Prometheus remote write

Deliver metric data to a Prometheus remote write endpoint

status: beta delivery: at-least-once acknowledgements: yes egress: batch state: stateless

Warnings

High cardinality metric names and labels are discouraged by Prometheus as they can provide performance and reliability problems. You should consider alternative strategies to reduce the cardinality. Vector offers a tag_cardinality_limit transform as a way to protect against this.

Configuration

Example configurations

{
  "sinks": {
    "my_sink_id": {
      "type": "prometheus_remote_write",
      "inputs": [
        "my-source-or-transform-id"
      ],
      "endpoint": "https://localhost:8087/",
      "default_namespace": "service",
      "acknowledgements": null,
      "healthcheck": null
    }
  }
}
[sinks.my_sink_id]
type = "prometheus_remote_write"
inputs = [ "my-source-or-transform-id" ]
endpoint = "https://localhost:8087/"
default_namespace = "service"
---
sinks:
  my_sink_id:
    type: prometheus_remote_write
    inputs:
      - my-source-or-transform-id
    endpoint: https://localhost:8087/
    default_namespace: service
    acknowledgements: null
    healthcheck: null
{
  "sinks": {
    "my_sink_id": {
      "type": "prometheus_remote_write",
      "inputs": [
        "my-source-or-transform-id"
      ],
      "endpoint": "https://localhost:8087/",
      "auth": null,
      "aws": null,
      "default_namespace": "service",
      "buckets": [
        0.005
      ],
      "quantiles": [
        0.5
      ],
      "buffer": null,
      "acknowledgements": null,
      "batch": null,
      "healthcheck": null,
      "request": null,
      "tls": null,
      "proxy": null,
      "tenant_id": "my-domain"
    }
  }
}
[sinks.my_sink_id]
type = "prometheus_remote_write"
inputs = [ "my-source-or-transform-id" ]
endpoint = "https://localhost:8087/"
default_namespace = "service"
buckets = [ 0.005 ]
quantiles = [ 0.5 ]
tenant_id = "my-domain"
---
sinks:
  my_sink_id:
    type: prometheus_remote_write
    inputs:
      - my-source-or-transform-id
    endpoint: https://localhost:8087/
    auth: null
    aws: null
    default_namespace: service
    buckets:
      - 0.005
    quantiles:
      - 0.5
    buffer: null
    acknowledgements: null
    batch: null
    healthcheck: null
    request: null
    tls: null
    proxy: null
    tenant_id: my-domain

acknowledgements

common optional object
Controls how acknowledgements are handled by this sink. When enabled, all connected sources that support end-to-end acknowledgements will wait for the destination of this sink to acknowledge receipt of events before providing acknowledgement to the sending source. These settings override the global acknowledgement settings.

acknowledgements.enabled

common optional bool
Controls if all connected sources will wait for this sink to deliver the events before acknowledging receipt.
default: false

auth

optional object
Options for authentication.

auth.access_key_id

optional string literal
The AWS access key id. Used for AWS authentication when communicating with AWS services.
Examples
"AKIAIOSFODNN7EXAMPLE"

auth.assume_role

optional string literal
The ARN of an IAM role to assume at startup.
Examples
"arn:aws:iam::123456789098:role/my_role"
The timeout for loading credentials. Relevant when the default credentials chain is used or assume_role.
Examples
30
default: 5 (seconds)

auth.password

required string literal
The basic authentication password.
Examples
"${PROMETHEUS_PASSWORD}"
"password"

auth.profile

optional string literal
The AWS profile name. Used to select AWS credentials from a provided credentials file.
Examples
"develop"
default: default

auth.region

optional string literal
The AWS region to send STS requests to. If not set, this will default to the configured region for the service itself.
Examples
"us-west-2"

auth.secret_access_key

optional string literal
The AWS secret access key. Used for AWS authentication when communicating with AWS services.
Examples
"wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY"

auth.strategy

required string literal enum
The authentication strategy to use.
Enum options
OptionDescription
awsAuthentication strategy used for the AWS managed Prometheus service.
basicThe basic authentication strategy.
Examples
"aws"
"basic"

auth.user

required string literal
The basic authentication user name.
Examples
"${PROMETHEUS_USERNAME}"
"username"

aws

optional object
Options for the AWS connections.

