that Sarama will attempt to parse. If a broker returns a response message larger than this value, Sarama will return a PacketDecodingError to protect the client from running out of memory. Please note that brokers do not have any natural limit on the size of responses they send. In particular, they can send arbitrarily large fetch responses to consumers (see https://issues.apache.org/jira/browse/KAFKA-2063).

View Source
var NoNode = &Broker{id: -1, addr: ":-1"}
View Source
var PanicHandler func(interface{})

PanicHandler is called for recovering from panics spawned internally to the library (and thus not recoverable by the caller's goroutine). Defaults to nil, which means panics are not recovered.

Functions

This section is empty.

Types

type AbortedTransaction added in v1.14.0

type AbortedTransaction struct {
	ProducerID  int64
	FirstOffset int64
}

type Acl added in v1.16.0

type Acl struct {
	Principal      string
	Host           string
	Operation      AclOperation
	PermissionType AclPermissionType
}

type AclCreation added in v1.16.0

type AclCreation struct {
	Resource
	Acl
}

type AclCreationResponse added in v1.16.0

type AclCreationResponse struct {
	Err    KError
	ErrMsg *string
}

type AclFilter added in v1.16.0

type AclFilter struct {
	ResourceType   AclResourceType
	ResourceName   *string
	Principal      *string
	Host           *string
	Operation      AclOperation
	PermissionType AclPermissionType
}

type AclOperation added in v1.16.0

type AclOperation int
const (
	AclOperationUnknown         AclOperation = 0
	AclOperationAny             AclOperation = 1
	AclOperationAll             AclOperation = 2
	AclOperationRead            AclOperation = 3
	AclOperationWrite           AclOperation = 4
	AclOperationCreate          AclOperation = 5
	AclOperationDelete          AclOperation = 6
	AclOperationAlter           AclOperation = 7
	AclOperationDescribe        AclOperation = 8
	AclOperationClusterAction   AclOperation = 9
	AclOperationDescribeConfigs AclOperation = 10
	AclOperationAlterConfigs    AclOperation = 11
	AclOperationIdempotentWrite AclOperation = 12
)

ref: https://github.com/apache/kafka/blob/trunk/clients/src/main/java/org/apache/kafka/common/acl/AclOperation.java

type AclPermissionType added in v1.16.0

type AclPermissionType int

type AclResourceType added in v1.16.0

type AclResourceType int
const (
	AclResourceUnknown         AclResourceType = 0
	AclResourceAny             AclResourceType = 1
	AclResourceTopic           AclResourceType = 2
	AclResourceGroup           AclResourceType = 3
	AclResourceCluster         AclResourceType = 4
	AclResourceTransactionalID AclResourceType = 5
)

ref: https://github.com/apache/kafka/blob/trunk/clients/src/main/java/org/apache/kafka/common/resource/ResourceType.java

type AddOffsetsToTxnRequest added in v1.16.0

type AddOffsetsToTxnRequest struct {
	TransactionalID string
	ProducerID      int64
	ProducerEpoch   int16
	GroupID         string
}

type AddOffsetsToTxnResponse added in v1.16.0

type AddOffsetsToTxnResponse struct {
	ThrottleTime time.Duration
	Err          KError
}

type AddPartitionsToTxnRequest added in v1.16.0

type AddPartitionsToTxnRequest struct {
	TransactionalID string
	ProducerID      int64
	ProducerEpoch   int16
	TopicPartitions map[string][]int32
}

type AddPartitionsToTxnResponse added in v1.16.0

type AddPartitionsToTxnResponse struct {
	ThrottleTime time.Duration
	Errors       map[string][]*PartitionError
}

type AlterConfigsRequest added in v1.16.0

type AlterConfigsRequest struct {
	Resources    []*AlterConfigsResource
	ValidateOnly bool
}

type AlterConfigsResource added in v1.16.0

type AlterConfigsResource struct {
	Type          ConfigResourceType
	Name          string
	ConfigEntries map[string]*string
}

type AlterConfigsResourceResponse added in v1.16.0

type AlterConfigsResourceResponse struct {
	ErrorCode int16
	ErrorMsg  string
	Type      ConfigResourceType
	Name      string
}

type AlterConfigsResponse added in v1.16.0

type AlterConfigsResponse struct {
	ThrottleTime time.Duration
	Resources    []*AlterConfigsResourceResponse
}

type ApiVersionsRequest

type ApiVersionsRequest struct {
}

type ApiVersionsResponse

type ApiVersionsResponse struct {
	Err         KError
	ApiVersions []*ApiVersionsResponseBlock
}

type ApiVersionsResponseBlock

type ApiVersionsResponseBlock struct {
	ApiKey     int16
	MinVersion int16
	MaxVersion int16
}

type AsyncProducer

type AsyncProducer interface {

	// AsyncClose triggers a shutdown of the producer. The shutdown has completed
	// when both the Errors and Successes channels have been closed. When calling
	// AsyncClose, you *must* continue to read from those channels in order to
	// drain the results of any messages in flight.
	AsyncClose()

	// Close shuts down the producer and waits for any buffered messages to be
	// flushed. You must call this function before a producer object passes out of
	// scope, as it may otherwise leak memory. You must call this before calling
	// Close on the underlying client.
	Close() error

	// Input is the input channel for the user to write messages to that they
	// wish to send.
	Input() chan<- *ProducerMessage

	// Successes is the success output channel back to the user when Return.Successes is
	// enabled. If Return.Successes is true, you MUST read from this channel or the
	// Producer will deadlock. It is suggested that you send and read messages
	// together in a single select statement.
	Successes() <-chan *ProducerMessage

	// Errors is the error output channel back to the user. You MUST read from this
	// channel or the Producer will deadlock when the channel is full. Alternatively,
	// you can set Producer.Return.Errors in your config to false, which prevents
	// errors to be returned.
	Errors() <-chan *ProducerError
}

AsyncProducer publishes Kafka messages using a non-blocking API. It routes messages to the correct broker for the provided topic-partition, refreshing metadata as appropriate, and parses responses for errors. You must read from the Errors() channel or the producer will deadlock. You must call Close() or AsyncClose() on a producer to avoid leaks: it will not be garbage-collected automatically when it passes out of scope.

Example (Goroutines)

This example shows how to use the producer with separate goroutines reading from the Successes and Errors channels. Note that in order for the Successes channel to be populated, you have to set config.Producer.Return.Successes to true.

config := NewConfig()
config.Producer.Return.Successes = true
producer, err := NewAsyncProducer([]string{"localhost:9092"}, config)
if err != nil {
	panic(err)
}

// Trap SIGINT to trigger a graceful shutdown.
signals := make(chan os.Signal, 1)
signal.Notify(signals, os.Interrupt)

var (
	wg                          sync.WaitGroup
	enqueued, successes, errors int
)

wg.Add(1)
go func() {
	defer wg.Done()
	for range producer.Successes() {
		successes++
	}
}()

wg.Add(1)
go func() {
	defer wg.Done()
	for err := range producer.Errors() {
		log.Println(err)
		errors++
	}
}()

ProducerLoop:
for {
	message := &ProducerMessage{Topic: "my_topic", Value: StringEncoder("testing 123")}
	select {
	case producer.Input() <- message:
		enqueued++

	case <-signals:
		producer.AsyncClose() // Trigger a shutdown of the producer.
		break ProducerLoop
	}
}

wg.Wait()

log.Printf("Successfully produced: %d; errors: %d\n", successes, errors)
Output:

Example (Select)

This example shows how to use the producer while simultaneously reading the Errors channel to know about any failures.

producer, err := NewAsyncProducer([]string{"localhost:9092"}, nil)
if err != nil {
	panic(err)
}

defer func() {
	if err := producer.Close(); err != nil {
		log.Fatalln(err)
	}
}()

// Trap SIGINT to trigger a shutdown.
signals := make(chan os.Signal, 1)
signal.Notify(signals, os.Interrupt)

var enqueued, errors int
ProducerLoop:
for {
	select {
	case producer.Input() <- &ProducerMessage{Topic: "my_topic", Key: nil, Value: StringEncoder("testing 123")}:
		enqueued++
	case err := <-producer.Errors():
		log.Println("Failed to produce message", err)
		errors++
	case <-signals:
		break ProducerLoop
	}
}

log.Printf("Enqueued: %d; errors: %d\n", enqueued, errors)
Output:

func NewAsyncProducer

func NewAsyncProducer(addrs []string, conf *Config) (AsyncProducer, error)

NewAsyncProducer creates a new AsyncProducer using the given broker addresses and configuration.

func NewAsyncProducerFromClient

func NewAsyncProducerFromClient(client Client) (AsyncProducer, error)

NewAsyncProducerFromClient creates a new Producer using the given client. It is still necessary to call Close() on the underlying client when shutting down this producer.

type BalanceStrategy added in v1.19.0

type BalanceStrategy interface {
	// Name uniquely identifies the strategy.
	Name() string

	// Plan accepts a map of `memberID -> metadata` and a map of `topic -> partitions`
	// and returns a distribution plan.
	Plan(members map[string]ConsumerGroupMemberMetadata, topics map[string][]int32) (BalanceStrategyPlan, error)
}

BalanceStrategy is used to balance topics and partitions across memebers of a consumer group

type BalanceStrategyPlan added in v1.19.0

type BalanceStrategyPlan map[string]map[string][]int32

BalanceStrategyPlan is the results of any BalanceStrategy.Plan attempt. It contains an allocation of topic/partitions by memberID in the form of a `memberID -> topic -> partitions` map.

func (BalanceStrategyPlan) Add added in v1.19.0

func (p BalanceStrategyPlan) Add(memberID, topic string, partitions ...int32)

Add assigns a topic with a number partitions to a member.

type Broker

type Broker struct {
	// contains filtered or unexported fields
}

Broker represents a single Kafka broker connection. All operations on this object are entirely concurrency-safe.

Example
broker := NewBroker("localhost:9092")
err := broker.Open(nil)
if err != nil {
	panic(err)
}

request := MetadataRequest{Topics: []string{"myTopic"}}
response, err := broker.GetMetadata(&request)
if err != nil {
	_ = broker.Close()
	panic(err)
}

fmt.Println("There are", len(response.Topics), "topics active in the cluster.")

if err = broker.Close(); err != nil {
	panic(err)
}
Output:

func NewBroker

func NewBroker(addr string) *Broker

NewBroker creates and returns a Broker targeting the given host:port address. This does not attempt to actually connect, you have to call Open() for that.

func (*Broker) AddOffsetsToTxn added in v1.16.0

func (b *Broker) AddOffsetsToTxn(request *AddOffsetsToTxnRequest) (*AddOffsetsToTxnResponse, error)

func (*Broker) AddPartitionsToTxn added in v1.16.0

func (b *Broker) AddPartitionsToTxn(request *AddPartitionsToTxnRequest) (*AddPartitionsToTxnResponse, error)

func (*Broker) Addr

func (b *Broker) Addr() string

Addr returns the broker address as either retrieved from Kafka's metadata or passed to NewBroker.

func (*Broker) AlterConfigs added in v1.16.0

func (b *Broker) AlterConfigs(request *AlterConfigsRequest) (*AlterConfigsResponse, error)

func (*Broker) ApiVersions

func (b *Broker) ApiVersions(request *ApiVersionsRequest) (*ApiVersionsResponse, error)

func (*Broker) Close

func (b *Broker) Close() error

func (*Broker) CommitOffset

func (b *Broker) CommitOffset(request *OffsetCommitRequest) (*OffsetCommitResponse, error)

func (*Broker) Connected

func (b *Broker) Connected() (bool, error)

Connected returns true if the broker is connected and false otherwise. If the broker is not connected but it had tried to connect, the error from that connection attempt is also returned.

func (*Broker) CreateAcls added in v1.16.0

func (b *Broker) CreateAcls(request *CreateAclsRequest) (*CreateAclsResponse, error)

func (*Broker) CreatePartitions added in v1.17.0

func (b *Broker) CreatePartitions(request *CreatePartitionsRequest) (*CreatePartitionsResponse, error)

func (*Broker) CreateTopics added in v1.16.0

func (b *Broker) CreateTopics(request *CreateTopicsRequest) (*CreateTopicsResponse, error)

func (*Broker) DeleteAcls added in v1.16.0

func (b *Broker) DeleteAcls(request *DeleteAclsRequest) (*DeleteAclsResponse, error)

func (*Broker) DeleteGroups added in v1.17.0

func (b *Broker) DeleteGroups(request *DeleteGroupsRequest) (*DeleteGroupsResponse, error)

func (*Broker) DeleteRecords added in v1.17.0

func (b *Broker) DeleteRecords(request *DeleteRecordsRequest) (*DeleteRecordsResponse, error)

func (*Broker) DeleteTopics added in v1.16.0

func (b *Broker) DeleteTopics(request *DeleteTopicsRequest) (*DeleteTopicsResponse, error)

func (*Broker) DescribeAcls added in v1.16.0

func (b *Broker) DescribeAcls(request *DescribeAclsRequest) (*DescribeAclsResponse, error)

func (*Broker) DescribeConfigs added in v1.16.0

func (b *Broker) DescribeConfigs(request *DescribeConfigsRequest) (*DescribeConfigsResponse, error)

func (*Broker) DescribeGroups

func (b *Broker) DescribeGroups(request *DescribeGroupsRequest) (*DescribeGroupsResponse, error)

func (*Broker) EndTxn added in v1.16.0

func (b *Broker) EndTxn(request *EndTxnRequest) (*EndTxnResponse, error)

func (*Broker) Fetch

func (b *Broker) Fetch(request *FetchRequest) (*FetchResponse, error)

func (*Broker) FetchOffset

func (b *Broker) FetchOffset(request *OffsetFetchRequest) (*OffsetFetchResponse, error)

func (*Broker) FindCoordinator added in v1.17.0

func (b *Broker) FindCoordinator(request *FindCoordinatorRequest) (*FindCoordinatorResponse, error)

func (*Broker) GetAvailableOffsets

func (b *Broker) GetAvailableOffsets(request *OffsetRequest) (*OffsetResponse, error)

func (*Broker) GetConsumerMetadata

func (b *Broker) GetConsumerMetadata(request *ConsumerMetadataRequest) (*ConsumerMetadataResponse, error)

func (*Broker) GetMetadata

func (b *Broker) GetMetadata(request *MetadataRequest) (*MetadataResponse, error)

func (*Broker) Heartbeat

func (b *Broker) Heartbeat(request *HeartbeatRequest) (*HeartbeatResponse, error)

func (*Broker) ID

func (b *Broker) ID() int32

ID returns the broker ID retrieved from Kafka's metadata, or -1 if that is not known.

