kafka output

The kafka output type writes a batch of messages to Kafka brokers and waits for acknowledgement before propagating it back to the input.

Common
Advanced

# Common config fields, showing default values
output:
  label: ""
  kafka:
    addresses: [] # No default (required)
    topic: "" # No default (required)
    target_version: "" # No default (optional)
    key: ""
    partitioner: "fnv1a_hash"
    compression: "none"
    static_headers: {} # No default (optional)
    metadata:
      exclude_prefixes: []
    max_in_flight: 64
    batching:
      count: 0
      byte_size: 0
      period: ""
      check: ""

# All config fields, showing default values
output:
  label: ""
  kafka:
    addresses: [] # No default (required)
    tls:
      enabled: false
      skip_cert_verify: false
      enable_renegotiation: false
      root_cas: ""
      root_cas_file: ""
      client_certs: []
    sasl:
      mechanism: "none"
      user: ""
      password: ""
      access_token: ""
      token_cache: ""
      token_key: ""
    topic: "" # No default (required)
    client_id: "benthos"
    target_version: "" # No default (optional)
    rack_id: ""
    key: ""
    partitioner: "fnv1a_hash"
    partition: ""
    custom_topic_creation:
      enabled: false
      partitions: -1
      replication_factor: -1
    compression: "none"
    static_headers: {} # No default (optional)
    metadata:
      exclude_prefixes: []
    inject_tracing_map: "" # No default (optional)
    max_in_flight: 64
    idempotent_write: false
    ack_replicas: false
    max_msg_bytes: 1000000
    timeout: "5s"
    retry_as_batch: false
    batching:
      count: 0
      byte_size: 0
      period: ""
      check: ""
      processors: [] # No default (optional)
    max_retries: 0
    backoff:
      initial_interval: "3s"
      max_interval: "10s"
      max_elapsed_time: "30s"
    timestamp_ms: "" # No default (optional)

The config field ack_replicas determines whether we wait for acknowledgement from all replicas or just a single broker.

Both the key and topic fields can be dynamically set using function interpolations described in Bloblang queries.

Metadata will be added to each message sent as headers (version 0.11+), but can be restricted using the field metadata.

Strict ordering and retries

When strict ordering is required for messages written to topic partitions it is important to ensure that both the field max_in_flight is set to 1 and that the field retry_as_batch is set to true.

You must also ensure that failed batches are never rerouted back to the same output. This can be done by setting the field max_retries to 0 and backoff.max_elapsed_time to empty, which will apply back pressure indefinitely until the batch is sent successfully.

However, this also means that manual intervention will eventually be required in cases where the batch cannot be sent due to configuration problems such as an incorrect max_msg_bytes estimate. A less strict but automated alternative would be to route failed batches to a dead letter queue using a fallback broker, but this would allow subsequent batches to be delivered in the meantime whilst those failed batches are dealt with.

Troubleshooting

If you're seeing issues writing to or reading from Kafka with this component then it's worth trying out the newer kafka_franz output.

I'm seeing logs that report Failed to connect to kafka: kafka: client has run out of available brokers to talk to (Is your cluster reachable?), but the brokers are definitely reachable.

Unfortunately this error message will appear for a wide range of connection problems even when the broker endpoint can be reached. Double check your authentication configuration and also ensure that you have enabled TLS if applicable.

Performance

This output benefits from sending multiple messages in flight in parallel for improved performance. You can tune the max number of in flight messages (or message batches) with the field max_in_flight.

This output benefits from sending messages as a batch for improved performance. Batches can be formed at both the input and output level. You can find out more in this doc.

Fields

`addresses`

A list of broker addresses to connect to. If an item of the list contains commas it will be expanded into multiple addresses.

Type: array of string

`tls`

Custom TLS settings can be used to override system defaults.

Type: object

`tls.enabled`

Whether custom TLS settings are enabled.

Type: bool
Default: false

`tls.skip_cert_verify`

Whether to skip server side certificate verification.

Type: bool
Default: false

`tls.enable_renegotiation`

Whether to allow the remote server to repeatedly request renegotiation. Enable this option if you're seeing the error message local error: tls: no renegotiation.

Type: bool
Default: false

`tls.root_cas`

An optional root certificate authority to use. This is a string, representing a certificate chain from the parent trusted root certificate, to possible intermediate signing certificates, to the host certificate.

Secret

This field contains sensitive information. Use a secret reference rather than a literal value.

Type: string
Default: ""

`tls.root_cas_file`

An optional path of a root certificate authority file to use. This is a file, often with a .pem extension, containing a certificate chain from the parent trusted root certificate, to possible intermediate signing certificates, to the host certificate.

Type: string
Default: ""

`tls.client_certs`

A list of client certificates to use. For each certificate either the fields cert and key, or cert_file and key_file should be specified, but not both.