aws.region

common optional string literal
The AWS region of the target service. This defaults to the region named in the endpoint parameter, or the value of the $AWS_REGION or $AWS_DEFAULT_REGION environment variables if that cannot be determined, or “us-east-1”.
Examples
"us-east-1"

batch

optional object
Configures the sink batching behavior.

batch.max_bytes

common optional uint
The maximum size of a batch that will be processed by a sink. This is based on the uncompressed size of the batched events, before they are serialized / compressed.

batch.max_events

common optional uint
The maximum size of a batch, in events, before it is flushed.

batch.timeout_secs

common optional float
The maximum age of a batch before it is flushed.
default: 1 (seconds)

buckets

optional [float]
Default buckets to use for aggregating distribution metrics into histograms.
Array float
Examples
[
  0.005,
  0.01
]
default: [0.005 0.01 0.025 0.05 0.1 0.25 0.5 1 2.5 5 10]

buffer

optional object

Configures the sink specific buffer behavior.

More information about the individual buffer types, and buffer behavior, can be found in the Buffering Model section.

buffer.max_events

common optional uint
The maximum number of events allowed in the buffer.
Relevant when: type = "memory"
default: 500 (events)

buffer.max_size

required uint
The maximum size of the buffer on the disk. Must be at least ~256 megabytes (268435488 bytes).
Relevant when: type = "disk"
Examples
268435488

buffer.type

common optional string literal enum
The type of buffer to use.
Enum options
OptionDescription
disk

Events are buffered on disk.

This is less performant, but more durable. Data that has been synchronized to disk will not be lost if Vector is restarted forcefully or crashes.

Data is synchronized to disk every 500ms.

memory

Events are buffered in memory.

This is more performant, but less durable. Data will be lost if Vector is restarted forcefully or crashes.

default: memory

buffer.when_full

optional string literal enum
The behavior when the buffer becomes full.
Enum options
OptionDescription
block

Waits for capacity in the buffer.

This will cause backpressure to propagate to upstream components, which can cause data to pile up on the edge.

drop_newest

Drops the event without waiting for capacity in the buffer.

The data is lost. This should only be used when performance is the highest priority.

default: block

default_namespace

common optional string literal
Used as a namespace for metrics that don’t have it. A namespace will be prefixed to a metric’s name. It should follow Prometheus naming conventions.
Examples
"service"

endpoint

required string literal
The endpoint URL to send data to.
Examples
"https://localhost:8087/"

healthcheck

common optional object
Health check options for the sink.

healthcheck.enabled

common optional bool
Enables/disables the healthcheck upon Vector boot.
default: true

inputs

required [string]

A list of upstream source or transform IDs. Wildcards (*) are supported.

See configuration for more info.

Array string literal
Examples
[
  "my-source-or-transform-id",
  "prefix-*"
]

proxy

optional object
Configures an HTTP(S) proxy for Vector to use. By default, the globally configured proxy is used.

proxy.enabled

optional bool
If false the proxy will be disabled.
default: true

proxy.http

optional string literal
The URL to proxy HTTP requests through.
Examples
"http://foo.bar:3128"

proxy.https

optional string literal
The URL to proxy HTTPS requests through.
Examples
"http://foo.bar:3128"

proxy.no_proxy

optional [string]

A list of hosts to avoid proxying. Allowed patterns here include:

PatternExample match
Domain namesexample.com matches requests to example.com
Wildcard domains.example.com matches requests to example.com and its subdomains
IP addresses127.0.0.1 matches requests to 127.0.0.1
CIDR blocks192.168.0.0./16 matches requests to any IP addresses in this range
Splat* matches all hosts

quantiles

optional [float]
Quantiles to use for aggregating distribution metrics into a summary.
Array float
Examples
[
  0.5,
  0.75,
  0.9,
  0.95,
  0.99
]
default: [0.5 0.75 0.9 0.95 0.99]

request

optional object
Configures the sink request behavior.
Configure the adaptive concurrency algorithms. These values have been tuned by optimizing simulated results. In general you should not need to adjust these.
The fraction of the current value to set the new concurrency limit when decreasing the limit. Valid values are greater than 0 and less than 1. Smaller values cause the algorithm to scale back rapidly when latency increases. Note that the new limit is rounded down after applying this ratio.
default: 0.9
The adaptive concurrency algorithm uses an exponentially weighted moving average (EWMA) of past RTT measurements as a reference to compare with the current RTT. This value controls how heavily new measurements are weighted compared to older ones. Valid values are greater than 0 and less than 1. Smaller values cause this reference to adjust more slowly, which may be useful if a service has unusually high response variability.
default: 0.7
When calculating the past RTT average, we also compute a secondary “deviation” value that indicates how variable those values are. We use that deviation when comparing the past RTT average to the current measurements, so we can ignore increases in RTT that are within an expected range. This factor is used to scale up the deviation to an appropriate range. Valid values are greater than or equal to 0, and we expect reasonable values to range from 1.0 to 3.0. Larger values cause the algorithm to ignore larger increases in the RTT.
default: 2

request.concurrency

common optional uint
The maximum number of in-flight requests allowed at any given time, or “adaptive” to allow Vector to automatically set the limit based on current network and service conditions.
The time window, in seconds, used for the rate_limit_num option.
default: 1 (seconds)

request.rate_limit_num

common optional uint
The maximum number of requests allowed within the rate_limit_duration_secs time window.
default: 5
The maximum number of retries to make for failed requests. The default, for all intents and purposes, represents an infinite number of retries.
default: 1.8446744073709552e+19
The amount of time to wait before attempting the first retry for a failed request. Once, the first retry has failed the fibonacci sequence will be used to select future backoffs.
default: 1 (seconds)
The maximum amount of time, in seconds, to wait between retries.
default: 10 (seconds)

request.timeout_secs

common optional uint
The maximum time a request can take before being aborted. It is highly recommended that you do not lower this value below the service’s internal timeout, as this could create orphaned requests, pile on retries, and result in duplicate data downstream.
default: 60 (seconds)

tenant_id

optional string template
If set, a header named X-Scope-OrgID will be added to outgoing requests with the text of this setting. This may be used by Cortex or other remote services to identify the tenant making the request.
Note: This parameter supports Vector's template syntax, which enables you to use dynamic per-event values.
Examples
"my-domain"

tls

optional object
Configures the TLS options for outgoing connections.

tls.alpn_protocols

optional [string]
Sets the list of supported ALPN protocols, which are used during negotiation with peer. Prioritized in the order they are defined.

tls.ca_file

optional string literal
Absolute path to an additional CA certificate file, in DER or PEM format (X.509), or an inline CA certificate in PEM format.
Examples
"/path/to/certificate_authority.crt"

tls.crt_file

common optional string literal
Absolute path to a certificate file used to identify this connection, in DER or PEM format (X.509) or PKCS#12, or an inline certificate in PEM format. If this is set and is not a PKCS#12 archive, key_file must also be set.
Examples
"/path/to/host_certificate.crt"

tls.key_file

common optional string literal
Absolute path to a private key file used to identify this connection, in DER or PEM format (PKCS#8), or an inline private key in PEM format. If this is set, crt_file must also be set.
Examples
"/path/to/host_certificate.key"

tls.key_pass

optional string literal
Pass phrase used to unlock the encrypted key file. This has no effect unless key_file is set.
Examples
"${KEY_PASS_ENV_VAR}"
"PassWord1"
If true (the default), Vector will validate the TLS certificate of the remote host. Specifically the issuer is checked but not CRLs (Certificate Revocation Lists).
default: true

tls.verify_hostname

optional bool
If true (the default), Vector will validate the configured remote host name against the remote host’s TLS certificate. Do NOT set this to false unless you understand the risks of not verifying the remote hostname.
default: true