func (*Broker) InitProducerID added in v1.16.0

func (b *Broker) InitProducerID(request *InitProducerIDRequest) (*InitProducerIDResponse, error)

func (*Broker) JoinGroup

func (b *Broker) JoinGroup(request *JoinGroupRequest) (*JoinGroupResponse, error)

func (*Broker) LeaveGroup

func (b *Broker) LeaveGroup(request *LeaveGroupRequest) (*LeaveGroupResponse, error)

func (*Broker) ListGroups

func (b *Broker) ListGroups(request *ListGroupsRequest) (*ListGroupsResponse, error)

func (*Broker) Open

func (b *Broker) Open(conf *Config) error

Open tries to connect to the Broker if it is not already connected or connecting, but does not block waiting for the connection to complete. This means that any subsequent operations on the broker will block waiting for the connection to succeed or fail. To get the effect of a fully synchronous Open call, follow it by a call to Connected(). The only errors Open will return directly are ConfigurationError or AlreadyConnected. If conf is nil, the result of NewConfig() is used.

func (*Broker) Produce

func (b *Broker) Produce(request *ProduceRequest) (*ProduceResponse, error)

func (*Broker) Rack added in v1.20.0

func (b *Broker) Rack() string

Rack returns the broker's rack as retrieved from Kafka's metadata or the empty string if it is not known. The returned value corresponds to the broker's broker.rack configuration setting. Requires protocol version to be at least v0.10.0.0.

func (*Broker) SyncGroup

func (b *Broker) SyncGroup(request *SyncGroupRequest) (*SyncGroupResponse, error)

func (*Broker) TxnOffsetCommit added in v1.16.0

func (b *Broker) TxnOffsetCommit(request *TxnOffsetCommitRequest) (*TxnOffsetCommitResponse, error)

type ByteEncoder

type ByteEncoder []byte

ByteEncoder implements the Encoder interface for Go byte slices so that they can be used as the Key or Value in a ProducerMessage.

func (ByteEncoder) Encode

func (b ByteEncoder) Encode() ([]byte, error)

func (ByteEncoder) Length

func (b ByteEncoder) Length() int

type Client

type Client interface {
	// Config returns the Config struct of the client. This struct should not be
	// altered after it has been created.
	Config() *Config

	// Controller returns the cluster controller broker. Requires Kafka 0.10 or higher.
	Controller() (*Broker, error)

	// Brokers returns the current set of active brokers as retrieved from cluster metadata.
	Brokers() []*Broker

	// Topics returns the set of available topics as retrieved from cluster metadata.
	Topics() ([]string, error)

	// Partitions returns the sorted list of all partition IDs for the given topic.
	Partitions(topic string) ([]int32, error)

	// WritablePartitions returns the sorted list of all writable partition IDs for
	// the given topic, where "writable" means "having a valid leader accepting
	// writes".
	WritablePartitions(topic string) ([]int32, error)

	// Leader returns the broker object that is the leader of the current
	// topic/partition, as determined by querying the cluster metadata.
	Leader(topic string, partitionID int32) (*Broker, error)

	// Replicas returns the set of all replica IDs for the given partition.
	Replicas(topic string, partitionID int32) ([]int32, error)

	// InSyncReplicas returns the set of all in-sync replica IDs for the given
	// partition. In-sync replicas are replicas which are fully caught up with
	// the partition leader.
	InSyncReplicas(topic string, partitionID int32) ([]int32, error)

	// RefreshMetadata takes a list of topics and queries the cluster to refresh the
	// available metadata for those topics. If no topics are provided, it will refresh
	// metadata for all topics.
	RefreshMetadata(topics ...string) error

	// GetOffset queries the cluster to get the most recent available offset at the
	// given time (in milliseconds) on the topic/partition combination.
	// Time should be OffsetOldest for the earliest available offset,
	// OffsetNewest for the offset of the message that will be produced next, or a time.
	GetOffset(topic string, partitionID int32, time int64) (int64, error)

	// Coordinator returns the coordinating broker for a consumer group. It will
	// return a locally cached value if it's available. You can call
	// RefreshCoordinator to update the cached value. This function only works on
	// Kafka 0.8.2 and higher.
	Coordinator(consumerGroup string) (*Broker, error)

	// RefreshCoordinator retrieves the coordinator for a consumer group and stores it
	// in local cache. This function only works on Kafka 0.8.2 and higher.
	RefreshCoordinator(consumerGroup string) error

	// InitProducerID retrieves information required for Idempotent Producer
	InitProducerID() (*InitProducerIDResponse, error)

	// Close shuts down all broker connections managed by this client. It is required
	// to call this function before a client object passes out of scope, as it will
	// otherwise leak memory. You must close any Producers or Consumers using a client
	// before you close the client.
	Close() error

	// Closed returns true if the client has already had Close called on it
	Closed() bool
}

Client is a generic Kafka client. It manages connections to one or more Kafka brokers. You MUST call Close() on a client to avoid leaks, it will not be garbage-collected automatically when it passes out of scope. It is safe to share a client amongst many users, however Kafka will process requests from a single client strictly in serial, so it is generally more efficient to use the default one client per producer/consumer.

func NewClient

func NewClient(addrs []string, conf *Config) (Client, error)

NewClient creates a new Client. It connects to one of the given broker addresses and uses that broker to automatically fetch metadata on the rest of the kafka cluster. If metadata cannot be retrieved from any of the given broker addresses, the client is not created.

type ClusterAdmin added in v1.18.0

type ClusterAdmin interface {
	// Creates a new topic. This operation is supported by brokers with version 0.10.1.0 or higher.
	// It may take several seconds after CreateTopic returns success for all the brokers
	// to become aware that the topic has been created. During this time, listTopics
	// may not return information about the new topic.The validateOnly option is supported from version 0.10.2.0.
	CreateTopic(topic string, detail *TopicDetail, validateOnly bool) error

	// Delete a topic. It may take several seconds after the DeleteTopic to returns success
	// and for all the brokers to become aware that the topics are gone.
	// During this time, listTopics  may continue to return information about the deleted topic.
	// If delete.topic.enable is false on the brokers, deleteTopic will mark
	// the topic for deletion, but not actually delete them.
	// This operation is supported by brokers with version 0.10.1.0 or higher.
	DeleteTopic(topic string) error

	// Increase the number of partitions of the topics  according to the corresponding values.
	// If partitions are increased for a topic that has a key, the partition logic or ordering of
	// the messages will be affected. It may take several seconds after this method returns
	// success for all the brokers to become aware that the partitions have been created.
	// During this time, ClusterAdmin#describeTopics may not return information about the
	// new partitions. This operation is supported by brokers with version 1.0.0 or higher.
	CreatePartitions(topic string, count int32, assignment [][]int32, validateOnly bool) error

	// Delete records whose offset is smaller than the given offset of the corresponding partition.
	// This operation is supported by brokers with version 0.11.0.0 or higher.
	DeleteRecords(topic string, partitionOffsets map[int32]int64) error

	// Get the configuration for the specified resources.
	// The returned configuration includes default values and the Default is true
	// can be used to distinguish them from user supplied values.
	// Config entries where ReadOnly is true cannot be updated.
	// The value of config entries where Sensitive is true is always nil so
	// sensitive information is not disclosed.
	// This operation is supported by brokers with version 0.11.0.0 or higher.
	DescribeConfig(resource ConfigResource) ([]ConfigEntry, error)

	// Update the configuration for the specified resources with the default options.
	// This operation is supported by brokers with version 0.11.0.0 or higher.
	// The resources with their configs (topic is the only resource type with configs
	// that can be updated currently Updates are not transactional so they may succeed
	// for some resources while fail for others. The configs for a particular resource are updated automatically.
	AlterConfig(resourceType ConfigResourceType, name string, entries map[string]*string, validateOnly bool) error

	// Creates access control lists (ACLs) which are bound to specific resources.
	// This operation is not transactional so it may succeed for some ACLs while fail for others.
	// If you attempt to add an ACL that duplicates an existing ACL, no error will be raised, but
	// no changes will be made. This operation is supported by brokers with version 0.11.0.0 or higher.
	CreateACL(resource Resource, acl Acl) error

	// Lists access control lists (ACLs) according to the supplied filter.
	// it may take some time for changes made by createAcls or deleteAcls to be reflected in the output of ListAcls
	// This operation is supported by brokers with version 0.11.0.0 or higher.
	ListAcls(filter AclFilter) ([]ResourceAcls, error)

	// Deletes access control lists (ACLs) according to the supplied filters.
	// This operation is not transactional so it may succeed for some ACLs while fail for others.
	// This operation is supported by brokers with version 0.11.0.0 or higher.
	DeleteACL(filter AclFilter, validateOnly bool) ([]MatchingAcl, error)

	// Close shuts down the admin and closes underlying client.
	Close() error
}

ClusterAdmin is the administrative client for Kafka, which supports managing and inspecting topics, brokers, configurations and ACLs. The minimum broker version required is 0.10.0.0. Methods with stricter requirements will specify the minimum broker version required. You MUST call Close() on a client to avoid leaks

func NewClusterAdmin added in v1.18.0

func NewClusterAdmin(addrs []string, conf *Config) (ClusterAdmin, error)

NewClusterAdmin creates a new ClusterAdmin using the given broker addresses and configuration.

type CompressionCodec

type CompressionCodec int8

CompressionCodec represents the various compression codecs recognized by Kafka in messages.

const (
	CompressionNone   CompressionCodec = 0
	CompressionGZIP   CompressionCodec = 1
	CompressionSnappy CompressionCodec = 2
	CompressionLZ4    CompressionCodec = 3
	CompressionZSTD   CompressionCodec = 4
)

func (CompressionCodec) String added in v1.17.0

func (cc CompressionCodec) String() string

type Config

type Config struct {
	// Admin is the namespace for ClusterAdmin properties used by the administrative Kafka client.
	Admin struct {
		// The maximum duration the administrative Kafka client will wait for ClusterAdmin operations,
		// including topics, brokers, configurations and ACLs (defaults to 3 seconds).
		Timeout time.Duration
	}

	// Net is the namespace for network-level properties used by the Broker, and
	// shared by the Client/Producer/Consumer.
	Net struct {
		// How many outstanding requests a connection is allowed to have before
		// sending on it blocks (default 5).
		MaxOpenRequests int

		// All three of the below configurations are similar to the
		// `socket.timeout.ms` setting in JVM kafka. All of them default
		// to 30 seconds.
		DialTimeout  time.Duration // How long to wait for the initial connection.
		ReadTimeout  time.Duration // How long to wait for a response.
		WriteTimeout time.Duration // How long to wait for a transmit.

		TLS struct {
			// Whether or not to use TLS when connecting to the broker
			// (defaults to false).
			Enable bool
			// The TLS configuration to use for secure connections if
			// enabled (defaults to nil).
			Config *tls.Config
		}

		// SASL based authentication with broker. While there are multiple SASL authentication methods
		// the current implementation is limited to plaintext (SASL/PLAIN) authentication
		SASL struct {
			// Whether or not to use SASL authentication when connecting to the broker
			// (defaults to false).
			Enable bool
			// Whether or not to send the Kafka SASL handshake first if enabled
			// (defaults to true). You should only set this to false if you're using
			// a non-Kafka SASL proxy.
			Handshake bool
			//username and password for SASL/PLAIN authentication
			User     string
			Password string
		}

		// KeepAlive specifies the keep-alive period for an active network connection.
		// If zero, keep-alives are disabled. (default is 0: disabled).
		KeepAlive time.Duration

		// LocalAddr is the local address to use when dialing an
		// address. The address must be of a compatible type for the
		// network being dialed.
		// If nil, a local address is automatically chosen.
		LocalAddr net.Addr
	}

	// Metadata is the namespace for metadata management properties used by the
	// Client, and shared by the Producer/Consumer.
	Metadata struct {
		Retry struct {
			// The total number of times to retry a metadata request when the
			// cluster is in the middle of a leader election (default 3).
			Max int
			// How long to wait for leader election to occur before retrying
			// (default 250ms). Similar to the JVM's `retry.backoff.ms`.
			Backoff time.Duration
		}
		// How frequently to refresh the cluster metadata in the background.
		// Defaults to 10 minutes. Set to 0 to disable. Similar to
		// `topic.metadata.refresh.interval.ms` in the JVM version.
		RefreshFrequency time.Duration

		// Whether to maintain a full set of metadata for all topics, or just
		// the minimal set that has been necessary so far. The full set is simpler
		// and usually more convenient, but can take up a substantial amount of
		// memory if you have many topics and partitions. Defaults to true.
		Full bool
	}

	// Producer is the namespace for configuration related to producing messages,
	// used by the Producer.
	Producer struct {
		// The maximum permitted size of a message (defaults to 1000000). Should be
		// set equal to or smaller than the broker's `message.max.bytes`.
		MaxMessageBytes int
		// The level of acknowledgement reliability needed from the broker (defaults
		// to WaitForLocal). Equivalent to the `request.required.acks` setting of the
		// JVM producer.
		RequiredAcks RequiredAcks
		// The maximum duration the broker will wait the receipt of the number of
		// RequiredAcks (defaults to 10 seconds). This is only relevant when
		// RequiredAcks is set to WaitForAll or a number > 1. Only supports
		// millisecond resolution, nanoseconds will be truncated. Equivalent to
		// the JVM producer's `request.timeout.ms` setting.
		Timeout time.Duration
		// The type of compression to use on messages (defaults to no compression).
		// Similar to `compression.codec` setting of the JVM producer.
		Compression CompressionCodec
		// The level of compression to use on messages. The meaning depends
		// on the actual compression type used and defaults to default compression
		// level for the codec.
		CompressionLevel int
		// Generates partitioners for choosing the partition to send messages to
		// (defaults to hashing the message key). Similar to the `partitioner.class`
		// setting for the JVM producer.
		Partitioner PartitionerConstructor
		// If enabled, the producer will ensure that exactly one copy of each message is
		// written.
		Idempotent bool