Type: array of object
Default: []

`tls.client_certs[].cert`

A plain text certificate to use.

Type: string
Default: ""

`tls.client_certs[].key`

A plain text certificate key to use.

Secret

This field contains sensitive information. Use a secret reference rather than a literal value.

Type: string
Default: ""

`tls.client_certs[].cert_file`

The path of a certificate to use.

Type: string
Default: ""

`tls.client_certs[].key_file`

The path of a certificate key to use.

Type: string
Default: ""

`tls.client_certs[].password`

A plain text password for when the private key is password encrypted in PKCS#1 or PKCS#8 format. The obsolete pbeWithMD5AndDES-CBC algorithm is not supported for the PKCS#8 format.

Because the obsolete pbeWithMD5AndDES-CBC algorithm does not authenticate the ciphertext, it is vulnerable to padding oracle attacks that can let an attacker recover the plaintext.

Secret

This field contains sensitive information. Use a secret reference rather than a literal value.

Type: string
Default: ""

`sasl`

Enables SASL authentication.

Type: object

`sasl.mechanism`

The SASL authentication mechanism, if left empty SASL authentication is not used.

Type: string
Default: "none"

Option	Summary
`OAUTHBEARER`	OAuth Bearer based authentication.
`PLAIN`	Plain text authentication. NOTE: When using plain text auth it is extremely likely that you'll also need to enable TLS.
`SCRAM-SHA-256`	Authentication using the SCRAM-SHA-256 mechanism.
`SCRAM-SHA-512`	Authentication using the SCRAM-SHA-512 mechanism.
`none`	Default, no SASL authentication.

`sasl.user`

A PLAIN username. It is recommended that you use environment variables to populate this field.

Type: string
Default: ""

`sasl.password`

A PLAIN password. It is recommended that you use environment variables to populate this field.

Secret

This field contains sensitive information. Use a secret reference rather than a literal value.

Type: string
Default: ""

`sasl.access_token`

A static OAUTHBEARER access token

Secret

This field contains sensitive information. Use a secret reference rather than a literal value.

Type: string
Default: ""

`sasl.token_cache`

Instead of using a static access_token allows you to query a cache resource to fetch OAUTHBEARER tokens from

Type: string
Default: ""

`sasl.token_key`

Required when using a token_cache, the key to query the cache with for tokens.

Type: string
Default: ""

`topic`

The topic to publish messages to.

This field supports interpolation functions.

Type: string

`client_id`

An identifier for the client connection.

Type: string
Default: "benthos"

`target_version`

The version of the Kafka protocol to use. This limits the capabilities used by the client and should ideally match the version of your brokers. Defaults to the oldest supported stable version.

Type: string

`rack_id`

A rack identifier for this client.

Type: string
Default: ""

`key`

The key to publish messages with.

This field supports interpolation functions.

Type: string
Default: ""

`partitioner`

The partitioning algorithm to use.

Type: string
Default: "fnv1a_hash"

Options: fnv1a_hash, murmur2_hash, random, round_robin, manual

`partition`

The manually-specified partition to publish messages to, relevant only when the field partitioner is set to manual. Must be able to parse as a 32-bit integer.

This field supports interpolation functions.

Type: string
Default: ""

`custom_topic_creation`

If enabled, topics will be created with the specified number of partitions and replication factor if they do not already exist.

Type: object

`custom_topic_creation.enabled`

Whether to enable custom topic creation.

Type: bool
Default: false

`custom_topic_creation.partitions`

The number of partitions to create for new topics. Leave at -1 to use the broker configured default. Must be >= 1.

Type: int
Default: -1

`custom_topic_creation.replication_factor`

The replication factor to use for new topics. Leave at -1 to use the broker configured default. Must be an odd number, and less then or equal to the number of brokers.

Type: int
Default: -1

`compression`

The compression algorithm to use.

Type: string
Default: "none"

Options: none, snappy, lz4, gzip, zstd

`static_headers`

An optional map of static headers that should be added to messages in addition to metadata.

Type: map of string

`metadata`

Specify criteria for which metadata values are sent with messages as headers.

Type: object

`metadata.exclude_prefixes`

Provide a list of explicit metadata key prefixes to be excluded when adding metadata to sent messages.

Type: array of string
Default: []

`inject_tracing_map`

EXPERIMENTAL: A Bloblang mapping used to inject an object containing tracing propagation information into outbound messages. The specification of the injected fields will match the format used by the service wide tracer.

Type: string

`max_in_flight`

The maximum number of messages to have in flight at a given time. Increase this to improve throughput.

Type: int
Default: 64

`idempotent_write`

Enable the idempotent write producer option. This requires the IDEMPOTENT_WRITE permission on CLUSTER and can be disabled if this permission is not available.

Type: bool
Default: false

`ack_replicas`

Ensure that messages have been copied across all replicas before acknowledging receipt.