Telemetry

Metrics

link

buffer_byte_size

gauge
The number of bytes current in the buffer.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
host optional
The hostname of the system Vector is running on.
pid optional
The process ID of the Vector instance.

buffer_discarded_events_total

counter
The number of events dropped by this non-blocking buffer.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
host optional
The hostname of the system Vector is running on.
pid optional
The process ID of the Vector instance.

buffer_events

gauge
The number of events currently in the buffer.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
host optional
The hostname of the system Vector is running on.
pid optional
The process ID of the Vector instance.

buffer_received_event_bytes_total

counter
The number of bytes received by this buffer.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
host optional
The hostname of the system Vector is running on.
pid optional
The process ID of the Vector instance.

buffer_received_events_total

counter
The number of events received by this buffer.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
host optional
The hostname of the system Vector is running on.
pid optional
The process ID of the Vector instance.

buffer_sent_event_bytes_total

counter
The number of bytes sent by this buffer.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
host optional
The hostname of the system Vector is running on.
pid optional
The process ID of the Vector instance.

buffer_sent_events_total

counter
The number of events sent by this buffer.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
host optional
The hostname of the system Vector is running on.
pid optional
The process ID of the Vector instance.

component_received_event_bytes_total

counter
The number of event bytes accepted by this component either from tagged origins like file and uri, or cumulatively from other origins.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
container_name optional
The name of the container from which the data originated.
file optional
The file from which the data originated.
host optional
The hostname of the system Vector is running on.
mode optional
The connection mode used by the component.
peer_addr optional
The IP from which the data originated.
peer_path optional
The pathname from which the data originated.
pid optional
The process ID of the Vector instance.
pod_name optional
The name of the pod from which the data originated.
uri optional
The sanitized URI from which the data originated.

component_received_events_count

histogram

A histogram of the number of events passed in each internal batch in Vector’s internal topology.

Note that this is separate than sink-level batching. It is mostly useful for low level debugging performance issues in Vector due to small internal batches.

component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
container_name optional
The name of the container from which the data originated.
file optional
The file from which the data originated.
host optional
The hostname of the system Vector is running on.
mode optional
The connection mode used by the component.
peer_addr optional
The IP from which the data originated.
peer_path optional
The pathname from which the data originated.
pid optional
The process ID of the Vector instance.
pod_name optional
The name of the pod from which the data originated.
uri optional
The sanitized URI from which the data originated.

component_received_events_total

counter
The number of events accepted by this component either from tagged origins like file and uri, or cumulatively from other origins.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
container_name optional
The name of the container from which the data originated.
file optional
The file from which the data originated.
host optional
The hostname of the system Vector is running on.
mode optional
The connection mode used by the component.
peer_addr optional
The IP from which the data originated.
peer_path optional
The pathname from which the data originated.
pid optional
The process ID of the Vector instance.
pod_name optional
The name of the pod from which the data originated.
uri optional
The sanitized URI from which the data originated.

component_sent_event_bytes_total

counter
The total number of event bytes emitted by this component.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
host optional
The hostname of the system Vector is running on.
output optional
The specific output of the component.
pid optional
The process ID of the Vector instance.

component_sent_events_total

counter
The total number of events emitted by this component.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
host optional
The hostname of the system Vector is running on.
output optional
The specific output of the component.
pid optional
The process ID of the Vector instance.