		// Return specifies what channels will be populated. If they are set to true,
		// you must read from the respective channels to prevent deadlock. If,
		// however, this config is used to create a `SyncProducer`, both must be set
		// to true and you shall not read from the channels since the producer does
		// this internally.
		Return struct {
			// If enabled, successfully delivered messages will be returned on the
			// Successes channel (default disabled).
			Successes bool

			// If enabled, messages that failed to deliver will be returned on the
			// Errors channel, including error (default enabled).
			Errors bool
		}

		// The following config options control how often messages are batched up and
		// sent to the broker. By default, messages are sent as fast as possible, and
		// all messages received while the current batch is in-flight are placed
		// into the subsequent batch.
		Flush struct {
			// The best-effort number of bytes needed to trigger a flush. Use the
			// global sarama.MaxRequestSize to set a hard upper limit.
			Bytes int
			// The best-effort number of messages needed to trigger a flush. Use
			// `MaxMessages` to set a hard upper limit.
			Messages int
			// The best-effort frequency of flushes. Equivalent to
			// `queue.buffering.max.ms` setting of JVM producer.
			Frequency time.Duration
			// The maximum number of messages the producer will send in a single
			// broker request. Defaults to 0 for unlimited. Similar to
			// `queue.buffering.max.messages` in the JVM producer.
			MaxMessages int
		}

		Retry struct {
			// The total number of times to retry sending a message (default 3).
			// Similar to the `message.send.max.retries` setting of the JVM producer.
			Max int
			// How long to wait for the cluster to settle between retries
			// (default 100ms). Similar to the `retry.backoff.ms` setting of the
			// JVM producer.
			Backoff time.Duration
		}
	}

	// Consumer is the namespace for configuration related to consuming messages,
	// used by the Consumer.
	Consumer struct {

		// Group is the namespace for configuring consumer group.
		Group struct {
			Session struct {
				// The timeout used to detect consumer failures when using Kafka's group management facility.
				// The consumer sends periodic heartbeats to indicate its liveness to the broker.
				// If no heartbeats are received by the broker before the expiration of this session timeout,
				// then the broker will remove this consumer from the group and initiate a rebalance.
				// Note that the value must be in the allowable range as configured in the broker configuration
				// by `group.min.session.timeout.ms` and `group.max.session.timeout.ms` (default 10s)
				Timeout time.Duration
			}
			Heartbeat struct {
				// The expected time between heartbeats to the consumer coordinator when using Kafka's group
				// management facilities. Heartbeats are used to ensure that the consumer's session stays active and
				// to facilitate rebalancing when new consumers join or leave the group.
				// The value must be set lower than Consumer.Group.Session.Timeout, but typically should be set no
				// higher than 1/3 of that value.
				// It can be adjusted even lower to control the expected time for normal rebalances (default 3s)
				Interval time.Duration
			}
			Rebalance struct {
				// Strategy for allocating topic partitions to members (default BalanceStrategyRange)
				Strategy BalanceStrategy
				// The maximum allowed time for each worker to join the group once a rebalance has begun.
				// This is basically a limit on the amount of time needed for all tasks to flush any pending
				// data and commit offsets. If the timeout is exceeded, then the worker will be removed from
				// the group, which will cause offset commit failures (default 60s).
				Timeout time.Duration

				Retry struct {
					// When a new consumer joins a consumer group the set of consumers attempt to "rebalance"
					// the load to assign partitions to each consumer. If the set of consumers changes while
					// this assignment is taking place the rebalance will fail and retry. This setting controls
					// the maximum number of attempts before giving up (default 4).
					Max int
					// Backoff time between retries during rebalance (default 2s)
					Backoff time.Duration
				}
			}
			Member struct {
				// Custom metadata to include when joining the group. The user data for all joined members
				// can be retrieved by sending a DescribeGroupRequest to the broker that is the
				// coordinator for the group.
				UserData []byte
			}
		}

		Retry struct {
			// How long to wait after a failing to read from a partition before
			// trying again (default 2s).
			Backoff time.Duration
		}

		// Fetch is the namespace for controlling how many bytes are retrieved by any
		// given request.
		Fetch struct {
			// The minimum number of message bytes to fetch in a request - the broker
			// will wait until at least this many are available. The default is 1,
			// as 0 causes the consumer to spin when no messages are available.
			// Equivalent to the JVM's `fetch.min.bytes`.
			Min int32
			// The default number of message bytes to fetch from the broker in each
			// request (default 1MB). This should be larger than the majority of
			// your messages, or else the consumer will spend a lot of time
			// negotiating sizes and not actually consuming. Similar to the JVM's
			// `fetch.message.max.bytes`.
			Default int32
			// The maximum number of message bytes to fetch from the broker in a
			// single request. Messages larger than this will return
			// ErrMessageTooLarge and will not be consumable, so you must be sure
			// this is at least as large as your largest message. Defaults to 0
			// (no limit). Similar to the JVM's `fetch.message.max.bytes`. The
			// global `sarama.MaxResponseSize` still applies.
			Max int32
		}
		// The maximum amount of time the broker will wait for Consumer.Fetch.Min
		// bytes to become available before it returns fewer than that anyways. The
		// default is 250ms, since 0 causes the consumer to spin when no events are
		// available. 100-500ms is a reasonable range for most cases. Kafka only
		// supports precision up to milliseconds; nanoseconds will be truncated.
		// Equivalent to the JVM's `fetch.wait.max.ms`.
		MaxWaitTime time.Duration

		// The maximum amount of time the consumer expects a message takes to
		// process for the user. If writing to the Messages channel takes longer
		// than this, that partition will stop fetching more messages until it
		// can proceed again.
		// Note that, since the Messages channel is buffered, the actual grace time is
		// (MaxProcessingTime * ChanneBufferSize). Defaults to 100ms.
		// If a message is not written to the Messages channel between two ticks
		// of the expiryTicker then a timeout is detected.
		// Using a ticker instead of a timer to detect timeouts should typically
		// result in many fewer calls to Timer functions which may result in a
		// significant performance improvement if many messages are being sent
		// and timeouts are infrequent.
		// The disadvantage of using a ticker instead of a timer is that
		// timeouts will be less accurate. That is, the effective timeout could
		// be between `MaxProcessingTime` and `2 * MaxProcessingTime`. For
		// example, if `MaxProcessingTime` is 100ms then a delay of 180ms
		// between two messages being sent may not be recognized as a timeout.
		MaxProcessingTime time.Duration

		// Return specifies what channels will be populated. If they are set to true,
		// you must read from them to prevent deadlock.
		Return struct {
			// If enabled, any errors that occurred while consuming are returned on
			// the Errors channel (default disabled).
			Errors bool
		}

		// Offsets specifies configuration for how and when to commit consumed
		// offsets. This currently requires the manual use of an OffsetManager
		// but will eventually be automated.
		Offsets struct {
			// How frequently to commit updated offsets. Defaults to 1s.
			CommitInterval time.Duration

			// The initial offset to use if no offset was previously committed.
			// Should be OffsetNewest or OffsetOldest. Defaults to OffsetNewest.
			Initial int64

			// The retention duration for committed offsets. If zero, disabled
			// (in which case the `offsets.retention.minutes` option on the
			// broker will be used).  Kafka only supports precision up to
			// milliseconds; nanoseconds will be truncated. Requires Kafka
			// broker version 0.9.0 or later.
			// (default is 0: disabled).
			Retention time.Duration

			Retry struct {
				// The total number of times to retry failing commit
				// requests during OffsetManager shutdown (default 3).
				Max int
			}
		}
	}

	// A user-provided string sent with every request to the brokers for logging,
	// debugging, and auditing purposes. Defaults to "sarama", but you should
	// probably set it to something specific to your application.
	ClientID string
	// The number of events to buffer in internal and external channels. This
	// permits the producer and consumer to continue processing some messages
	// in the background while user code is working, greatly improving throughput.
	// Defaults to 256.
	ChannelBufferSize int
	// The version of Kafka that Sarama will assume it is running against.
	// Defaults to the oldest supported stable version. Since Kafka provides
	// backwards-compatibility, setting it to a version older than you have
	// will not break anything, although it may prevent you from using the
	// latest features. Setting it to a version greater than you are actually
	// running may lead to random breakage.
	Version KafkaVersion
	// The registry to define metrics into.
	// Defaults to a local registry.
	// If you want to disable metrics gathering, set "metrics.UseNilMetrics" to "true"
	// prior to starting Sarama.
	// See Examples on how to use the metrics registry
	MetricRegistry metrics.Registry
}

Config is used to pass multiple configuration options to Sarama's constructors.

Example (Metrics)

This example shows how to integrate with an existing registry as well as publishing metrics on the standard output

// Our application registry
appMetricRegistry := metrics.NewRegistry()
appGauge := metrics.GetOrRegisterGauge("m1", appMetricRegistry)
appGauge.Update(1)

config := NewConfig()
// Use a prefix registry instead of the default local one
config.MetricRegistry = metrics.NewPrefixedChildRegistry(appMetricRegistry, "sarama.")

// Simulate a metric created by sarama without starting a broker
saramaGauge := metrics.GetOrRegisterGauge("m2", config.MetricRegistry)
saramaGauge.Update(2)

metrics.WriteOnce(appMetricRegistry, os.Stdout)
Output:

gauge m1
  value:               1
gauge sarama.m2
  value:               2

func NewConfig

func NewConfig() *Config

NewConfig returns a new configuration instance with sane defaults.

func (*Config) Validate

func (c *Config) Validate() error

Validate checks a Config instance. It will return a ConfigurationError if the specified values don't make sense.

type ConfigEntry added in v1.16.0

type ConfigEntry struct {
	Name      string
	Value     string
	ReadOnly  bool
	Default   bool
	Source    ConfigSource
	Sensitive bool
	Synonyms  []*ConfigSynonym
}

type ConfigResource added in v1.16.0

type ConfigResource struct {
	Type        ConfigResourceType
	Name        string
	ConfigNames []string
}

type ConfigResourceType added in v1.16.0

type ConfigResourceType int8
const (
	UnknownResource ConfigResourceType = 0
	AnyResource     ConfigResourceType = 1
	TopicResource   ConfigResourceType = 2
	GroupResource   ConfigResourceType = 3
	ClusterResource ConfigResourceType = 4
	BrokerResource  ConfigResourceType = 5
)

type ConfigSource added in v1.20.1

type ConfigSource int8
const (
	SourceUnknown              ConfigSource = 0
	SourceTopic                ConfigSource = 1
	SourceDynamicBroker        ConfigSource = 2
	SourceDynamicDefaultBroker ConfigSource = 3
	SourceStaticBroker         ConfigSource = 4
	SourceDefault              ConfigSource = 5
)

func (ConfigSource) String added in v1.20.1

func (s ConfigSource) String() string

type ConfigSynonym added in v1.20.1

type ConfigSynonym struct {
	ConfigName  string
	ConfigValue string
	Source      ConfigSource
}

type ConfigurationError

type ConfigurationError string

ConfigurationError is the type of error returned from a constructor (e.g. NewClient, or NewConsumer) when the specified configuration is invalid.

func (ConfigurationError) Error

func (err ConfigurationError) Error() string

type Consumer

type Consumer interface {

	// Topics returns the set of available topics as retrieved from the cluster
	// metadata. This method is the same as Client.Topics(), and is provided for
	// convenience.
	Topics() ([]string, error)

	// Partitions returns the sorted list of all partition IDs for the given topic.
	// This method is the same as Client.Partitions(), and is provided for convenience.
	Partitions(topic string) ([]int32, error)

	// ConsumePartition creates a PartitionConsumer on the given topic/partition with
	// the given offset. It will return an error if this Consumer is already consuming
	// on the given topic/partition. Offset can be a literal offset, or OffsetNewest
	// or OffsetOldest
	ConsumePartition(topic string, partition int32, offset int64) (PartitionConsumer, error)

	// HighWaterMarks returns the current high water marks for each topic and partition.
	// Consistency between partitions is not guaranteed since high water marks are updated separately.
	HighWaterMarks() map[string]map[int32]int64

	// Close shuts down the consumer. It must be called after all child
	// PartitionConsumers have already been closed.
	Close() error
}

Consumer manages PartitionConsumers which process Kafka messages from brokers. You MUST call Close() on a consumer to avoid leaks, it will not be garbage-collected automatically when it passes out of scope.

Sarama's Consumer type does not currently support automatic consumer-group rebalancing and offset tracking. For Zookeeper-based tracking (Kafka 0.8.2 and earlier), the https://github.com/wvanbergen/kafka library builds on Sarama to add this support. For Kafka-based tracking (Kafka 0.9 and later), the https://github.com/bsm/sarama-cluster library builds on Sarama to add this support.