Type: bool
Default: false

`max_msg_bytes`

The maximum size in bytes of messages sent to the target topic.

Type: int
Default: 1000000

`timeout`

The maximum period of time to wait for message sends before abandoning the request and retrying.

Type: string
Default: "5s"

`retry_as_batch`

When enabled forces an entire batch of messages to be retried if any individual message fails on a send, otherwise only the individual messages that failed are retried. Disabling this helps to reduce message duplicates during intermittent errors, but also makes it impossible to guarantee strict ordering of messages.

Type: bool
Default: false

`batching`

Allows you to configure a batching policy.

Type: object

`batching.count`

A number of messages at which the batch should be flushed. If 0 disables count based batching.

Type: int
Default: 0

`batching.byte_size`

An amount of bytes at which the batch should be flushed. If 0 disables size based batching.

Type: int
Default: 0

`batching.period`

A period in which an incomplete batch should be flushed regardless of its size.

Type: string
Default: ""

`batching.check`

A Bloblang query that should return a boolean value indicating whether a message should end a batch.

Type: string
Default: ""

`batching.processors`

A list of processors to apply to a batch as it is flushed. This allows you to aggregate and archive the batch however you see fit. Please note that all resulting messages are flushed as a single batch, therefore splitting the batch into smaller batches using these processors is a no-op.

Type: array of processor

`max_retries`

The maximum number of retries before giving up on the request. If set to zero there is no discrete limit.

Type: int
Default: 0

`backoff`

Control time intervals between retry attempts.

Type: object

`backoff.initial_interval`

The initial period to wait between retry attempts.

Type: string
Default: "3s"

`backoff.max_interval`

The maximum period to wait between retry attempts

Type: string
Default: "10s"

`backoff.max_elapsed_time`

The maximum overall period of time to spend on retry attempts before the request is aborted. Setting this value to a zeroed duration (such as 0s) will result in unbounded retries.

Type: string
Default: "30s"

`timestamp_ms`

An optional timestamp to set for each message expressed in milliseconds. When left empty, the current timestamp is used.

This field supports interpolation functions.

Type: string

Strict ordering and retries
Troubleshooting
Performance
Fields

Strict ordering and retries​

Troubleshooting​

Performance​

Fields​

addresses​

tls​

tls.enabled​

tls.skip_cert_verify​

tls.enable_renegotiation​

tls.root_cas​

tls.root_cas_file​

tls.client_certs​

tls.client_certs[].cert​

tls.client_certs[].key​

tls.client_certs[].cert_file​

tls.client_certs[].key_file​

tls.client_certs[].password​

sasl​

sasl.mechanism​

sasl.user​

sasl.password​

sasl.access_token​

sasl.token_cache​

sasl.token_key​

topic​

client_id​

target_version​

rack_id​

key​

partitioner​

partition​

custom_topic_creation​

custom_topic_creation.enabled​

custom_topic_creation.partitions​

custom_topic_creation.replication_factor​

compression​

static_headers​

metadata​

metadata.exclude_prefixes​

inject_tracing_map​

max_in_flight​

idempotent_write​

ack_replicas​

max_msg_bytes​

timeout​

retry_as_batch​

batching​

batching.count​

batching.byte_size​

batching.period​

batching.check​

batching.processors​

max_retries​

backoff​

backoff.initial_interval​

backoff.max_interval​

backoff.max_elapsed_time​

timestamp_ms​

Strict ordering and retries

Troubleshooting

Performance

Fields

`addresses`

`tls`

`tls.enabled`

`tls.skip_cert_verify`

`tls.enable_renegotiation`

`tls.root_cas`

`tls.root_cas_file`

`tls.client_certs`

`tls.client_certs[].cert`

`tls.client_certs[].key`

`tls.client_certs[].cert_file`

`tls.client_certs[].key_file`

`tls.client_certs[].password`

`sasl`

`sasl.mechanism`

`sasl.user`

`sasl.password`

`sasl.access_token`

`sasl.token_cache`

`sasl.token_key`

`topic`

`client_id`

`target_version`

`rack_id`

`key`

`partitioner`

`partition`

`custom_topic_creation`

`custom_topic_creation.enabled`

`custom_topic_creation.partitions`

`custom_topic_creation.replication_factor`

`compression`

`static_headers`

`metadata`

`metadata.exclude_prefixes`

`inject_tracing_map`

`max_in_flight`

`idempotent_write`

`ack_replicas`

`max_msg_bytes`

`timeout`

`retry_as_batch`

`batching`

`batching.count`

`batching.byte_size`

`batching.period`

`batching.check`

`batching.processors`

`max_retries`

`backoff`

`backoff.initial_interval`

`backoff.max_interval`

`backoff.max_elapsed_time`

`timestamp_ms`