events_in_total

counter
The number of events accepted by this component either from tagged origins like file and uri, or cumulatively from other origins. This metric is deprecated and will be removed in a future version. Use component_received_events_total instead.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
container_name optional
The name of the container from which the data originated.
file optional
The file from which the data originated.
host optional
The hostname of the system Vector is running on.
mode optional
The connection mode used by the component.
peer_addr optional
The IP from which the data originated.
peer_path optional
The pathname from which the data originated.
pid optional
The process ID of the Vector instance.
pod_name optional
The name of the pod from which the data originated.
uri optional
The sanitized URI from which the data originated.

processing_errors_total

counter
The total number of processing errors encountered by this component. This metric is deprecated in favor of component_errors_total.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
error_type
The type of the error
host optional
The hostname of the system Vector is running on.
pid optional
The process ID of the Vector instance.

utilization

gauge
A ratio from 0 to 1 of the load on a component. A value of 0 would indicate a completely idle component that is simply waiting for input. A value of 1 would indicate a that is never idle. This value is updated every 5 seconds.
component_id
The Vector component ID.
component_kind
The Vector component kind.
component_name
Deprecated, use component_id instead. The value is the same as component_id.
component_type
The Vector component type.
host optional
The hostname of the system Vector is running on.
pid optional
The process ID of the Vector instance.

How it works

Buffers and batches

This component buffers & batches data as shown in the diagram above. You’ll notice that Vector treats these concepts differently, instead of treating them as global concepts, Vector treats them as sink specific concepts. This isolates sinks, ensuring services disruptions are contained and delivery guarantees are honored.

Batches are flushed when 1 of 2 conditions are met:

  1. The batch age meets or exceeds the configured timeout_secs.
  2. The batch size meets or exceeds the configured max_bytes or max_events.

Buffers are controlled via the buffer.* options.

Health checks

Health checks ensure that the downstream service is accessible and ready to accept data. This check is performed upon sink initialization. If the health check fails an error will be logged and Vector will proceed to start.

Require health checks

If you’d like to exit immediately upon a health check failure, you can pass the --require-healthy flag:

vector --config /etc/vector/vector.toml --require-healthy

Disable health checks

If you’d like to disable health checks for this sink you can set the healthcheck option to false.

Partitioning

Vector supports dynamic configuration values through a simple template syntax. If an option supports templating, it will be noted with a badge and you can use event fields to create dynamic values. For example:

[sinks.my-sink]
dynamic_option = "application={{ application_id }}"

In the above example, the application_id for each event will be used to partition outgoing data.

Rate limits & adaptive concurrency

Adaptive Request Concurrency (ARC)

Adaptive Request Concurrency is a feature of Vector that does away with static concurrency limits and automatically optimizes HTTP concurrency based on downstream service responses. The underlying mechanism is a feedback loop inspired by TCP congestion control algorithms. Checkout the announcement blog post,

We highly recommend enabling this feature as it improves performance and reliability of Vector and the systems it communicates with. As such, we have made it the default, and no further configuration is required.

Static concurrency

If Adaptive Request Concurrency is not for you, you can manually set static concurrency limits by specifying an integer for request.concurrency:

[sinks.my-sink]
  request.concurrency = 10

Rate limits

In addition to limiting request concurrency, you can also limit the overall request throughput via the request.rate_limit_duration_secs and request.rate_limit_num options.

[sinks.my-sink]
  request.rate_limit_duration_secs = 1
  request.rate_limit_num = 10

These will apply to both adaptive and fixed request.concurrency values.

Retry policy

Vector will retry failed requests (status == 429, >= 500, and != 501). Other responses will not be retried. You can control the number of retry attempts and backoff rate with the request.retry_attempts and request.retry_backoff_secs options.

State

This component is stateless, meaning its behavior is consistent across each input.

Transport Layer Security (TLS)

Vector uses OpenSSL for TLS protocols due to OpenSSL’s maturity. You can enable and adjust TLS behavior using the tls.* options.