Example

This example shows how to use the consumer to read messages from a single partition.

consumer, err := NewConsumer([]string{"localhost:9092"}, nil)
if err != nil {
	panic(err)
}

defer func() {
	if err := consumer.Close(); err != nil {
		log.Fatalln(err)
	}
}()

partitionConsumer, err := consumer.ConsumePartition("my_topic", 0, OffsetNewest)
if err != nil {
	panic(err)
}

defer func() {
	if err := partitionConsumer.Close(); err != nil {
		log.Fatalln(err)
	}
}()

// Trap SIGINT to trigger a shutdown.
signals := make(chan os.Signal, 1)
signal.Notify(signals, os.Interrupt)

consumed := 0
ConsumerLoop:
for {
	select {
	case msg := <-partitionConsumer.Messages():
		log.Printf("Consumed message offset %d\n", msg.Offset)
		consumed++
	case <-signals:
		break ConsumerLoop
	}
}

log.Printf("Consumed: %d\n", consumed)
Output:

func NewConsumer

func NewConsumer(addrs []string, config *Config) (Consumer, error)

NewConsumer creates a new consumer using the given broker addresses and configuration.

func NewConsumerFromClient

func NewConsumerFromClient(client Client) (Consumer, error)

NewConsumerFromClient creates a new consumer using the given client. It is still necessary to call Close() on the underlying client when shutting down this consumer.

type ConsumerError

type ConsumerError struct {
	Topic     string
	Partition int32
	Err       error
}

ConsumerError is what is provided to the user when an error occurs. It wraps an error and includes the topic and partition.

func (ConsumerError) Error

func (ce ConsumerError) Error() string

type ConsumerErrors

type ConsumerErrors []*ConsumerError

ConsumerErrors is a type that wraps a batch of errors and implements the Error interface. It can be returned from the PartitionConsumer's Close methods to avoid the need to manually drain errors when stopping.

func (ConsumerErrors) Error

func (ce ConsumerErrors) Error() string

type ConsumerGroup added in v1.19.0

type ConsumerGroup interface {
	// Consume joins a cluster of consumers for a given list of topics and
	// starts a blocking ConsumerGroupSession through the ConsumerGroupHandler.
	//
	// The life-cycle of a session is represented by the following steps:
	//
	// 1. The consumers join the group (as explained in https://kafka.apache.org/documentation/#intro_consumers)
	//    and is assigned their "fair share" of partitions, aka 'claims'.
	// 2. Before processing starts, the handler's Setup() hook is called to notify the user
	//    of the claims and allow any necessary preparation or alteration of state.
	// 3. For each of the assigned claims the handler's ConsumeClaim() function is then called
	//    in a separate goroutine which requires it to be thread-safe. Any state must be carefully protected
	//    from concurrent reads/writes.
	// 4. The session will persist until one of the ConsumeClaim() functions exits. This can be either when the
	//    parent context is cancelled or when a server-side rebalance cycle is initiated.
	// 5. Once all the ConsumeClaim() loops have exited, the handler's Cleanup() hook is called
	//    to allow the user to perform any final tasks before a rebalance.
	// 6. Finally, marked offsets are committed one last time before claims are released.
	//
	// Please note, that once a rebalance is triggered, sessions must be completed within
	// Config.Consumer.Group.Rebalance.Timeout. This means that ConsumeClaim() functions must exit
	// as quickly as possible to allow time for Cleanup() and the final offset commit. If the timeout
	// is exceeded, the consumer will be removed from the group by Kafka, which will cause offset
	// commit failures.
	Consume(ctx context.Context, topics []string, handler ConsumerGroupHandler) error

	// Errors returns a read channel of errors that occurred during the consumer life-cycle.
	// By default, errors are logged and not returned over this channel.
	// If you want to implement any custom error handling, set your config's
	// Consumer.Return.Errors setting to true, and read from this channel.
	Errors() <-chan error

	// Close stops the ConsumerGroup and detaches any running sessions. It is required to call
	// this function before the object passes out of scope, as it will otherwise leak memory.
	Close() error
}

ConsumerGroup is responsible for dividing up processing of topics and partitions over a collection of processes (the members of the consumer group).

Example
package main

import (
	"context"
	"fmt"
)

type exampleConsumerGroupHandler struct{}

func (exampleConsumerGroupHandler) Setup(_ ConsumerGroupSession) error   { return nil }
func (exampleConsumerGroupHandler) Cleanup(_ ConsumerGroupSession) error { return nil }
func (h exampleConsumerGroupHandler) ConsumeClaim(sess ConsumerGroupSession, claim ConsumerGroupClaim) error {
	for msg := range claim.Messages() {
		fmt.Printf("Message topic:%q partition:%d offset:%d\n", msg.Topic, msg.Partition, msg.Offset)
		sess.MarkMessage(msg, "")
	}
	return nil
}

func main() {
	// Init config, specify appropriate version
	config := NewConfig()
	config.Version = V1_0_0_0
	config.Consumer.Return.Errors = true

	// Start with a client
	client, err := NewClient([]string{"localhost:9092"}, config)
	if err != nil {
		panic(err)
	}
	defer func() { _ = client.Close() }()

	// Start a new consumer group
	group, err := NewConsumerGroupFromClient("my-group", client)
	if err != nil {
		panic(err)
	}
	defer func() { _ = group.Close() }()

	// Track errors
	go func() {
		for err := range group.Errors() {
			fmt.Println("ERROR", err)
		}
	}()

	// Iterate over consumer sessions.
	ctx := context.Background()
	for {
		topics := []string{"my-topic"}
		handler := exampleConsumerGroupHandler{}

		err := group.Consume(ctx, topics, handler)
		if err != nil {
			panic(err)
		}
	}
}
Output:

func NewConsumerGroup added in v1.19.0

func NewConsumerGroup(addrs []string, groupID string, config *Config) (ConsumerGroup, error)

NewConsumerGroup creates a new consumer group the given broker addresses and configuration.

func NewConsumerGroupFromClient added in v1.19.0

func NewConsumerGroupFromClient(groupID string, client Client) (ConsumerGroup, error)

NewConsumerGroupFromClient creates a new consumer group using the given client. It is still necessary to call Close() on the underlying client when shutting down this consumer. PLEASE NOTE: consumer groups can only re-use but not share clients.

type ConsumerGroupClaim added in v1.19.0

type ConsumerGroupClaim interface {
	// Topic returns the consumed topic name.
	Topic() string

	// Partition returns the consumed partition.
	Partition() int32

	// InitialOffset returns the initial offset that was used as a starting point for this claim.
	InitialOffset() int64

	// HighWaterMarkOffset returns the high water mark offset of the partition,
	// i.e. the offset that will be used for the next message that will be produced.
	// You can use this to determine how far behind the processing is.
	HighWaterMarkOffset() int64

	// Messages returns the read channel for the messages that are returned by
	// the broker. The messages channel will be closed when a new rebalance cycle
	// is due. You must finish processing and mark offsets within
	// Config.Consumer.Group.Session.Timeout before the topic/partition is eventually
	// re-assigned to another group member.
	Messages() <-chan *ConsumerMessage
}

ConsumerGroupClaim processes Kafka messages from a given topic and partition within a consumer group.

type ConsumerGroupHandler added in v1.19.0

type ConsumerGroupHandler interface {
	// Setup is run at the beginning of a new session, before ConsumeClaim.
	Setup(ConsumerGroupSession) error

	// Cleanup is run at the end of a session, once all ConsumeClaim goroutines have exited
	// but before the offsets are committed for the very last time.
	Cleanup(ConsumerGroupSession) error

	// ConsumeClaim must start a consumer loop of ConsumerGroupClaim's Messages().
	// Once the Messages() channel is closed, the Handler must finish its processing
	// loop and exit.
	ConsumeClaim(ConsumerGroupSession, ConsumerGroupClaim) error
}

ConsumerGroupHandler instances are used to handle individual topic/partition claims. It also provides hooks for your consumer group session life-cycle and allow you to trigger logic before or after the consume loop(s).

PLEASE NOTE that handlers are likely be called from several goroutines concurrently, ensure that all state is safely protected against race conditions.

type ConsumerGroupMemberAssignment

type ConsumerGroupMemberAssignment struct {
	Version  int16
	Topics   map[string][]int32
	UserData []byte
}

ConsumerGroupMemberAssignment holds the member assignment for a consume group

type ConsumerGroupMemberMetadata

type ConsumerGroupMemberMetadata struct {
	Version  int16
	Topics   []string
	UserData []byte
}

ConsumerGroupMemberMetadata holds the metadata for consumer group

type ConsumerGroupSession added in v1.19.0

type ConsumerGroupSession interface {
	// Claims returns information about the claimed partitions by topic.
	Claims() map[string][]int32

	// MemberID returns the cluster member ID.
	MemberID() string

	// GenerationID returns the current generation ID.
	GenerationID() int32

	// MarkOffset marks the provided offset, alongside a metadata string
	// that represents the state of the partition consumer at that point in time. The
	// metadata string can be used by another consumer to restore that state, so it
	// can resume consumption.
	//
	// To follow upstream conventions, you are expected to mark the offset of the
	// next message to read, not the last message read. Thus, when calling `MarkOffset`
	// you should typically add one to the offset of the last consumed message.
	//
	// Note: calling MarkOffset does not necessarily commit the offset to the backend
	// store immediately for efficiency reasons, and it may never be committed if
	// your application crashes. This means that you may end up processing the same
	// message twice, and your processing should ideally be idempotent.
	MarkOffset(topic string, partition int32, offset int64, metadata string)

	// ResetOffset resets to the provided offset, alongside a metadata string that
	// represents the state of the partition consumer at that point in time. Reset
	// acts as a counterpart to MarkOffset, the difference being that it allows to
	// reset an offset to an earlier or smaller value, where MarkOffset only
	// allows incrementing the offset. cf MarkOffset for more details.
	ResetOffset(topic string, partition int32, offset int64, metadata string)

	// MarkMessage marks a message as consumed.
	MarkMessage(msg *ConsumerMessage, metadata string)

	// Context returns the session context.
	Context() context.Context
}

ConsumerGroupSession represents a consumer group member session.

type ConsumerMessage

type ConsumerMessage struct {
	Key, Value     []byte
	Topic          string
	Partition      int32
	Offset         int64
	Timestamp      time.Time       // only set if kafka is version 0.10+, inner message timestamp
	BlockTimestamp time.Time       // only set if kafka is version 0.10+, outer (compressed) block timestamp
	Headers        []*RecordHeader // only set if kafka is version 0.11+
}

ConsumerMessage encapsulates a Kafka message returned by the consumer.

type ConsumerMetadataRequest

type ConsumerMetadataRequest struct {
	ConsumerGroup string
}

ConsumerMetadataRequest is used for metadata requests

type ConsumerMetadataResponse

type ConsumerMetadataResponse struct {
	Err             KError
	Coordinator     *Broker
	CoordinatorID   int32  // deprecated: use Coordinator.ID()
	CoordinatorHost string // deprecated: use Coordinator.Addr()
	CoordinatorPort int32  // deprecated: use Coordinator.Addr()
}

ConsumerMetadataResponse holds the response for a consumer group meta data requests

type CoordinatorType added in v1.17.0

type CoordinatorType int8
const (
	CoordinatorGroup       CoordinatorType = 0
	CoordinatorTransaction CoordinatorType = 1
)

type CreateAclsRequest added in v1.16.0

type CreateAclsRequest struct {
	AclCreations []*AclCreation
}

type CreateAclsResponse added in v1.16.0

type CreateAclsResponse struct {
	ThrottleTime         time.Duration
	AclCreationResponses []*AclCreationResponse
}

type CreatePartitionsRequest added in v1.15.0

type CreatePartitionsRequest struct {
	TopicPartitions map[string]*TopicPartition
	Timeout         time.Duration
	ValidateOnly    bool
}

type CreatePartitionsResponse added in v1.15.0

type CreatePartitionsResponse struct {
	ThrottleTime         time.Duration
	TopicPartitionErrors map[string]*TopicPartitionError
}

type CreateTopicsRequest added in v1.16.0

type CreateTopicsRequest struct {
	Version int16

	TopicDetails map[string]*TopicDetail
	Timeout      time.Duration
	ValidateOnly bool
}

type CreateTopicsResponse added in v1.16.0

type CreateTopicsResponse struct {
	Version      int16
	ThrottleTime time.Duration
	TopicErrors  map[string]*TopicError
}

type DeleteAclsRequest added in v1.16.0

type DeleteAclsRequest struct {
	Filters []*AclFilter
}

type DeleteAclsResponse added in v1.16.0

type DeleteAclsResponse struct {
	ThrottleTime    time.Duration
	FilterResponses []*FilterResponse
}

type DeleteGroupsRequest added in v1.17.0

type DeleteGroupsRequest struct {
	Groups []string
}

func (*DeleteGroupsRequest) AddGroup added in v1.17.0

func (r *DeleteGroupsRequest) AddGroup(group string)

type DeleteGroupsResponse added in v1.17.0

type DeleteGroupsResponse struct {
	ThrottleTime    time.Duration
	GroupErrorCodes map[string]KError
}

type DeleteRecordsRequest added in v1.17.0

type DeleteRecordsRequest struct {
	Topics  map[string]*DeleteRecordsRequestTopic
	Timeout time.Duration
}

type DeleteRecordsRequestTopic added in v1.17.0

type DeleteRecordsRequestTopic struct {
	PartitionOffsets map[int32]int64 // partition => offset
}

type DeleteRecordsResponse added in v1.17.0

type DeleteRecordsResponse struct {
	Version      int16
	ThrottleTime time.Duration
	Topics       map[string]*DeleteRecordsResponseTopic
}

type DeleteRecordsResponsePartition added in v1.17.0

type DeleteRecordsResponsePartition struct {
	LowWatermark int64
	Err          KError
}

type DeleteRecordsResponseTopic added in v1.17.0

type DeleteRecordsResponseTopic struct {
	Partitions map[int32]*DeleteRecordsResponsePartition
}

type DeleteTopicsRequest added in v1.16.0

type DeleteTopicsRequest struct {
	Version int16
	Topics  []string
	Timeout time.Duration
}

type DeleteTopicsResponse added in v1.16.0

type DeleteTopicsResponse struct {
	Version         int16
	ThrottleTime    time.Duration
	TopicErrorCodes map[string]KError
}

type DescribeAclsRequest added in v1.16.0

type DescribeAclsRequest struct {
	AclFilter
}

type DescribeAclsResponse added in v1.16.0

type DescribeAclsResponse struct {
	ThrottleTime time.Duration
	Err          KError
	ErrMsg       *string
	ResourceAcls []*ResourceAcls
}

type DescribeConfigsRequest added in v1.16.0

type DescribeConfigsRequest struct {
	Version         int16
	Resources       []*ConfigResource
	IncludeSynonyms bool
}

type DescribeConfigsResponse added in v1.16.0

type DescribeConfigsResponse struct {
	Version      int16
	ThrottleTime time.Duration
	Resources    []*ResourceResponse
}

type DescribeGroupsRequest

type DescribeGroupsRequest struct {
	Groups []string
}

func (*DescribeGroupsRequest) AddGroup

func (r *DescribeGroupsRequest) AddGroup(group string)

type DescribeGroupsResponse

type DescribeGroupsResponse struct {
	Groups []*GroupDescription
}

type DynamicConsistencyPartitioner added in v1.18.0

type DynamicConsistencyPartitioner interface {
	Partitioner

	// MessageRequiresConsistency is similar to Partitioner.RequiresConsistency,
	// but takes in the message being partitioned so that the partitioner can
	// make a per-message determination.
	MessageRequiresConsistency(message *ProducerMessage) bool
}

DynamicConsistencyPartitioner can optionally be implemented by Partitioners in order to allow more flexibility than is originally allowed by the RequiresConsistency method in the Partitioner interface. This allows partitioners to require consistency sometimes, but not all times. It's useful for, e.g., the HashPartitioner, which does not require consistency if the message key is nil.

type Encoder

type Encoder interface {
	Encode() ([]byte, error)
	Length() int
}

Encoder is a simple interface for any type that can be encoded as an array of bytes in order to be sent as the key or value of a Kafka message. Length() is provided as an optimization, and must return the same as len() on the result of Encode().

type EndTxnRequest added in v1.16.0

type EndTxnRequest struct {
	TransactionalID   string
	ProducerID        int64
	ProducerEpoch     int16
	TransactionResult bool
}

type EndTxnResponse added in v1.16.0

type EndTxnResponse struct {
	ThrottleTime time.Duration
	Err          KError
}

type FetchRequest

type FetchRequest struct {
	MaxWaitTime int32
	MinBytes    int32
	MaxBytes    int32
	Version     int16
	Isolation   IsolationLevel
	// contains filtered or unexported fields
}

FetchRequest (API key 1) will fetch Kafka messages. Version 3 introduced the MaxBytes field. See https://issues.apache.org/jira/browse/KAFKA-2063 for a discussion of the issues leading up to that. The KIP is at https://cwiki.apache.org/confluence/display/KAFKA/KIP-74%3A+Add+Fetch+Response+Size+Limit+in+Bytes

func (*FetchRequest) AddBlock

func (r *FetchRequest) AddBlock(topic string, partitionID int32, fetchOffset int64, maxBytes int32)

type FetchResponse

type FetchResponse struct {
	Blocks       map[string]map[int32]*FetchResponseBlock
	ThrottleTime time.Duration
	Version      int16 // v1 requires 0.9+, v2 requires 0.10+
}

func (*FetchResponse) AddError

func (r *FetchResponse) AddError(topic string, partition int32, err KError)

func (*FetchResponse) AddMessage

func (r *FetchResponse) AddMessage(topic string, partition int32, key, value Encoder, offset int64)

func (*FetchResponse) AddRecord added in v1.14.0

func (r *FetchResponse) AddRecord(topic string, partition int32, key, value Encoder, offset int64)

func (*FetchResponse) GetBlock

func (r *FetchResponse) GetBlock(topic string, partition int32) *FetchResponseBlock

func (*FetchResponse) SetLastOffsetDelta added in v1.16.0

func (r *FetchResponse) SetLastOffsetDelta(topic string, partition int32, offset int32)

func (*FetchResponse) SetLastStableOffset added in v1.14.0

func (r *FetchResponse) SetLastStableOffset(topic string, partition int32, offset int64)

type FetchResponseBlock

type FetchResponseBlock struct {
	Err                 KError
	HighWaterMarkOffset int64
	LastStableOffset    int64
	AbortedTransactions []*AbortedTransaction
	Records             *Records // deprecated: use FetchResponseBlock.RecordsSet
	RecordsSet          []*Records
	Partial             bool
}

type FilterResponse added in v1.16.0

type FilterResponse struct {
	Err          KError
	ErrMsg       *string
	MatchingAcls []*MatchingAcl
}

type FindCoordinatorRequest added in v1.17.0

type FindCoordinatorRequest struct {
	Version         int16
	CoordinatorKey  string
	CoordinatorType CoordinatorType
}

type FindCoordinatorResponse added in v1.17.0

type FindCoordinatorResponse struct {
	Version      int16
	ThrottleTime time.Duration
	Err          KError
	ErrMsg       *string
	Coordinator  *Broker
}

type GroupDescription

type GroupDescription struct {
	Err          KError
	GroupId      string
	State        string
	ProtocolType string
	Protocol     string
	Members      map[string]*GroupMemberDescription
}

type GroupMemberDescription

type GroupMemberDescription struct {
	ClientId         string
	ClientHost       string
	MemberMetadata   []byte
	MemberAssignment []byte
}

func (*GroupMemberDescription) GetMemberAssignment

func (gmd *GroupMemberDescription) GetMemberAssignment() (*ConsumerGroupMemberAssignment, error)

func (*GroupMemberDescription) GetMemberMetadata

func (gmd *GroupMemberDescription) GetMemberMetadata() (*ConsumerGroupMemberMetadata, error)

type GroupProtocol

type GroupProtocol struct {
	Name     string
	Metadata []byte
}

type HashPartitionerOption added in v1.18.0

type HashPartitionerOption func(*hashPartitioner)

HashPartitionOption lets you modify default values of the partitioner

func WithAbsFirst added in v1.18.0

func WithAbsFirst() HashPartitionerOption

WithAbsFirst means that the partitioner handles absolute values in the same way as the reference Java implementation

func WithCustomFallbackPartitioner added in v1.18.0

func WithCustomFallbackPartitioner(randomHP *hashPartitioner) HashPartitionerOption

WithCustomFallbackPartitioner lets you specify what HashPartitioner should be used in case a Distribution Key is empty

func WithCustomHashFunction added in v1.18.0

func WithCustomHashFunction(hasher func() hash.Hash32) HashPartitionerOption

WithCustomHashFunction lets you specify what hash function to use for the partitioning

type HeartbeatRequest

type HeartbeatRequest struct {
	GroupId      string
	GenerationId int32
	MemberId     string
}

type HeartbeatResponse

type HeartbeatResponse struct {
	Err KError
}

type InitProducerIDRequest added in v1.16.0

type InitProducerIDRequest struct {
	TransactionalID    *string
	TransactionTimeout time.Duration
}

type InitProducerIDResponse added in v1.16.0

type InitProducerIDResponse struct {
	ThrottleTime  time.Duration
	Err           KError
	ProducerID    int64
	ProducerEpoch int16
}

type IsolationLevel added in v1.14.0

type IsolationLevel int8
const (
	ReadUncommitted IsolationLevel = 0
	ReadCommitted   IsolationLevel = 1
)

type JoinGroupRequest

type JoinGroupRequest struct {
	Version               int16
	GroupId               string
	SessionTimeout        int32
	RebalanceTimeout      int32
	MemberId              string
	ProtocolType          string
	GroupProtocols        map[string][]byte // deprecated; use OrderedGroupProtocols
	OrderedGroupProtocols []*GroupProtocol
}

func (*JoinGroupRequest) AddGroupProtocol

func (r *JoinGroupRequest) AddGroupProtocol(name string, metadata []byte)

func (*JoinGroupRequest) AddGroupProtocolMetadata

func (r *JoinGroupRequest) AddGroupProtocolMetadata(name string, metadata *ConsumerGroupMemberMetadata) error

type JoinGroupResponse

type JoinGroupResponse struct {
	Version       int16
	ThrottleTime  int32
	Err           KError
	GenerationId  int32
	GroupProtocol string
	LeaderId      string
	MemberId      string
	Members       map[string][]byte
}

func (*JoinGroupResponse) GetMembers

type KError

type KError int16

KError is the type of error that can be returned directly by the Kafka broker. See https://cwiki.apache.org/confluence/display/KAFKA/A+Guide+To+The+Kafka+Protocol#AGuideToTheKafkaProtocol-ErrorCodes

const (
	ErrNoError                            KError = 0
	ErrUnknown                            KError = -1
	ErrOffsetOutOfRange                   KError = 1
	ErrInvalidMessage                     KError = 2
	ErrUnknownTopicOrPartition            KError = 3
	ErrInvalidMessageSize                 KError = 4
	ErrLeaderNotAvailable                 KError = 5
	ErrNotLeaderForPartition              KError = 6
	ErrRequestTimedOut                    KError = 7
	ErrBrokerNotAvailable                 KError = 8
	ErrReplicaNotAvailable                KError = 9
	ErrMessageSizeTooLarge                KError = 10
	ErrStaleControllerEpochCode           KError = 11
	ErrOffsetMetadataTooLarge             KError = 12
	ErrNetworkException                   KError = 13
	ErrOffsetsLoadInProgress              KError = 14
	ErrConsumerCoordinatorNotAvailable    KError = 15
	ErrNotCoordinatorForConsumer          KError = 16
	ErrInvalidTopic                       KError = 17
	ErrMessageSetSizeTooLarge             KError = 18
	ErrNotEnoughReplicas                  KError = 19
	ErrNotEnoughReplicasAfterAppend       KError = 20
	ErrInvalidRequiredAcks                KError = 21
	ErrIllegalGeneration                  KError = 22
	ErrInconsistentGroupProtocol          KError = 23
	ErrInvalidGroupId                     KError = 24
	ErrUnknownMemberId                    KError = 25
	ErrInvalidSessionTimeout              KError = 26
	ErrRebalanceInProgress                KError = 27
	ErrInvalidCommitOffsetSize            KError = 28
	ErrTopicAuthorizationFailed           KError = 29
	ErrGroupAuthorizationFailed           KError = 30
	ErrClusterAuthorizationFailed         KError = 31
	ErrInvalidTimestamp                   KError = 32
	ErrUnsupportedSASLMechanism           KError = 33
	ErrIllegalSASLState                   KError = 34
	ErrUnsupportedVersion                 KError = 35
	ErrTopicAlreadyExists                 KError = 36
	ErrInvalidPartitions                  KError = 37
	ErrInvalidReplicationFactor           KError = 38
	ErrInvalidReplicaAssignment           KError = 39
	ErrInvalidConfig                      KError = 40
	ErrNotController                      KError = 41
	ErrInvalidRequest                     KError = 42
	ErrUnsupportedForMessageFormat        KError = 43
	ErrPolicyViolation                    KError = 44
	ErrOutOfOrderSequenceNumber           KError = 45
	ErrDuplicateSequenceNumber            KError = 46
	ErrInvalidProducerEpoch               KError = 47
	ErrInvalidTxnState                    KError = 48
	ErrInvalidProducerIDMapping           KError = 49
	ErrInvalidTransactionTimeout          KError = 50
	ErrConcurrentTransactions             KError = 51
	ErrTransactionCoordinatorFenced       KError = 52
	ErrTransactionalIDAuthorizationFailed KError = 53
	ErrSecurityDisabled                   KError = 54
	ErrOperationNotAttempted              KError = 55
	ErrKafkaStorageError                  KError = 56
	ErrLogDirNotFound                     KError = 57
	ErrSASLAuthenticationFailed           KError = 58
	ErrUnknownProducerID                  KError = 59
	ErrReassignmentInProgress             KError = 60
	ErrDelegationTokenAuthDisabled        KError = 61
	ErrDelegationTokenNotFound            KError = 62
	ErrDelegationTokenOwnerMismatch       KError = 63
	ErrDelegationTokenRequestNotAllowed   KError = 64
	ErrDelegationTokenAuthorizationFailed KError = 65
	ErrDelegationTokenExpired             KError = 66
	ErrInvalidPrincipalType               KError = 67
	ErrNonEmptyGroup                      KError = 68
	ErrGroupIDNotFound                    KError = 69
	ErrFetchSessionIDNotFound             KError = 70
	ErrInvalidFetchSessionEpoch           KError = 71
	ErrListenerNotFound                   KError = 72
)

Numeric error codes returned by the Kafka server.

func (KError) Error

func (err KError) Error() string

type KafkaVersion

type KafkaVersion struct {
	// contains filtered or unexported fields
}

KafkaVersion instances represent versions of the upstream Kafka broker.

func ParseKafkaVersion added in v1.15.0

func ParseKafkaVersion(s string) (KafkaVersion, error)

func (KafkaVersion) IsAtLeast

func (v KafkaVersion) IsAtLeast(other KafkaVersion) bool

IsAtLeast return true if and only if the version it is called on is greater than or equal to the version passed in:

V1.IsAtLeast(V2) // false
V2.IsAtLeast(V1) // true

func (KafkaVersion) String added in v1.15.0

func (v KafkaVersion) String() string

type LeaveGroupRequest

type LeaveGroupRequest struct {
	GroupId  string
	MemberId string
}

type LeaveGroupResponse

type LeaveGroupResponse struct {
	Err KError
}

type ListGroupsRequest

type ListGroupsRequest struct {
}

type ListGroupsResponse

type ListGroupsResponse struct {
	Err    KError
	Groups map[string]string
}

type MatchingAcl added in v1.16.0

type MatchingAcl struct {
	Err    KError
	ErrMsg *string
	Resource
	Acl
}

type Message

type Message struct {
	Codec            CompressionCodec // codec used to compress the message contents
	CompressionLevel int              // compression level
	Key              []byte           // the message key, may be nil
	Value            []byte           // the message contents
	Set              *MessageSet      // the message set a message might wrap
	Version          int8             // v1 requires Kafka 0.10
	Timestamp        time.Time        // the timestamp of the message (version 1+ only)
	// contains filtered or unexported fields
}

type MessageBlock

type MessageBlock struct {
	Offset int64
	Msg    *Message
}

func (*MessageBlock) Messages

func (msb *MessageBlock) Messages() []*MessageBlock

Messages convenience helper which returns either all the messages that are wrapped in this block

type MessageSet

type MessageSet struct {
	PartialTrailingMessage bool // whether the set on the wire contained an incomplete trailing MessageBlock
	OverflowMessage        bool // whether the set on the wire contained an overflow message
	Messages               []*MessageBlock
}

type MetadataRequest

type MetadataRequest struct {
	Version                int16
	Topics                 []string
	AllowAutoTopicCreation bool
}

type MetadataResponse

type MetadataResponse struct {
	Version        int16
	ThrottleTimeMs int32
	Brokers        []*Broker
	ClusterID      *string
	ControllerID   int32
	Topics         []*TopicMetadata
}

func (*MetadataResponse) AddBroker

func (r *MetadataResponse) AddBroker(addr string, id int32)

func (*MetadataResponse) AddTopic

func (r *MetadataResponse) AddTopic(topic string, err KError) *TopicMetadata

func (*MetadataResponse) AddTopicPartition

func (r *MetadataResponse) AddTopicPartition(topic string, partition, brokerID int32, replicas, isr []int32, err KError)

type MockAlterConfigsResponse added in v1.18.0

type MockAlterConfigsResponse struct {
	// contains filtered or unexported fields
}

func NewMockAlterConfigsResponse added in v1.18.0

func NewMockAlterConfigsResponse(t TestReporter) *MockAlterConfigsResponse

func (*MockAlterConfigsResponse) For added in v1.18.0

func (mr *MockAlterConfigsResponse) For(reqBody versionedDecoder) encoder

type MockBroker

type MockBroker struct {
	// contains filtered or unexported fields
}

MockBroker is a mock Kafka broker that is used in unit tests. It is exposed to facilitate testing of higher level or specialized consumers and producers built on top of Sarama. Note that it does not 'mimic' the Kafka API protocol, but rather provides a facility to do that. It takes care of the TCP transport, request unmarshaling, response marshaling, and makes it the test writer responsibility to program correct according to the Kafka API protocol MockBroker behaviour.

MockBroker is implemented as a TCP server listening on a kernel-selected localhost port that can accept many connections. It reads Kafka requests from that connection and returns responses programmed by the SetHandlerByMap function. If a MockBroker receives a request that it has no programmed response for, then it returns nothing and the request times out.

A set of MockRequest builders to define mappings used by MockBroker is provided by Sarama. But users can develop MockRequests of their own and use them along with or instead of the standard ones.

When running tests with MockBroker it is strongly recommended to specify a timeout to `go test` so that if the broker hangs waiting for a response, the test panics.

It is not necessary to prefix message length or correlation ID to your response bytes, the server does that automatically as a convenience.

func NewMockBroker

func NewMockBroker(t TestReporter, brokerID int32) *MockBroker

NewMockBroker launches a fake Kafka broker. It takes a TestReporter as provided by the test framework and a channel of responses to use. If an error occurs it is simply logged to the TestReporter and the broker exits.

func NewMockBrokerAddr

func NewMockBrokerAddr(t TestReporter, brokerID int32, addr string) *MockBroker

NewMockBrokerAddr behaves like newMockBroker but listens on the address you give it rather than just some ephemeral port.

func NewMockBrokerListener added in v1.16.0

func NewMockBrokerListener(t TestReporter, brokerID int32, listener net.Listener) *MockBroker

NewMockBrokerListener behaves like newMockBrokerAddr but accepts connections on the listener specified.

func (*MockBroker) Addr

func (b *MockBroker) Addr() string

Addr returns the broker connection string in the form "<address>:<port>".

func (*MockBroker) BrokerID

func (b *MockBroker) BrokerID() int32

BrokerID returns broker ID assigned to the broker.

func (*MockBroker) Close

func (b *MockBroker) Close()

Close terminates the broker blocking until it stops internal goroutines and releases all resources.

func (*MockBroker) History

func (b *MockBroker) History() []RequestResponse

History returns a slice of RequestResponse pairs in the order they were processed by the broker. Note that in case of multiple connections to the broker the order expected by a test can be different from the order recorded in the history, unless some synchronization is implemented in the test.

func (*MockBroker) Port

func (b *MockBroker) Port() int32

Port returns the TCP port number the broker is listening for requests on.

func (*MockBroker) Returns

func (b *MockBroker) Returns(e encoder)

func (*MockBroker) SetHandlerByMap

func (b *MockBroker) SetHandlerByMap(handlerMap map[string]MockResponse)

SetHandlerByMap defines mapping of Request types to MockResponses. When a request is received by the broker, it looks up the request type in the map and uses the found MockResponse instance to generate an appropriate reply. If the request type is not found in the map then nothing is sent.

func (*MockBroker) SetLatency

func (b *MockBroker) SetLatency(latency time.Duration)

SetLatency makes broker pause for the specified period every time before replying.

func (*MockBroker) SetNotifier

func (b *MockBroker) SetNotifier(notifier RequestNotifierFunc)

SetNotifier set a function that will get invoked whenever a request has been processed successfully and will provide the number of bytes read and written

type MockConsumerMetadataResponse

type MockConsumerMetadataResponse struct {
	// contains filtered or unexported fields
}

MockConsumerMetadataResponse is a `ConsumerMetadataResponse` builder.

func NewMockConsumerMetadataResponse

func NewMockConsumerMetadataResponse(t TestReporter) *MockConsumerMetadataResponse

func (*MockConsumerMetadataResponse) For

func (mr *MockConsumerMetadataResponse) For(reqBody versionedDecoder) encoder

func (*MockConsumerMetadataResponse) SetCoordinator

func (mr *MockConsumerMetadataResponse) SetCoordinator(group string, broker *MockBroker) *MockConsumerMetadataResponse

func (*MockConsumerMetadataResponse) SetError

type MockCreateAclsResponse added in v1.18.0

type MockCreateAclsResponse struct {
	// contains filtered or unexported fields
}

func NewMockCreateAclsResponse added in v1.18.0

func NewMockCreateAclsResponse(t TestReporter) *MockCreateAclsResponse

func (*MockCreateAclsResponse) For added in v1.18.0

func (mr *MockCreateAclsResponse) For(reqBody versionedDecoder) encoder

type MockCreatePartitionsResponse added in v1.18.0

type MockCreatePartitionsResponse struct {
	// contains filtered or unexported fields
}

func NewMockCreatePartitionsResponse added in v1.18.0

func NewMockCreatePartitionsResponse(t TestReporter) *MockCreatePartitionsResponse

func (*MockCreatePartitionsResponse) For added in v1.18.0

func (mr *MockCreatePartitionsResponse) For(reqBody versionedDecoder) encoder

type MockCreateTopicsResponse added in v1.18.0

type MockCreateTopicsResponse struct {
	// contains filtered or unexported fields
}

func NewMockCreateTopicsResponse added in v1.18.0

func NewMockCreateTopicsResponse(t TestReporter) *MockCreateTopicsResponse

func (*MockCreateTopicsResponse) For added in v1.18.0

func (mr *MockCreateTopicsResponse) For(reqBody versionedDecoder) encoder

type MockDeleteAclsResponse added in v1.18.0

type MockDeleteAclsResponse struct {
	// contains filtered or unexported fields
}

func NewMockDeleteAclsResponse added in v1.18.0

func NewMockDeleteAclsResponse(t TestReporter) *MockDeleteAclsResponse

func (*MockDeleteAclsResponse) For added in v1.18.0

func (mr *MockDeleteAclsResponse) For(reqBody versionedDecoder) encoder

type MockDeleteRecordsResponse added in v1.18.0

type MockDeleteRecordsResponse struct {
	// contains filtered or unexported fields
}

func NewMockDeleteRecordsResponse added in v1.18.0

func NewMockDeleteRecordsResponse(t TestReporter) *MockDeleteRecordsResponse

func (*MockDeleteRecordsResponse) For added in v1.18.0

func (mr *MockDeleteRecordsResponse) For(reqBody versionedDecoder) encoder

type MockDeleteTopicsResponse added in v1.18.0

type MockDeleteTopicsResponse struct {
	// contains filtered or unexported fields
}

func NewMockDeleteTopicsResponse added in v1.18.0

func NewMockDeleteTopicsResponse(t TestReporter) *MockDeleteTopicsResponse

func (*MockDeleteTopicsResponse) For added in v1.18.0

func (mr *MockDeleteTopicsResponse) For(reqBody versionedDecoder) encoder

type MockDescribeConfigsResponse added in v1.18.0

type MockDescribeConfigsResponse struct {
	// contains filtered or unexported fields
}

func NewMockDescribeConfigsResponse added in v1.18.0

func NewMockDescribeConfigsResponse(t TestReporter) *MockDescribeConfigsResponse

func (*MockDescribeConfigsResponse) For added in v1.18.0

func (mr *MockDescribeConfigsResponse) For(reqBody versionedDecoder) encoder

type MockFetchResponse

type MockFetchResponse struct {
	// contains filtered or unexported fields
}

MockFetchResponse is a `FetchResponse` builder.

func NewMockFetchResponse

func NewMockFetchResponse(t TestReporter, batchSize int) *MockFetchResponse

func (*MockFetchResponse) For

func (mfr *MockFetchResponse) For(reqBody versionedDecoder) encoder

func (*MockFetchResponse) SetHighWaterMark

func (mfr *MockFetchResponse) SetHighWaterMark(topic string, partition int32, offset int64) *MockFetchResponse

func (*MockFetchResponse) SetMessage

func (mfr *MockFetchResponse) SetMessage(topic string, partition int32, offset int64, msg Encoder) *MockFetchResponse

func (*MockFetchResponse) SetVersion added in v1.13.0

func (mfr *MockFetchResponse) SetVersion(version int16) *MockFetchResponse

type MockFindCoordinatorResponse added in v1.17.0

type MockFindCoordinatorResponse struct {
	// contains filtered or unexported fields
}

MockFindCoordinatorResponse is a `FindCoordinatorResponse` builder.

func NewMockFindCoordinatorResponse added in v1.17.0

func NewMockFindCoordinatorResponse(t TestReporter) *MockFindCoordinatorResponse

func (*MockFindCoordinatorResponse) For added in v1.17.0

func (mr *MockFindCoordinatorResponse) For(reqBody versionedDecoder) encoder

func (*MockFindCoordinatorResponse) SetCoordinator added in v1.17.0

func (mr *MockFindCoordinatorResponse) SetCoordinator(coordinatorType CoordinatorType, group string, broker *MockBroker) *MockFindCoordinatorResponse

func (*MockFindCoordinatorResponse) SetError added in v1.17.0

func (mr *MockFindCoordinatorResponse) SetError(coordinatorType CoordinatorType, group string, kerror KError) *MockFindCoordinatorResponse

type MockListAclsResponse added in v1.18.0

type MockListAclsResponse struct {
	// contains filtered or unexported fields
}

func NewMockListAclsResponse added in v1.18.0

func NewMockListAclsResponse(t TestReporter) *MockListAclsResponse

func (*MockListAclsResponse) For added in v1.18.0

func (mr *MockListAclsResponse) For(reqBody versionedDecoder) encoder

type MockMetadataResponse

type MockMetadataResponse struct {
	// contains filtered or unexported fields
}

MockMetadataResponse is a `MetadataResponse` builder.

func NewMockMetadataResponse

func NewMockMetadataResponse(t TestReporter) *MockMetadataResponse

func (*MockMetadataResponse) For

func (mmr *MockMetadataResponse) For(reqBody versionedDecoder) encoder

func (*MockMetadataResponse) SetBroker

func (mmr *MockMetadataResponse) SetBroker(addr string, brokerID int32) *MockMetadataResponse

func (*MockMetadataResponse) SetController added in v1.17.0

func (mmr *MockMetadataResponse) SetController(brokerID int32) *MockMetadataResponse

func (*MockMetadataResponse) SetLeader

func (mmr *MockMetadataResponse) SetLeader(topic string, partition, brokerID int32) *MockMetadataResponse

type MockOffsetCommitResponse

type MockOffsetCommitResponse struct {
	// contains filtered or unexported fields
}

MockOffsetCommitResponse is a `OffsetCommitResponse` builder.

func NewMockOffsetCommitResponse

func NewMockOffsetCommitResponse(t TestReporter) *MockOffsetCommitResponse

func (*MockOffsetCommitResponse) For

func (mr *MockOffsetCommitResponse) For(reqBody versionedDecoder) encoder

func (*MockOffsetCommitResponse) SetError

func (mr *MockOffsetCommitResponse) SetError(group, topic string, partition int32, kerror KError) *MockOffsetCommitResponse

type MockOffsetFetchResponse

type MockOffsetFetchResponse struct {
	// contains filtered or unexported fields
}

MockOffsetFetchResponse is a `OffsetFetchResponse` builder.

func NewMockOffsetFetchResponse

func NewMockOffsetFetchResponse(t TestReporter) *MockOffsetFetchResponse

func (*MockOffsetFetchResponse) For

func (mr *MockOffsetFetchResponse) For(reqBody versionedDecoder) encoder

func (*MockOffsetFetchResponse) SetOffset

func (mr *MockOffsetFetchResponse) SetOffset(group, topic string, partition int32, offset int64, metadata string, kerror KError) *MockOffsetFetchResponse

type MockOffsetResponse

type MockOffsetResponse struct {
	// contains filtered or unexported fields
}

MockOffsetResponse is an `OffsetResponse` builder.

func NewMockOffsetResponse

func NewMockOffsetResponse(t TestReporter) *MockOffsetResponse

func (*MockOffsetResponse) For

func (mor *MockOffsetResponse) For(reqBody versionedDecoder) encoder

func (*MockOffsetResponse) SetOffset

func (mor *MockOffsetResponse) SetOffset(topic string, partition int32, time, offset int64) *MockOffsetResponse

func (*MockOffsetResponse) SetVersion added in v1.14.0

func (mor *MockOffsetResponse) SetVersion(version int16) *MockOffsetResponse

type MockProduceResponse

type MockProduceResponse struct {
	// contains filtered or unexported fields
}

MockProduceResponse is a `ProduceResponse` builder.

func NewMockProduceResponse

func NewMockProduceResponse(t TestReporter) *MockProduceResponse

func (*MockProduceResponse) For

func (mr *MockProduceResponse) For(reqBody versionedDecoder) encoder

func (*MockProduceResponse) SetError

func (mr *MockProduceResponse) SetError(topic string, partition int32, kerror KError) *MockProduceResponse

func (*MockProduceResponse) SetVersion added in v1.16.0

func (mr *MockProduceResponse) SetVersion(version int16) *MockProduceResponse

type MockResponse

type MockResponse interface {
	For(reqBody versionedDecoder) (res encoder)
}

MockResponse is a response builder interface it defines one method that allows generating a response based on a request body. MockResponses are used to program behavior of MockBroker in tests.

type MockSequence

type MockSequence struct {
	// contains filtered or unexported fields
}

MockSequence is a mock response builder that is created from a sequence of concrete responses. Every time when a `MockBroker` calls its `For` method the next response from the sequence is returned. When the end of the sequence is reached the last element from the sequence is returned.

func NewMockSequence

func NewMockSequence(responses ...interface{}) *MockSequence

func (*MockSequence) For

func (mc *MockSequence) For(reqBody versionedDecoder) (res encoder)

type MockWrapper

type MockWrapper struct {
	// contains filtered or unexported fields
}

MockWrapper is a mock response builder that returns a particular concrete response regardless of the actual request passed to the `For` method.

func NewMockWrapper

func NewMockWrapper(res encoder) *MockWrapper

func (*MockWrapper) For

func (mw *MockWrapper) For(reqBody versionedDecoder) (res encoder)

type OffsetCommitRequest

type OffsetCommitRequest struct {
	ConsumerGroup           string
	ConsumerGroupGeneration int32  // v1 or later
	ConsumerID              string // v1 or later
	RetentionTime           int64  // v2 or later

	// Version can be:
	// - 0 (kafka 0.8.1 and later)
	// - 1 (kafka 0.8.2 and later)
	// - 2 (kafka 0.9.0 and later)
	// - 3 (kafka 0.11.0 and later)
	// - 4 (kafka 2.0.0 and later)
	Version int16
	// contains filtered or unexported fields
}

func (*OffsetCommitRequest) AddBlock

func (r *OffsetCommitRequest) AddBlock(topic string, partitionID int32, offset int64, timestamp int64, metadata string)

func (*OffsetCommitRequest) Offset added in v1.17.0

func (r *OffsetCommitRequest) Offset(topic string, partitionID int32) (int64, string, error)

type OffsetCommitResponse

type OffsetCommitResponse struct {
	Version        int16
	ThrottleTimeMs int32
	Errors         map[string]map[int32]KError
}

func (*OffsetCommitResponse) AddError

func (r *OffsetCommitResponse) AddError(topic string, partition int32, kerror KError)

type OffsetFetchRequest

type OffsetFetchRequest struct {
	Version       int16
	ConsumerGroup string
	// contains filtered or unexported fields
}

func (*OffsetFetchRequest) AddPartition

func (r *OffsetFetchRequest) AddPartition(topic string, partitionID int32)

func (*OffsetFetchRequest) ZeroPartitions added in v1.20.0

func (r *OffsetFetchRequest) ZeroPartitions()

type OffsetFetchResponse

type OffsetFetchResponse struct {
	Version        int16
	ThrottleTimeMs int32
	Blocks         map[string]map[int32]*OffsetFetchResponseBlock
	Err            KError
}

func (*OffsetFetchResponse) AddBlock

func (r *OffsetFetchResponse) AddBlock(topic string, partition int32, block *OffsetFetchResponseBlock)

func (*OffsetFetchResponse) GetBlock

func (r *OffsetFetchResponse) GetBlock(topic string, partition int32) *OffsetFetchResponseBlock

type OffsetFetchResponseBlock

type OffsetFetchResponseBlock struct {
	Offset      int64
	LeaderEpoch int32
	Metadata    string
	Err         KError
}

type OffsetManager

type OffsetManager interface {
	// ManagePartition creates a PartitionOffsetManager on the given topic/partition.
	// It will return an error if this OffsetManager is already managing the given
	// topic/partition.
	ManagePartition(topic string, partition int32) (PartitionOffsetManager, error)

	// Close stops the OffsetManager from managing offsets. It is required to call
	// this function before an OffsetManager object passes out of scope, as it
	// will otherwise leak memory. You must call this after all the
	// PartitionOffsetManagers are closed.
	Close() error
}

OffsetManager uses Kafka to store and fetch consumed partition offsets.

func NewOffsetManagerFromClient

func NewOffsetManagerFromClient(group string, client Client) (OffsetManager, error)

NewOffsetManagerFromClient creates a new OffsetManager from the given client. It is still necessary to call Close() on the underlying client when finished with the partition manager.

type OffsetRequest

type OffsetRequest struct {
	Version int16
	// contains filtered or unexported fields
}

func (*OffsetRequest) AddBlock

func (r *OffsetRequest) AddBlock(topic string, partitionID int32, time int64, maxOffsets int32)

func (*OffsetRequest) ReplicaID added in v1.20.1

func (r *OffsetRequest) ReplicaID() int32

func (*OffsetRequest) SetReplicaID added in v1.20.1

func (r *OffsetRequest) SetReplicaID(id int32)

type OffsetResponse

type OffsetResponse struct {
	Version int16
	Blocks  map[string]map[int32]*OffsetResponseBlock
}

func (*OffsetResponse) AddTopicPartition

func (r *OffsetResponse) AddTopicPartition(topic string, partition int32, offset int64)

func (*OffsetResponse) GetBlock

func (r *OffsetResponse) GetBlock(topic string, partition int32) *OffsetResponseBlock

type OffsetResponseBlock

type OffsetResponseBlock struct {
	Err       KError
	Offsets   []int64 // Version 0
	Offset    int64   // Version 1
	Timestamp int64   // Version 1
}

type PacketDecodingError

type PacketDecodingError struct {
	Info string
}

PacketDecodingError is returned when there was an error (other than truncated data) decoding the Kafka broker's response. This can be a bad CRC or length field, or any other invalid value.

func (PacketDecodingError) Error

func (err PacketDecodingError) Error() string

type PacketEncodingError

type PacketEncodingError struct {
	Info string
}

PacketEncodingError is returned from a failure while encoding a Kafka packet. This can happen, for example, if you try to encode a string over 2^15 characters in length, since Kafka's encoding rules do not permit that.

func (PacketEncodingError) Error

func (err PacketEncodingError) Error() string

type PartitionConsumer

type PartitionConsumer interface {

	// AsyncClose initiates a shutdown of the PartitionConsumer. This method will return immediately, after which you
	// should continue to service the 'Messages' and 'Errors' channels until they are empty. It is required to call this
	// function, or Close before a consumer object passes out of scope, as it will otherwise leak memory. You must call
	// this before calling Close on the underlying client.
	AsyncClose()

	// Close stops the PartitionConsumer from fetching messages. It will initiate a shutdown just like AsyncClose, drain
	// the Messages channel, harvest any errors & return them to the caller. Note that if you are continuing to service
	// the Messages channel when this function is called, you will be competing with Close for messages; consider
	// calling AsyncClose, instead. It is required to call this function (or AsyncClose) before a consumer object passes
	// out of scope, as it will otherwise leak memory. You must call this before calling Close on the underlying client.
	Close() error

	// Messages returns the read channel for the messages that are returned by
	// the broker.
	Messages() <-chan *ConsumerMessage

	// Errors returns a read channel of errors that occurred during consuming, if
	// enabled. By default, errors are logged and not returned over this channel.
	// If you want to implement any custom error handling, set your config's
	// Consumer.Return.Errors setting to true, and read from this channel.
	Errors() <-chan *ConsumerError

	// HighWaterMarkOffset returns the high water mark offset of the partition,
	// i.e. the offset that will be used for the next message that will be produced.
	// You can use this to determine how far behind the processing is.
	HighWaterMarkOffset() int64
}

PartitionConsumer processes Kafka messages from a given topic and partition. You MUST call one of Close() or AsyncClose() on a PartitionConsumer to avoid leaks; it will not be garbage-collected automatically when it passes out of scope.

The simplest way of using a PartitionConsumer is to loop over its Messages channel using a for/range loop. The PartitionConsumer will only stop itself in one case: when the offset being consumed is reported as out of range by the brokers. In this case you should decide what you want to do (try a different offset, notify a human, etc) and handle it appropriately. For all other error cases, it will just keep retrying. By default, it logs these errors to sarama.Logger; if you want to be notified directly of all errors, set your config's Consumer.Return.Errors to true and read from the Errors channel, using a select statement or a separate goroutine. Check out the Consumer examples to see implementations of these different approaches.

To terminate such a for/range loop while the loop is executing, call AsyncClose. This will kick off the process of consumer tear-down & return imediately. Continue to loop, servicing the Messages channel until the teardown process AsyncClose initiated closes it (thus terminating the for/range loop). If you've already ceased reading Messages, call Close; this will signal the PartitionConsumer's goroutines to begin shutting down (just like AsyncClose), but will also drain the Messages channel, harvest all errors & return them once cleanup has completed.

type PartitionError added in v1.16.0

type PartitionError struct {
	Partition int32
	Err       KError
}

type PartitionMetadata

type PartitionMetadata struct {
	Err             KError
	ID              int32
	Leader          int32
	Replicas        []int32
	Isr             []int32
	OfflineReplicas []int32
}

type PartitionOffsetManager

type PartitionOffsetManager interface {
	// NextOffset returns the next offset that should be consumed for the managed
	// partition, accompanied by metadata which can be used to reconstruct the state
	// of the partition consumer when it resumes. NextOffset() will return
	// `config.Consumer.Offsets.Initial` and an empty metadata string if no offset
	// was committed for this partition yet.
	NextOffset() (int64, string)

	// MarkOffset marks the provided offset, alongside a metadata string
	// that represents the state of the partition consumer at that point in time. The
	// metadata string can be used by another consumer to restore that state, so it
	// can resume consumption.
	//
	// To follow upstream conventions, you are expected to mark the offset of the
	// next message to read, not the last message read. Thus, when calling `MarkOffset`
	// you should typically add one to the offset of the last consumed message.
	//
	// Note: calling MarkOffset does not necessarily commit the offset to the backend
	// store immediately for efficiency reasons, and it may never be committed if
	// your application crashes. This means that you may end up processing the same
	// message twice, and your processing should ideally be idempotent.
	MarkOffset(offset int64, metadata string)

	// ResetOffset resets to the provided offset, alongside a metadata string that
	// represents the state of the partition consumer at that point in time. Reset
	// acts as a counterpart to MarkOffset, the difference being that it allows to
	// reset an offset to an earlier or smaller value, where MarkOffset only
	// allows incrementing the offset. cf MarkOffset for more details.
	ResetOffset(offset int64, metadata string)

	// Errors returns a read channel of errors that occur during offset management, if
	// enabled. By default, errors are logged and not returned over this channel. If
	// you want to implement any custom error handling, set your config's
	// Consumer.Return.Errors setting to true, and read from this channel.
	Errors() <-chan *ConsumerError

	// AsyncClose initiates a shutdown of the PartitionOffsetManager. This method will
	// return immediately, after which you should wait until the 'errors' channel has
	// been drained and closed. It is required to call this function, or Close before
	// a consumer object passes out of scope, as it will otherwise leak memory. You
	// must call this before calling Close on the underlying client.
	AsyncClose()

	// Close stops the PartitionOffsetManager from managing offsets. It is required to
	// call this function (or AsyncClose) before a PartitionOffsetManager object
	// passes out of scope, as it will otherwise leak memory. You must call this
	// before calling Close on the underlying client.
	Close() error
}

PartitionOffsetManager uses Kafka to store and fetch consumed partition offsets. You MUST call Close() on a partition offset manager to avoid leaks, it will not be garbage-collected automatically when it passes out of scope.

type PartitionOffsetMetadata added in v1.16.0

type PartitionOffsetMetadata struct {
	Partition int32
	Offset    int64
	Metadata  *string
}

type Partitioner

type Partitioner interface {
	// Partition takes a message and partition count and chooses a partition
	Partition(message *ProducerMessage, numPartitions int32) (int32, error)

	// RequiresConsistency indicates to the user of the partitioner whether the
	// mapping of key->partition is consistent or not. Specifically, if a
	// partitioner requires consistency then it must be allowed to choose from all
	// partitions (even ones known to be unavailable), and its choice must be
	// respected by the caller. The obvious example is the HashPartitioner.
	RequiresConsistency() bool
}

Partitioner is anything that, given a Kafka message and a number of partitions indexed [0...numPartitions-1], decides to which partition to send the message. RandomPartitioner, RoundRobinPartitioner and HashPartitioner are provided as simple default implementations.

Example (Manual)

This example shows how to assign partitions to your messages manually.

config := NewConfig()

// First, we tell the producer that we are going to partition ourselves.
config.Producer.Partitioner = NewManualPartitioner

producer, err := NewSyncProducer([]string{"localhost:9092"}, config)
if err != nil {
	log.Fatal(err)
}
defer func() {
	if err := producer.Close(); err != nil {
		log.Println("Failed to close producer:", err)
	}
}()

// Now, we set the Partition field of the ProducerMessage struct.
msg := &ProducerMessage{Topic: "test", Partition: 6, Value: StringEncoder("test")}

partition, offset, err := producer.SendMessage(msg)
if err != nil {
	log.Fatalln("Failed to produce message to kafka cluster.")
}

if partition != 6 {
	log.Fatal("Message should have been produced to partition 6!")
}

log.Printf("Produced message to partition %d with offset %d", partition, offset)
Output:

Example (Per_topic)

This example shows how to set a different partitioner depending on the topic.

config := NewConfig()
config.Producer.Partitioner = func(topic string) Partitioner {
	switch topic {
	case "access_log", "error_log":
		return NewRandomPartitioner(topic)

	default:
		return NewHashPartitioner(topic)
	}
}

// ...
Output:

Example (Random)

By default, Sarama uses the message's key to consistently assign a partition to a message using hashing. If no key is set, a random partition will be chosen. This example shows how you can partition messages randomly, even when a key is set, by overriding Config.Producer.Partitioner.

config := NewConfig()
config.Producer.Partitioner = NewRandomPartitioner

producer, err := NewSyncProducer([]string{"localhost:9092"}, config)
if err != nil {
	log.Fatal(err)
}
defer func() {
	if err := producer.Close(); err != nil {
		log.Println("Failed to close producer:", err)
	}
}()

msg := &ProducerMessage{Topic: "test", Key: StringEncoder("key is set"), Value: StringEncoder("test")}
partition, offset, err := producer.SendMessage(msg)
if err != nil {
	log.Fatalln("Failed to produce message to kafka cluster.")
}

log.Printf("Produced message to partition %d with offset %d", partition, offset)
Output:

func NewHashPartitioner

func NewHashPartitioner(topic string) Partitioner

NewHashPartitioner returns a Partitioner which behaves as follows. If the message's key is nil then a random partition is chosen. Otherwise the FNV-1a hash of the encoded bytes of the message key is used, modulus the number of partitions. This ensures that messages with the same key always end up on the same partition.

func NewManualPartitioner

func NewManualPartitioner(topic string) Partitioner

NewManualPartitioner returns a Partitioner which uses the partition manually set in the provided ProducerMessage's Partition field as the partition to produce to.

func NewRandomPartitioner

func NewRandomPartitioner(topic string) Partitioner

NewRandomPartitioner returns a Partitioner which chooses a random partition each time.

func NewReferenceHashPartitioner added in v1.18.0

func NewReferenceHashPartitioner(topic string) Partitioner

NewReferenceHashPartitioner is like NewHashPartitioner except that it handles absolute values in the same way as the reference Java implementation. NewHashPartitioner was supposed to do that but it had a mistake and now there are people depending on both behaviours. This will all go away on the next major version bump.

func NewRoundRobinPartitioner

func NewRoundRobinPartitioner(topic string) Partitioner

NewRoundRobinPartitioner returns a Partitioner which walks through the available partitions one at a time.

type PartitionerConstructor

type PartitionerConstructor func(topic string) Partitioner

PartitionerConstructor is the type for a function capable of constructing new Partitioners.

func NewCustomHashPartitioner

func NewCustomHashPartitioner(hasher func() hash.Hash32) PartitionerConstructor

NewCustomHashPartitioner is a wrapper around NewHashPartitioner, allowing the use of custom hasher. The argument is a function providing the instance, implementing the hash.Hash32 interface. This is to ensure that each partition dispatcher gets its own hasher, to avoid concurrency issues by sharing an instance.

func NewCustomPartitioner added in v1.18.0

func NewCustomPartitioner(options ...HashPartitionerOption) PartitionerConstructor

NewCustomPartitioner creates a default Partitioner but lets you specify the behavior of each component via options

type ProduceRequest

type ProduceRequest struct {
	TransactionalID *string
	RequiredAcks    RequiredAcks
	Timeout         int32
	Version         int16 // v1 requires Kafka 0.9, v2 requires Kafka 0.10, v3 requires Kafka 0.11
	// contains filtered or unexported fields
}

func (*ProduceRequest) AddBatch added in v1.14.0

func (r *ProduceRequest) AddBatch(topic string, partition int32, batch *RecordBatch)

func (*ProduceRequest) AddMessage

func (r *ProduceRequest) AddMessage(topic string, partition int32, msg *Message)

func (*ProduceRequest) AddSet

func (r *ProduceRequest) AddSet(topic string, partition int32, set *MessageSet)

type ProduceResponse

type ProduceResponse struct {
	Blocks       map[string]map[int32]*ProduceResponseBlock
	Version      int16
	ThrottleTime time.Duration // only provided if Version >= 1
}

func (*ProduceResponse) AddTopicPartition

func (r *ProduceResponse) AddTopicPartition(topic string, partition int32, err KError)

func (*ProduceResponse) GetBlock

func (r *ProduceResponse) GetBlock(topic string, partition int32) *ProduceResponseBlock

type ProduceResponseBlock

type ProduceResponseBlock struct {
	Err    KError
	Offset int64
	// only provided if Version >= 2 and the broker is configured with `LogAppendTime`
	Timestamp time.Time
}

type ProducerError

type ProducerError struct {
	Msg *ProducerMessage
	Err error
}

ProducerError is the type of error generated when the producer fails to deliver a message. It contains the original ProducerMessage as well as the actual error value.

func (ProducerError) Error

func (pe ProducerError) Error() string

type ProducerErrors

type ProducerErrors []*ProducerError

ProducerErrors is a type that wraps a batch of "ProducerError"s and implements the Error interface. It can be returned from the Producer's Close method to avoid the need to manually drain the Errors channel when closing a producer.

func (ProducerErrors) Error

func (pe ProducerErrors) Error() string

type ProducerMessage

type ProducerMessage struct {
	Topic string // The Kafka topic for this message.
	// The partitioning key for this message. Pre-existing Encoders include
	// StringEncoder and ByteEncoder.
	Key Encoder
	// The actual message to store in Kafka. Pre-existing Encoders include
	// StringEncoder and ByteEncoder.
	Value Encoder

	// The headers are key-value pairs that are transparently passed
	// by Kafka between producers and consumers.
	Headers []RecordHeader

	// This field is used to hold arbitrary data you wish to include so it
	// will be available when receiving on the Successes and Errors channels.
	// Sarama completely ignores this field and is only to be used for
	// pass-through data.
	Metadata interface{}

	// Offset is the offset of the message stored on the broker. This is only
	// guaranteed to be defined if the message was successfully delivered and
	// RequiredAcks is not NoResponse.
	Offset int64
	// Partition is the partition that the message was sent to. This is only
	// guaranteed to be defined if the message was successfully delivered.
	Partition int32
	// Timestamp is the timestamp assigned to the message by the broker. This
	// is only guaranteed to be defined if the message was successfully
	// delivered, RequiredAcks is not NoResponse, and the Kafka broker is at
	// least version 0.10.0.
	Timestamp time.Time
	// contains filtered or unexported fields
}

ProducerMessage is the collection of elements passed to the Producer in order to send a message.

type Record added in v1.14.0

type Record struct {
	Attributes     int8
	TimestampDelta time.Duration
	OffsetDelta    int64
	Key            []byte
	Value          []byte
	Headers        []*RecordHeader
	// contains filtered or unexported fields
}

type RecordBatch added in v1.14.0

type RecordBatch struct {
	FirstOffset           int64
	PartitionLeaderEpoch  int32
	Version               int8
	Codec                 CompressionCodec
	CompressionLevel      int
	Control               bool
	LastOffsetDelta       int32
	FirstTimestamp        time.Time
	MaxTimestamp          time.Time
	ProducerID            int64
	ProducerEpoch         int16
	FirstSequence         int32
	Records               []*Record
	PartialTrailingRecord bool
	// contains filtered or unexported fields
}

type RecordHeader added in v1.14.0

type RecordHeader struct {
	Key   []byte
	Value []byte
}

type Records added in v1.14.0

type Records struct {
	MsgSet      *MessageSet
	RecordBatch *RecordBatch
	// contains filtered or unexported fields
}

Records implements a union type containing either a RecordBatch or a legacy MessageSet.

type RequestNotifierFunc

type RequestNotifierFunc func(bytesRead, bytesWritten int)

RequestNotifierFunc is invoked when a mock broker processes a request successfully and will provides the number of bytes read and written.

type RequestResponse

type RequestResponse struct {
	Request  protocolBody
	Response encoder
}

RequestResponse represents a Request/Response pair processed by MockBroker.

type RequiredAcks

type RequiredAcks int16

RequiredAcks is used in Produce Requests to tell the broker how many replica acknowledgements it must see before responding. Any of the constants defined here are valid. On broker versions prior to 0.8.2.0 any other positive int16 is also valid (the broker will wait for that many acknowledgements) but in 0.8.2.0 and later this will raise an exception (it has been replaced by setting the `min.isr` value in the brokers configuration).

const (
	// NoResponse doesn't send any response, the TCP ACK is all you get.
	NoResponse RequiredAcks = 0
	// WaitForLocal waits for only the local commit to succeed before responding.
	WaitForLocal RequiredAcks = 1
	// WaitForAll waits for all in-sync replicas to commit before responding.
	// The minimum number of in-sync replicas is configured on the broker via
	// the `min.insync.replicas` configuration key.
	WaitForAll RequiredAcks = -1
)

type Resource added in v1.16.0

type Resource struct {
	ResourceType AclResourceType
	ResourceName string
}

type ResourceAcls added in v1.16.0

type ResourceAcls struct {
	Resource
	Acls []*Acl
}

type ResourceResponse added in v1.16.0

type ResourceResponse struct {
	ErrorCode int16
	ErrorMsg  string
	Type      ConfigResourceType
	Name      string
	Configs   []*ConfigEntry
}

type SaslHandshakeRequest

type SaslHandshakeRequest struct {
	Mechanism string
}

type SaslHandshakeResponse

type SaslHandshakeResponse struct {
	Err               KError
	EnabledMechanisms []string
}

type StdLogger

type StdLogger interface {
	Print(v ...interface{})
	Printf(format string, v ...interface{})
	Println(v ...interface{})
}

StdLogger is used to log error messages.

var Logger StdLogger = log.New(ioutil.Discard, "[Sarama] ", log.LstdFlags)

Logger is the instance of a StdLogger interface that Sarama writes connection management events to. By default it is set to discard all log messages via ioutil.Discard, but you can set it to redirect wherever you want.

type StringEncoder

type StringEncoder string

StringEncoder implements the Encoder interface for Go strings so that they can be used as the Key or Value in a ProducerMessage.

func (StringEncoder) Encode

func (s StringEncoder) Encode() ([]byte, error)

func (StringEncoder) Length

func (s StringEncoder) Length() int

type SyncGroupRequest

type SyncGroupRequest struct {
	GroupId          string
	GenerationId     int32
	MemberId         string
	GroupAssignments map[string][]byte
}

func (*SyncGroupRequest) AddGroupAssignment

func (r *SyncGroupRequest) AddGroupAssignment(memberId string, memberAssignment []byte)

func (*SyncGroupRequest) AddGroupAssignmentMember

func (r *SyncGroupRequest) AddGroupAssignmentMember(memberId string, memberAssignment *ConsumerGroupMemberAssignment) error

type SyncGroupResponse

type SyncGroupResponse struct {
	Err              KError
	MemberAssignment []byte
}

func (*SyncGroupResponse) GetMemberAssignment

func (r *SyncGroupResponse) GetMemberAssignment() (*ConsumerGroupMemberAssignment, error)

type SyncProducer

type SyncProducer interface {

	// SendMessage produces a given message, and returns only when it either has
	// succeeded or failed to produce. It will return the partition and the offset
	// of the produced message, or an error if the message failed to produce.
	SendMessage(msg *ProducerMessage) (partition int32, offset int64, err error)

	// SendMessages produces a given set of messages, and returns only when all
	// messages in the set have either succeeded or failed. Note that messages
	// can succeed and fail individually; if some succeed and some fail,
	// SendMessages will return an error.
	SendMessages(msgs []*ProducerMessage) error

	// Close shuts down the producer and waits for any buffered messages to be
	// flushed. You must call this function before a producer object passes out of
	// scope, as it may otherwise leak memory. You must call this before calling
	// Close on the underlying client.
	Close() error
}

SyncProducer publishes Kafka messages, blocking until they have been acknowledged. It routes messages to the correct broker, refreshing metadata as appropriate, and parses responses for errors. You must call Close() on a producer to avoid leaks, it may not be garbage-collected automatically when it passes out of scope.

The SyncProducer comes with two caveats: it will generally be less efficient than the AsyncProducer, and the actual durability guarantee provided when a message is acknowledged depend on the configured value of `Producer.RequiredAcks`. There are configurations where a message acknowledged by the SyncProducer can still sometimes be lost.

For implementation reasons, the SyncProducer requires `Producer.Return.Errors` and `Producer.Return.Successes` to be set to true in its configuration.

Example

This example shows the basic usage pattern of the SyncProducer.

producer, err := NewSyncProducer([]string{"localhost:9092"}, nil)
if err != nil {
	log.Fatalln(err)
}
defer func() {
	if err := producer.Close(); err != nil {
		log.Fatalln(err)
	}
}()

msg := &ProducerMessage{Topic: "my_topic", Value: StringEncoder("testing 123")}
partition, offset, err := producer.SendMessage(msg)
if err != nil {
	log.Printf("FAILED to send message: %s\n", err)
} else {
	log.Printf("> message sent to partition %d at offset %d\n", partition, offset)
}
Output:

func NewSyncProducer

func NewSyncProducer(addrs []string, config *Config) (SyncProducer, error)

NewSyncProducer creates a new SyncProducer using the given broker addresses and configuration.

func NewSyncProducerFromClient

func NewSyncProducerFromClient(client Client) (SyncProducer, error)

NewSyncProducerFromClient creates a new SyncProducer using the given client. It is still necessary to call Close() on the underlying client when shutting down this producer.

type TestReporter

type TestReporter interface {
	Error(...interface{})
	Errorf(string, ...interface{})
	Fatal(...interface{})
	Fatalf(string, ...interface{})
}

TestReporter has methods matching go's testing.T to avoid importing `testing` in the main part of the library.

type Timestamp added in v1.14.0

type Timestamp struct {
	*time.Time
}

type TopicDetail added in v1.16.0

type TopicDetail struct {
	NumPartitions     int32
	ReplicationFactor int16
	ReplicaAssignment map[int32][]int32
	ConfigEntries     map[string]*string
}

type TopicError added in v1.16.0

type TopicError struct {
	Err    KError
	ErrMsg *string
}

type TopicMetadata

type TopicMetadata struct {
	Err        KError
	Name       string
	IsInternal bool // Only valid for Version >= 1
	Partitions []*PartitionMetadata
}

type TopicPartition added in v1.15.0

type TopicPartition struct {
	Count      int32
	Assignment [][]int32
}

type TopicPartitionError added in v1.15.0

type TopicPartitionError struct {
	Err    KError
	ErrMsg *string
}

type TxnOffsetCommitRequest added in v1.16.0

type TxnOffsetCommitRequest struct {
	TransactionalID string
	GroupID         string
	ProducerID      int64
	ProducerEpoch   int16
	Topics          map[string][]*PartitionOffsetMetadata
}

type TxnOffsetCommitResponse added in v1.16.0

type TxnOffsetCommitResponse struct {
	ThrottleTime time.Duration
	Topics       map[string][]*PartitionError
}

Source Files

Directories

Path Synopsis
examples
Package mocks provides mocks that can be used for testing applications that use Sarama.
Package mocks provides mocks that can be used for testing applications that use Sarama.
tools

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL