Flyte Propeller Configuration#

Section: admin#

endpoint (config.URL)#

For admin types, specify where the uri of the service is located.

Default Value:

""

insecure (bool)#

Use insecure connection.

Default Value:

"false"

insecureSkipVerify (bool)#

InsecureSkipVerify controls whether a client verifies the server’s certificate chain and host name. Caution : shouldn’t be use for production usecases’

Default Value:

"false"

caCertFilePath (string)#

Use specified certificate file to verify the admin server peer.

Default Value:

""

maxBackoffDelay (config.Duration)#

Max delay for grpc backoff

Default Value:

8s

perRetryTimeout (config.Duration)#

gRPC per retry timeout

Default Value:

15s

maxRetries (int)#

Max number of gRPC retries

Default Value:

"4"

authType (uint8)#

Type of OAuth2 flow used for communicating with admin.ClientSecret,Pkce,ExternalCommand are valid values

Default Value:

ClientSecret

tokenRefreshWindow (config.Duration)#

Max duration between token refresh attempt and token expiry.

Default Value:

0s

useAuth (bool)#

Deprecated: Auth will be enabled/disabled based on admin’s dynamically discovered information.

Default Value:

"false"

clientId (string)#

Client ID

Default Value:

flytepropeller

clientSecretLocation (string)#

File containing the client secret

Default Value:

/etc/secrets/client_secret

clientSecretEnvVar (string)#

Environment variable containing the client secret

Default Value:

""

scopes ([]string)#

List of scopes to request

Default Value:

[]

useAudienceFromAdmin (bool)#

Use Audience configured from admins public endpoint config.

Default Value:

"false"

audience (string)#

Audience to use when initiating OAuth2 authorization requests.

Default Value:

""

authorizationServerUrl (string)#

This is the URL to your IdP’s authorization server. It’ll default to Endpoint

Default Value:

""

tokenUrl (string)#

OPTIONAL: Your IdP’s token endpoint. It’ll be discovered from flyte admin’s OAuth Metadata endpoint if not provided.

Default Value:

""

authorizationHeader (string)#

Custom metadata header to pass JWT

Default Value:

""

pkceConfig (pkce.Config)#

Config for Pkce authentication flow.

Default Value:

refreshTime: 5m0s
timeout: 2m0s

deviceFlowConfig (deviceflow.Config)#

Config for Device authentication flow.

Default Value:

pollInterval: 5s
refreshTime: 5m0s
timeout: 10m0s

command ([]string)#

Command for external authentication token generation

Default Value:

[]

proxyCommand ([]string)#

Command for external proxy-authorization token generation

Default Value:

[]

defaultServiceConfig (string)#

Default Value:

""

httpProxyURL (config.URL)#

OPTIONAL: HTTP Proxy to be used for OAuth requests.

Default Value:

""

config.Duration#

Duration (int64)#

Default Value:

8s

config.URL#

URL (url.URL)#

Default Value:

ForceQuery: false
Fragment: ""
Host: ""
OmitHost: false
Opaque: ""
Path: ""
RawFragment: ""
RawPath: ""
RawQuery: ""
Scheme: ""
User: null

url.URL#

Scheme (string)#

Default Value:

""
Opaque (string)#

Default Value:

""
User (url.Userinfo)#

Default Value:

null
Host (string)#

Default Value:

""
Path (string)#

Default Value:

""
RawPath (string)#

Default Value:

""
OmitHost (bool)#

Default Value:

"false"
ForceQuery (bool)#

Default Value:

"false"
RawQuery (string)#

Default Value:

""
Fragment (string)#

Default Value:

""
RawFragment (string)#

Default Value:

""

deviceflow.Config#

refreshTime (config.Duration)#

grace period from the token expiry after which it would refresh the token.

Default Value:

5m0s
timeout (config.Duration)#

amount of time the device flow should complete or else it will be cancelled.

Default Value:

10m0s
pollInterval (config.Duration)#

amount of time the device flow would poll the token endpoint if auth server doesn’t return a polling interval. Okta and google IDP do return an interval’

Default Value:

5s

pkce.Config#

timeout (config.Duration)#

Amount of time the browser session would be active for authentication from client app.

Default Value:

2m0s
refreshTime (config.Duration)#

grace period from the token expiry after which it would refresh the token.

Default Value:

5m0s

Section: catalog-cache#

type (string)#

Catalog Implementation to use

Default Value:

noop

endpoint (string)#

Endpoint for catalog service

Default Value:

""

insecure (bool)#

Use insecure grpc connection

Default Value:

"false"

max-cache-age (config.Duration)#

Cache entries past this age will incur cache miss. 0 means cache never expires

Default Value:

0s

use-admin-auth (bool)#

Use the same gRPC credentials option as the flyteadmin client

Default Value:

"false"

max-retries (int)#

The max number of retries for event recording.

Default Value:

"5"

base-scalar (int)#

The base/scalar backoff duration in milliseconds for event recording retries.

Default Value:

"100"

backoff-jitter (string)#

A string representation of a floating point number between 0 and 1 specifying the jitter factor for event recording retries.

Default Value:

"0.1"

default-service-config (string)#

Set the default service config for the catalog gRPC client

Default Value:

""

Section: event#

type (string)#

Sets the type of EventSink to configure [log/admin/file].

Default Value:

admin

file-path (string)#

For file types, specify where the file should be located.

Default Value:

""

rate (int64)#

Max rate at which events can be recorded per second.

Default Value:

"500"

capacity (int)#

The max bucket size for event recording tokens.

Default Value:

"1000"

max-retries (int)#

The max number of retries for event recording.

Default Value:

"5"

base-scalar (int)#

The base/scalar backoff duration in milliseconds for event recording retries.

Default Value:

"100"

backoff-jitter (string)#

A string representation of a floating point number between 0 and 1 specifying the jitter factor for event recording retries.

Default Value:

"0.1"

Section: logger#

show-source (bool)#

Includes source code location in logs.

Default Value:

"false"

mute (bool)#

Mutes all logs regardless of severity. Intended for benchmarks/tests only.

Default Value:

"false"

level (int)#

Sets the minimum logging level.

Default Value:

"3"

formatter (logger.FormatterConfig)#

Sets logging format.

Default Value:

type: json

logger.FormatterConfig#

type (string)#

Sets logging format type.

Default Value:

json

Section: otel#

type (string)#

Sets the type of exporter to configure [noop/file/jaeger/otlpgrpc/otlphttp].

Default Value:

noop

file (otelutils.FileConfig)#

Configuration for exporting telemetry traces to a file

Default Value:

filename: /tmp/trace.txt

jaeger (otelutils.JaegerConfig)#

Configuration for exporting telemetry traces to a jaeger

Default Value:

endpoint: http://localhost:14268/api/traces

otlpgrpc (otelutils.OtlpGrpcConfig)#

Configuration for exporting telemetry traces to an OTLP gRPC collector

Default Value:

endpoint: http://localhost:4317

otlphttp (otelutils.OtlpHttpConfig)#

Configuration for exporting telemetry traces to an OTLP HTTP collector

Default Value:

endpoint: http://localhost:4318/v1/traces

sampler (otelutils.SamplerConfig)#

Configuration for the sampler to use for the tracer

Default Value:

parentSampler: always
traceIdRatio: 0.01

otelutils.FileConfig#

filename (string)#

Filename to store exported telemetry traces

Default Value:

/tmp/trace.txt

otelutils.JaegerConfig#

endpoint (string)#

Endpoint for the jaeger telemetry trace ingestor

Default Value:

http://localhost:14268/api/traces

otelutils.OtlpGrpcConfig#

endpoint (string)#

Endpoint for the OTLP telemetry trace collector

Default Value:

http://localhost:4317

otelutils.OtlpHttpConfig#

endpoint (string)#

Endpoint for the OTLP telemetry trace collector

Default Value:

http://localhost:4318/v1/traces

otelutils.SamplerConfig#

parentSampler (string)#

Sets the parent sampler to use for the tracer

Default Value:

always
traceIdRatio (float64)#

Default Value:

"0.01"

Section: plugins#

agent-service (agent.Config)#

Default Value:

agentForTaskTypes: null
agents: null
defaultAgent:
  defaultServiceConfig: ""
  defaultTimeout: 10s
  endpoint: ""
  insecure: true
  timeouts: null
pollInterval: 10s
resourceConstraints:
  NamespaceScopeResourceConstraint:
    Value: 50
  ProjectScopeResourceConstraint:
    Value: 100
supportedTaskTypes:
- task_type_1
- task_type_2
webApi:
  caching:
    maxSystemFailures: 5
    resyncInterval: 30s
    size: 500000
    workers: 10
  readRateLimiter:
    burst: 100
    qps: 10
  resourceMeta: null
  resourceQuotas:
    default: 1000
  writeRateLimiter:
    burst: 100
    qps: 10

athena (athena.Config)#

Default Value:

defaultCatalog: AwsDataCatalog
defaultWorkGroup: primary
resourceConstraints:
  NamespaceScopeResourceConstraint:
    Value: 50
  ProjectScopeResourceConstraint:
    Value: 100
webApi:
  caching:
    maxSystemFailures: 5
    resyncInterval: 30s
    size: 500000
    workers: 10
  readRateLimiter:
    burst: 100
    qps: 10
  resourceMeta: null
  resourceQuotas:
    default: 1000
  writeRateLimiter:
    burst: 100
    qps: 10

aws (aws.Config)#

Default Value:

accountId: ""
logLevel: 0
region: us-east-2
retries: 3

bigquery (bigquery.Config)#

Default Value:

googleTokenSource:
  gke-task-workload-identity:
    remoteClusterConfig:
      auth:
        caCertPath: ""
        tokenPath: ""
      enabled: false
      endpoint: ""
      name: ""
  type: default
resourceConstraints:
  NamespaceScopeResourceConstraint:
    Value: 50
  ProjectScopeResourceConstraint:
    Value: 100
webApi:
  caching:
    maxSystemFailures: 5
    resyncInterval: 30s
    size: 500000
    workers: 10
  readRateLimiter:
    burst: 100
    qps: 10
  resourceMeta: null
  resourceQuotas:
    default: 1000
  writeRateLimiter:
    burst: 100
    qps: 10

catalogcache (catalog.Config)#

Default Value:

reader:
  maxItems: 10000
  maxRetries: 3
  workers: 10
writer:
  maxItems: 10000
  maxRetries: 3
  workers: 10

databricks (databricks.Config)#

Default Value:

databricksInstance: ""
databricksTokenKey: FLYTE_DATABRICKS_API_TOKEN
defaultWarehouse: COMPUTE_CLUSTER
entrypointFile: ""
resourceConstraints:
  NamespaceScopeResourceConstraint:
    Value: 50
  ProjectScopeResourceConstraint:
    Value: 100
webApi:
  caching:
    maxSystemFailures: 5
    resyncInterval: 30s
    size: 500000
    workers: 10
  readRateLimiter:
    burst: 100
    qps: 10
  resourceMeta: null
  resourceQuotas:
    default: 1000
  writeRateLimiter:
    burst: 100
    qps: 10

echo (testing.Config)#

Default Value:

sleep-duration: 0s

k8s (config.K8sPluginConfig)#

Default Value:

co-pilot:
  cpu: 500m
  default-input-path: /var/flyte/inputs
  default-output-path: /var/flyte/outputs
  image: cr.flyte.org/flyteorg/flytecopilot:v0.0.15
  input-vol-name: flyte-inputs
  memory: 128Mi
  name: flyte-copilot-
  output-vol-name: flyte-outputs
  start-timeout: 1m40s
  storage: ""
create-container-config-error-grace-period: 0s
create-container-error-grace-period: 3m0s
default-annotations:
  cluster-autoscaler.kubernetes.io/safe-to-evict: "false"
default-cpus: "1"
default-env-from-configmaps: null
default-env-from-secrets: null
default-env-vars: null
default-env-vars-from-env: null
default-labels: null
default-memory: 1Gi
default-node-selector: null
default-pod-dns-config: null
default-pod-security-context: null
default-pod-template-name: ""
default-pod-template-resync: 30s
default-security-context: null
default-tolerations: null
delete-resource-on-finalize: false
enable-host-networking-pod: null
gpu-device-node-label: k8s.amazonaws.com/accelerator
gpu-partition-size-node-label: k8s.amazonaws.com/gpu-partition-size
gpu-resource-name: nvidia.com/gpu
gpu-unpartitioned-node-selector-requirement: null
gpu-unpartitioned-toleration: null
image-pull-backoff-grace-period: 3m0s
image-pull-policy: ""
inject-finalizer: false
interruptible-node-selector: null
interruptible-node-selector-requirement: null
interruptible-tolerations: null
non-interruptible-node-selector-requirement: null
pod-pending-timeout: 0s
resource-tolerations: null
scheduler-name: ""
send-object-events: false
update-backoff-retries: 5
update-base-backoff-duration: 10

k8s-array (k8s.Config)#

Default Value:

ErrorAssembler:
  maxItems: 100000
  maxRetries: 5
  workers: 10
OutputAssembler:
  maxItems: 100000
  maxRetries: 5
  workers: 10
logs:
  config:
    cloudwatch-enabled: false
    cloudwatch-log-group: ""
    cloudwatch-region: ""
    cloudwatch-template-uri: ""
    dynamic-log-links: null
    gcp-project: ""
    kubernetes-enabled: true
    kubernetes-template-uri: http://localhost:30082/#!/log/{{ .namespace }}/{{ .podName
      }}/pod?namespace={{ .namespace }}
    kubernetes-url: ""
    stackdriver-enabled: false
    stackdriver-logresourcename: ""
    stackdriver-template-uri: ""
    templates: null
maxArrayJobSize: 5000
maxErrorLength: 1000
namespaceTemplate: ""
node-selector: null
remoteClusterConfig:
  auth:
    certPath: ""
    tokenPath: ""
    type: ""
  enabled: false
  endpoint: ""
  name: ""
resourceConfig:
  limit: 0
  primaryLabel: ""
scheduler: ""
tolerations: null

kf-operator (common.Config)#

Default Value:

timeout: 1m0s

logs (logs.LogConfig)#

Default Value:

cloudwatch-enabled: false
cloudwatch-log-group: ""
cloudwatch-region: ""
cloudwatch-template-uri: ""
dynamic-log-links: null
gcp-project: ""
kubernetes-enabled: true
kubernetes-template-uri: http://localhost:30082/#!/log/{{ .namespace }}/{{ .podName
  }}/pod?namespace={{ .namespace }}
kubernetes-url: ""
stackdriver-enabled: false
stackdriver-logresourcename: ""
stackdriver-template-uri: ""
templates: null

qubole (config.Config)#

Default Value:

analyzeLinkPath: /v2/analyze
clusterConfigs:
- labels:
  - default
  limit: 100
  namespaceScopeQuotaProportionCap: 0.7
  primaryLabel: default
  projectScopeQuotaProportionCap: 0.7
commandApiPath: /api/v1.2/commands/
defaultClusterLabel: default
destinationClusterConfigs: []
endpoint: https://wellness.qubole.com
lruCacheSize: 2000
quboleTokenKey: FLYTE_QUBOLE_CLIENT_TOKEN
workers: 15

ray (ray.Config)#

Default Value:

dashboardHost: 0.0.0.0
dashboardURLTemplate: null
defaults:
  headNode:
    ipAddress: $MY_POD_IP
    startParameters:
      disable-usage-stats: "true"
  workerNode:
    ipAddress: $MY_POD_IP
    startParameters:
      disable-usage-stats: "true"
enableUsageStats: false
includeDashboard: true
logs:
  cloudwatch-enabled: false
  cloudwatch-log-group: ""
  cloudwatch-region: ""
  cloudwatch-template-uri: ""
  dynamic-log-links: null
  gcp-project: ""
  kubernetes-enabled: false
  kubernetes-template-uri: ""
  kubernetes-url: ""
  stackdriver-enabled: false
  stackdriver-logresourcename: ""
  stackdriver-template-uri: ""
  templates: null
logsSidecar: null
remoteClusterConfig:
  auth:
    caCertPath: ""
    tokenPath: ""
  enabled: false
  endpoint: ""
  name: ""
serviceAccount: default
serviceType: NodePort
shutdownAfterJobFinishes: true
ttlSecondsAfterFinished: 3600

snowflake (snowflake.Config)#

Default Value:

defaultWarehouse: COMPUTE_WH
resourceConstraints:
  NamespaceScopeResourceConstraint:
    Value: 50
  ProjectScopeResourceConstraint:
    Value: 100
snowflakeTokenKey: FLYTE_SNOWFLAKE_CLIENT_TOKEN
webApi:
  caching:
    maxSystemFailures: 5
    resyncInterval: 30s
    size: 500000
    workers: 10
  readRateLimiter:
    burst: 100
    qps: 10
  resourceMeta: null
  resourceQuotas:
    default: 1000
  writeRateLimiter:
    burst: 100
    qps: 10

spark (spark.Config)#

Default Value:

features: null
logs:
  all-user:
    cloudwatch-enabled: false
    cloudwatch-log-group: ""
    cloudwatch-region: ""
    cloudwatch-template-uri: ""
    dynamic-log-links: null
    gcp-project: ""
    kubernetes-enabled: false
    kubernetes-template-uri: ""
    kubernetes-url: ""
    stackdriver-enabled: false
    stackdriver-logresourcename: ""
    stackdriver-template-uri: ""
    templates: null
  mixed:
    cloudwatch-enabled: false
    cloudwatch-log-group: ""
    cloudwatch-region: ""
    cloudwatch-template-uri: ""
    dynamic-log-links: null
    gcp-project: ""
    kubernetes-enabled: true
    kubernetes-template-uri: http://localhost:30082/#!/log/{{ .namespace }}/{{ .podName
      }}/pod?namespace={{ .namespace }}
    kubernetes-url: ""
    stackdriver-enabled: false
    stackdriver-logresourcename: ""
    stackdriver-template-uri: ""
    templates: null
  system:
    cloudwatch-enabled: false
    cloudwatch-log-group: ""
    cloudwatch-region: ""
    cloudwatch-template-uri: ""
    dynamic-log-links: null
    gcp-project: ""
    kubernetes-enabled: false
    kubernetes-template-uri: ""
    kubernetes-url: ""
    stackdriver-enabled: false
    stackdriver-logresourcename: ""
    stackdriver-template-uri: ""
    templates: null
  user:
    cloudwatch-enabled: false
    cloudwatch-log-group: ""
    cloudwatch-region: ""
    cloudwatch-template-uri: ""
    dynamic-log-links: null
    gcp-project: ""
    kubernetes-enabled: false
    kubernetes-template-uri: ""
    kubernetes-url: ""
    stackdriver-enabled: false
    stackdriver-logresourcename: ""
    stackdriver-template-uri: ""
    templates: null
spark-config-default: null
spark-history-server-url: ""

agent.Config#

webApi (webapi.PluginConfig)#

Defines config for the base WebAPI plugin.

Default Value:

caching:
  maxSystemFailures: 5
  resyncInterval: 30s
  size: 500000
  workers: 10
readRateLimiter:
  burst: 100
  qps: 10
resourceMeta: null
resourceQuotas:
  default: 1000
writeRateLimiter:
  burst: 100
  qps: 10
resourceConstraints (core.ResourceConstraintsSpec)#

Default Value:

NamespaceScopeResourceConstraint:
  Value: 50
ProjectScopeResourceConstraint:
  Value: 100
defaultAgent (agent.Deployment)#

The default agent.

Default Value:

defaultServiceConfig: ""
defaultTimeout: 10s
endpoint: ""
insecure: true
timeouts: null
agents (map[string]*agent.Deployment)#

The agents.

Default Value:

null
agentForTaskTypes (map[string]string)#

Default Value:

null
supportedTaskTypes ([]string)#

Default Value:

- task_type_1
- task_type_2
pollInterval (config.Duration)#

The interval at which the plugin should poll the agent for metadata updates.

Default Value:

10s

agent.Deployment#

endpoint (string)#

Default Value:

""
insecure (bool)#

Default Value:

"true"
defaultServiceConfig (string)#

Default Value:

""
timeouts (map[string]config.Duration)#

Default Value:

null
defaultTimeout (config.Duration)#

Default Value:

10s

core.ResourceConstraintsSpec#

ProjectScopeResourceConstraint (core.ResourceConstraint)#

Default Value:

Value: 100
NamespaceScopeResourceConstraint (core.ResourceConstraint)#

Default Value:

Value: 50

core.ResourceConstraint#

Value (int64)#

Default Value:

"100"

webapi.PluginConfig#

resourceQuotas (webapi.ResourceQuotas)#

Default Value:

default: 1000
readRateLimiter (webapi.RateLimiterConfig)#

Defines rate limiter properties for read actions (e.g. retrieve status).

Default Value:

burst: 100
qps: 10
writeRateLimiter (webapi.RateLimiterConfig)#

Defines rate limiter properties for write actions.

Default Value:

burst: 100
qps: 10
caching (webapi.CachingConfig)#

Defines caching characteristics.

Default Value:

maxSystemFailures: 5
resyncInterval: 30s
size: 500000
workers: 10
resourceMeta (interface)#

Default Value:

<nil>

webapi.CachingConfig#

size (int)#

Defines the maximum number of items to cache.

Default Value:

"500000"
resyncInterval (config.Duration)#

Defines the sync interval.

Default Value:

30s
workers (int)#

Defines the number of workers to start up to process items.

Default Value:

"10"
maxSystemFailures (int)#

Defines the number of failures to fetch a task before failing the task.

Default Value:

"5"

webapi.RateLimiterConfig#

qps (int)#

Defines the max rate of calls per second.

Default Value:

"10"
burst (int)#

Defines the maximum burst size.

Default Value:

"100"

athena.Config#

webApi (webapi.PluginConfig)#

Defines config for the base WebAPI plugin.

Default Value:

caching:
  maxSystemFailures: 5
  resyncInterval: 30s
  size: 500000
  workers: 10
readRateLimiter:
  burst: 100
  qps: 10
resourceMeta: null
resourceQuotas:
  default: 1000
writeRateLimiter:
  burst: 100
  qps: 10
resourceConstraints (core.ResourceConstraintsSpec)#

Default Value:

NamespaceScopeResourceConstraint:
  Value: 50
ProjectScopeResourceConstraint:
  Value: 100
defaultWorkGroup (string)#

Defines the default workgroup to use when running on Athena unless overwritten by the task.

Default Value:

primary
defaultCatalog (string)#

Defines the default catalog to use when running on Athena unless overwritten by the task.

Default Value:

AwsDataCatalog

aws.Config#

region (string)#

AWS Region to connect to.

Default Value:

us-east-2
accountId (string)#

AWS Account Identifier.

Default Value:

""
retries (int)#

Number of retries.

Default Value:

"3"
logLevel (uint64)#

Default Value:

"0"

bigquery.Config#

webApi (webapi.PluginConfig)#

Defines config for the base WebAPI plugin.

Default Value:

caching:
  maxSystemFailures: 5
  resyncInterval: 30s
  size: 500000
  workers: 10
readRateLimiter:
  burst: 100
  qps: 10
resourceMeta: null
resourceQuotas:
  default: 1000
writeRateLimiter:
  burst: 100
  qps: 10
resourceConstraints (core.ResourceConstraintsSpec)#

Default Value:

NamespaceScopeResourceConstraint:
  Value: 50
ProjectScopeResourceConstraint:
  Value: 100
googleTokenSource (google.TokenSourceFactoryConfig)#

Defines Google token source

Default Value:

gke-task-workload-identity:
  remoteClusterConfig:
    auth:
      caCertPath: ""
      tokenPath: ""
    enabled: false
    endpoint: ""
    name: ""
type: default
bigQueryEndpoint (string)#

Default Value:

""

google.TokenSourceFactoryConfig#

type (string)#

Defines type of TokenSourceFactory, possible values are ‘default’ and ‘gke-task-workload-identity’

Default Value:

default
gke-task-workload-identity (google.GkeTaskWorkloadIdentityTokenSourceFactoryConfig)#

Extra configuration for GKE task workload identity token source factory

Default Value:

remoteClusterConfig:
  auth:
    caCertPath: ""
    tokenPath: ""
  enabled: false
  endpoint: ""
  name: ""

google.GkeTaskWorkloadIdentityTokenSourceFactoryConfig#

remoteClusterConfig (k8s.ClusterConfig)#

Configuration of remote GKE cluster

Default Value:

auth:
  caCertPath: ""
  tokenPath: ""
enabled: false
endpoint: ""
name: ""

k8s.ClusterConfig#

name (string)#

Friendly name of the remote cluster

Default Value:

""
endpoint (string)#

Remote K8s cluster endpoint

Default Value:

""
auth (k8s.Auth)#

Default Value:

caCertPath: ""
tokenPath: ""
enabled (bool)#

Boolean flag to enable or disable

Default Value:

"false"

k8s.Auth#

tokenPath (string)#

Token path

Default Value:

""
caCertPath (string)#

Certificate path

Default Value:

""

catalog.Config#

reader (workqueue.Config)#

Catalog reader workqueue config. Make sure the index cache must be big enough to accommodate the biggest array task allowed to run on the system.

Default Value:

maxItems: 10000
maxRetries: 3
workers: 10
writer (workqueue.Config)#

Catalog writer workqueue config. Make sure the index cache must be big enough to accommodate the biggest array task allowed to run on the system.

Default Value:

maxItems: 10000
maxRetries: 3
workers: 10

workqueue.Config#

workers (int)#

Number of concurrent workers to start processing the queue.

Default Value:

"10"
maxRetries (int)#

Maximum number of retries per item.

Default Value:

"3"
maxItems (int)#

Maximum number of entries to keep in the index.

Default Value:

"10000"

common.Config#

timeout (config.Duration)#

Default Value:

1m0s

config.Config#

endpoint (config.URL)#

Endpoint for qubole to use

Default Value:

https://wellness.qubole.com
commandApiPath (config.URL)#

API Path where commands can be launched on Qubole. Should be a valid url.

Default Value:

/api/v1.2/commands/
analyzeLinkPath (config.URL)#

URL path where queries can be visualized on qubole website. Should be a valid url.

Default Value:

/v2/analyze
quboleTokenKey (string)#

Name of the key where to find Qubole token in the secret manager.

Default Value:

FLYTE_QUBOLE_CLIENT_TOKEN
lruCacheSize (int)#

Size of the AutoRefreshCache

Default Value:

"2000"
workers (int)#

Number of parallel workers to refresh the cache

Default Value:

"15"
defaultClusterLabel (string)#

The default cluster label. This will be used if label is not specified on the hive job.

Default Value:

default
clusterConfigs ([]config.ClusterConfig)#

Default Value:

- labels:
  - default
  limit: 100
  namespaceScopeQuotaProportionCap: 0.7
  primaryLabel: default
  projectScopeQuotaProportionCap: 0.7
destinationClusterConfigs ([]config.DestinationClusterConfig)#

Default Value:

[]

config.K8sPluginConfig#

inject-finalizer (bool)#

Instructs the plugin to inject a finalizer on startTask and remove it on task termination.

Default Value:

"false"
default-annotations (map[string]string)#

Default Value:

cluster-autoscaler.kubernetes.io/safe-to-evict: "false"
default-labels (map[string]string)#

Default Value:

null
default-env-vars (map[string]string)#

Default Value:

null
default-env-vars-from-env (map[string]string)#

Default Value:

null
default-env-from-configmaps ([]string)#

Default Value:

null
default-env-from-secrets ([]string)#

Default Value:

null
default-cpus (resource.Quantity)#

Defines a default value for cpu for containers if not specified.

Default Value:

"1"
default-memory (resource.Quantity)#

Defines a default value for memory for containers if not specified.

Default Value:

1Gi
default-tolerations ([]v1.Toleration)#

Default Value:

null
default-node-selector (map[string]string)#

Default Value:

null
default-affinity (v1.Affinity)#

Default Value:

null
scheduler-name (string)#

Defines scheduler name.

Default Value:

""
interruptible-tolerations ([]v1.Toleration)#

Default Value:

null
interruptible-node-selector (map[string]string)#

Default Value:

null
interruptible-node-selector-requirement (v1.NodeSelectorRequirement)#

Default Value:

null
non-interruptible-node-selector-requirement (v1.NodeSelectorRequirement)#

Default Value:

null
resource-tolerations (map[v1.ResourceName][]v1.Toleration)#

Default Value:

null
co-pilot (config.FlyteCoPilotConfig)#

Co-Pilot Configuration

Default Value:

cpu: 500m
default-input-path: /var/flyte/inputs
default-output-path: /var/flyte/outputs
image: cr.flyte.org/flyteorg/flytecopilot:v0.0.15
input-vol-name: flyte-inputs
memory: 128Mi
name: flyte-copilot-
output-vol-name: flyte-outputs
start-timeout: 1m40s
storage: ""
delete-resource-on-finalize (bool)#

Instructs the system to delete the resource upon successful execution of a k8s pod rather than have the k8s garbage collector clean it up. This ensures that no resources are kept around (potentially consuming cluster resources). This, however, will cause k8s log links to expire as soon as the resource is finalized.

Default Value:

"false"
create-container-error-grace-period (config.Duration)#

Default Value:

3m0s
create-container-config-error-grace-period (config.Duration)#

Default Value:

0s
image-pull-backoff-grace-period (config.Duration)#

Default Value:

3m0s
image-pull-policy (string)#

Default Value:

""
pod-pending-timeout (config.Duration)#

Default Value:

0s
gpu-device-node-label (string)#

Default Value:

k8s.amazonaws.com/accelerator
gpu-partition-size-node-label (string)#

Default Value:

k8s.amazonaws.com/gpu-partition-size
gpu-unpartitioned-node-selector-requirement (v1.NodeSelectorRequirement)#

Default Value:

null
gpu-unpartitioned-toleration (v1.Toleration)#

Default Value:

null
gpu-resource-name (string)#

Default Value:

nvidia.com/gpu
default-pod-security-context (v1.PodSecurityContext)#

Default Value:

null
default-security-context (v1.SecurityContext)#

Default Value:

null
enable-host-networking-pod (bool)#

Default Value:

<invalid reflect.Value>
default-pod-dns-config (v1.PodDNSConfig)#

Default Value:

null
default-pod-template-name (string)#

Name of the PodTemplate to use as the base for all k8s pods created by FlytePropeller.

Default Value:

""
default-pod-template-resync (config.Duration)#

Frequency of resyncing default pod templates

Default Value:

30s
send-object-events (bool)#

If true, will send k8s object events in TaskExecutionEvent updates.

Default Value:

"false"
update-base-backoff-duration (int)#

Initial delay in exponential backoff when updating a resource in milliseconds.

Default Value:

"10"
update-backoff-retries (int)#

Number of retries for exponential backoff when updating a resource.

Default Value:

"5"

config.FlyteCoPilotConfig#

name (string)#

Flyte co-pilot sidecar container name prefix. (additional bits will be added after this)

Default Value:

flyte-copilot-
image (string)#

Flyte co-pilot Docker Image FQN

Default Value:

cr.flyte.org/flyteorg/flytecopilot:v0.0.15
default-input-path (string)#

Default path where the volume should be mounted

Default Value:

/var/flyte/inputs
default-output-path (string)#

Default path where the volume should be mounted

Default Value:

/var/flyte/outputs
input-vol-name (string)#

Name of the data volume that is created for storing inputs

Default Value:

flyte-inputs
output-vol-name (string)#

Name of the data volume that is created for storing outputs

Default Value:

flyte-outputs
start-timeout (config.Duration)#

Default Value:

1m40s
cpu (string)#

Used to set cpu for co-pilot containers

Default Value:

500m
memory (string)#

Used to set memory for co-pilot containers

Default Value:

128Mi
storage (string)#

Default storage limit for individual inputs / outputs

Default Value:

""

resource.Quantity#

i (resource.int64Amount)#

Default Value:

{}
d (resource.infDecAmount)#

Default Value:

<nil>
s (string)#

Default Value:

"1"
Format (string)#

Default Value:

DecimalSI

resource.infDecAmount#

Dec (inf.Dec)#

Default Value:

null

resource.int64Amount#

value (int64)#

Default Value:

"1"
scale (int32)#

Default Value:

"0"

databricks.Config#

webApi (webapi.PluginConfig)#

Defines config for the base WebAPI plugin.

Default Value:

caching:
  maxSystemFailures: 5
  resyncInterval: 30s
  size: 500000
  workers: 10
readRateLimiter:
  burst: 100
  qps: 10
resourceMeta: null
resourceQuotas:
  default: 1000
writeRateLimiter:
  burst: 100
  qps: 10
resourceConstraints (core.ResourceConstraintsSpec)#

Default Value:

NamespaceScopeResourceConstraint:
  Value: 50
ProjectScopeResourceConstraint:
  Value: 100
defaultWarehouse (string)#

Defines the default warehouse to use when running on Databricks unless overwritten by the task.

Default Value:

COMPUTE_CLUSTER
databricksTokenKey (string)#

Name of the key where to find Databricks token in the secret manager.

Default Value:

FLYTE_DATABRICKS_API_TOKEN
databricksInstance (string)#

Databricks workspace instance name.

Default Value:

""
entrypointFile (string)#

A URL of the entrypoint file. DBFS and cloud storage (s3://, gcs://, adls://, etc) locations are supported.

Default Value:

""
databricksEndpoint (string)#

Default Value:

""

k8s.Config#

scheduler (string)#

Decides the scheduler to use when launching array-pods.

Default Value:

""
maxErrorLength (int)#

Determines the maximum length of the error string returned for the array.

Default Value:

"1000"
maxArrayJobSize (int64)#

Maximum size of array job.

Default Value:

"5000"
resourceConfig (k8s.ResourceConfig)#

Default Value:

limit: 0
primaryLabel: ""
remoteClusterConfig (k8s.ClusterConfig (remoteClusterConfig))#

Default Value:

auth:
  certPath: ""
  tokenPath: ""
  type: ""
enabled: false
endpoint: ""
name: ""
node-selector (map[string]string)#

Default Value:

null
tolerations ([]v1.Toleration)#

Default Value:

null
namespaceTemplate (string)#

Default Value:

""
OutputAssembler (workqueue.Config)#

Default Value:

maxItems: 100000
maxRetries: 5
workers: 10
ErrorAssembler (workqueue.Config)#

Default Value:

maxItems: 100000
maxRetries: 5
workers: 10
logs (k8s.LogConfig)#

Config for log links for k8s array jobs.

Default Value:

config:
  cloudwatch-enabled: false
  cloudwatch-log-group: ""
  cloudwatch-region: ""
  cloudwatch-template-uri: ""
  dynamic-log-links: null
  gcp-project: ""
  kubernetes-enabled: true
  kubernetes-template-uri: http://localhost:30082/#!/log/{{ .namespace }}/{{ .podName
    }}/pod?namespace={{ .namespace }}
  kubernetes-url: ""
  stackdriver-enabled: false
  stackdriver-logresourcename: ""
  stackdriver-template-uri: ""
  templates: null

k8s.ClusterConfig (remoteClusterConfig)#

name (string)#

Friendly name of the remote cluster

Default Value:

""
endpoint (string)#

Remote K8s cluster endpoint

Default Value:

""
auth (k8s.Auth (auth))#

Default Value:

certPath: ""
tokenPath: ""
type: ""
enabled (bool)#

Boolean flag to enable or disable

Default Value:

"false"

k8s.Auth (auth)#

type (string)#

Authentication type

Default Value:

""
tokenPath (string)#

Token path

Default Value:

""
certPath (string)#

Certificate path

Default Value:

""

k8s.LogConfig#

config (logs.LogConfig (config))#

Defines the log config for k8s logs.

Default Value:

cloudwatch-enabled: false
cloudwatch-log-group: ""
cloudwatch-region: ""
cloudwatch-template-uri: ""
dynamic-log-links: null
gcp-project: ""
kubernetes-enabled: true
kubernetes-template-uri: http://localhost:30082/#!/log/{{ .namespace }}/{{ .podName
  }}/pod?namespace={{ .namespace }}
kubernetes-url: ""
stackdriver-enabled: false
stackdriver-logresourcename: ""
stackdriver-template-uri: ""
templates: null

logs.LogConfig (config)#

cloudwatch-enabled (bool)#

Enable Cloudwatch Logging

Default Value:

"false"
cloudwatch-region (string)#

AWS region in which Cloudwatch logs are stored.

Default Value:

""
cloudwatch-log-group (string)#

Log group to which streams are associated.

Default Value:

""
cloudwatch-template-uri (string)#

Template Uri to use when building cloudwatch log links

Default Value:

""
kubernetes-enabled (bool)#

Enable Kubernetes Logging

Default Value:

"true"
kubernetes-url (string)#

Console URL for Kubernetes logs

Default Value:

""
kubernetes-template-uri (string)#

Template Uri to use when building kubernetes log links

Default Value:

http://localhost:30082/#!/log/{{ .namespace }}/{{ .podName }}/pod?namespace={{ .namespace
  }}
stackdriver-enabled (bool)#

Enable Log-links to stackdriver

Default Value:

"false"
gcp-project (string)#

Name of the project in GCP

Default Value:

""
stackdriver-logresourcename (string)#

Name of the logresource in stackdriver

Default Value:

""
stackdriver-template-uri (string)#

Template Uri to use when building stackdriver log links

Default Value:

""
templates ([]tasklog.TemplateLogPlugin)#

Default Value:

null

k8s.ResourceConfig#

primaryLabel (string)#

PrimaryLabel of a given service cluster

Default Value:

""
limit (int)#

Resource quota (in the number of outstanding requests) for the cluster

Default Value:

"0"

logs.LogConfig#

cloudwatch-enabled (bool)#

Enable Cloudwatch Logging

Default Value:

"false"
cloudwatch-region (string)#

AWS region in which Cloudwatch logs are stored.

Default Value:

""
cloudwatch-log-group (string)#

Log group to which streams are associated.

Default Value:

""
cloudwatch-template-uri (string)#

Template Uri to use when building cloudwatch log links

Default Value:

""
kubernetes-enabled (bool)#

Enable Kubernetes Logging

Default Value:

"true"
kubernetes-url (string)#

Console URL for Kubernetes logs

Default Value:

""
kubernetes-template-uri (string)#

Template Uri to use when building kubernetes log links

Default Value:

http://localhost:30082/#!/log/{{ .namespace }}/{{ .podName }}/pod?namespace={{ .namespace
  }}
stackdriver-enabled (bool)#

Enable Log-links to stackdriver

Default Value:

"false"
gcp-project (string)#

Name of the project in GCP

Default Value:

""
stackdriver-logresourcename (string)#

Name of the logresource in stackdriver

Default Value:

""
stackdriver-template-uri (string)#

Template Uri to use when building stackdriver log links

Default Value:

""
dynamic-log-links (map[string]tasklog.TemplateLogPlugin)#

Default Value:

null
templates ([]tasklog.TemplateLogPlugin)#

Default Value:

null

ray.Config#

shutdownAfterJobFinishes (bool)#

Default Value:

"true"
ttlSecondsAfterFinished (int32)#

Default Value:

"3600"
serviceType (string)#

Default Value:

NodePort
includeDashboard (bool)#

Default Value:

"true"
dashboardHost (string)#

Default Value:

0.0.0.0
nodeIPAddress (string)#

Default Value:

""
remoteClusterConfig (k8s.ClusterConfig)#

Configuration of remote K8s cluster for ray jobs

Default Value:

auth:
  caCertPath: ""
  tokenPath: ""
enabled: false
endpoint: ""
name: ""
logs (logs.LogConfig)#

Default Value:

cloudwatch-enabled: false
cloudwatch-log-group: ""
cloudwatch-region: ""
cloudwatch-template-uri: ""
dynamic-log-links: null
gcp-project: ""
kubernetes-enabled: false
kubernetes-template-uri: ""
kubernetes-url: ""
stackdriver-enabled: false
stackdriver-logresourcename: ""
stackdriver-template-uri: ""
templates: null
logsSidecar (v1.Container)#

Default Value:

null
dashboardURLTemplate (tasklog.TemplateLogPlugin)#

Default Value:

null
defaults (ray.DefaultConfig)#

Default Value:

headNode:
  ipAddress: $MY_POD_IP
  startParameters:
    disable-usage-stats: "true"
workerNode:
  ipAddress: $MY_POD_IP
  startParameters:
    disable-usage-stats: "true"
enableUsageStats (bool)#

Enable usage stats for ray jobs. These stats are submitted to usage-stats.ray.io per https://docs.ray.io/en/latest/cluster/usage-stats.html

Default Value:

"false"
serviceAccount (string)#

The k8s service account to run as

Default Value:

default

ray.DefaultConfig#

headNode (ray.NodeConfig)#

Default Value:

ipAddress: $MY_POD_IP
startParameters:
  disable-usage-stats: "true"
workerNode (ray.NodeConfig)#

Default Value:

ipAddress: $MY_POD_IP
startParameters:
  disable-usage-stats: "true"

ray.NodeConfig#

startParameters (map[string]string)#

Default Value:

disable-usage-stats: "true"
ipAddress (string)#

Default Value:

$MY_POD_IP

snowflake.Config#

webApi (webapi.PluginConfig)#

Defines config for the base WebAPI plugin.

Default Value:

caching:
  maxSystemFailures: 5
  resyncInterval: 30s
  size: 500000
  workers: 10
readRateLimiter:
  burst: 100
  qps: 10
resourceMeta: null
resourceQuotas:
  default: 1000
writeRateLimiter:
  burst: 100
  qps: 10
resourceConstraints (core.ResourceConstraintsSpec)#

Default Value:

NamespaceScopeResourceConstraint:
  Value: 50
ProjectScopeResourceConstraint:
  Value: 100
defaultWarehouse (string)#

Defines the default warehouse to use when running on Snowflake unless overwritten by the task.

Default Value:

COMPUTE_WH
snowflakeTokenKey (string)#

Name of the key where to find Snowflake token in the secret manager.

Default Value:

FLYTE_SNOWFLAKE_CLIENT_TOKEN
snowflakeEndpoint (string)#

Default Value:

""

spark.Config#

spark-config-default (map[string]string)#

Default Value:

null
spark-history-server-url (string)#

URL for SparkHistory Server that each job will publish the execution history to.

Default Value:

""
features ([]spark.Feature)#

Default Value:

null
logs (spark.LogConfig)#

Config for log links for spark applications.

Default Value:

all-user:
  cloudwatch-enabled: false
  cloudwatch-log-group: ""
  cloudwatch-region: ""
  cloudwatch-template-uri: ""
  dynamic-log-links: null
  gcp-project: ""
  kubernetes-enabled: false
  kubernetes-template-uri: ""
  kubernetes-url: ""
  stackdriver-enabled: false
  stackdriver-logresourcename: ""
  stackdriver-template-uri: ""
  templates: null
mixed:
  cloudwatch-enabled: false
  cloudwatch-log-group: ""
  cloudwatch-region: ""
  cloudwatch-template-uri: ""
  dynamic-log-links: null
  gcp-project: ""
  kubernetes-enabled: true
  kubernetes-template-uri: http://localhost:30082/#!/log/{{ .namespace }}/{{ .podName
    }}/pod?namespace={{ .namespace }}
  kubernetes-url: ""
  stackdriver-enabled: false
  stackdriver-logresourcename: ""
  stackdriver-template-uri: ""
  templates: null
system:
  cloudwatch-enabled: false
  cloudwatch-log-group: ""
  cloudwatch-region: ""
  cloudwatch-template-uri: ""
  dynamic-log-links: null
  gcp-project: ""
  kubernetes-enabled: false
  kubernetes-template-uri: ""
  kubernetes-url: ""
  stackdriver-enabled: false
  stackdriver-logresourcename: ""
  stackdriver-template-uri: ""
  templates: null
user:
  cloudwatch-enabled: false
  cloudwatch-log-group: ""
  cloudwatch-region: ""
  cloudwatch-template-uri: ""
  dynamic-log-links: null
  gcp-project: ""
  kubernetes-enabled: false
  kubernetes-template-uri: ""
  kubernetes-url: ""
  stackdriver-enabled: false
  stackdriver-logresourcename: ""
  stackdriver-template-uri: ""
  templates: null

spark.LogConfig#

mixed (logs.LogConfig)#

Defines the log config that’s not split into user/system.

Default Value:

cloudwatch-enabled: false
cloudwatch-log-group: ""
cloudwatch-region: ""
cloudwatch-template-uri: ""
dynamic-log-links: null
gcp-project: ""
kubernetes-enabled: true
kubernetes-template-uri: http://localhost:30082/#!/log/{{ .namespace }}/{{ .podName
  }}/pod?namespace={{ .namespace }}
kubernetes-url: ""
stackdriver-enabled: false
stackdriver-logresourcename: ""
stackdriver-template-uri: ""
templates: null
user (logs.LogConfig)#

Defines the log config for user logs.

Default Value:

cloudwatch-enabled: false
cloudwatch-log-group: ""
cloudwatch-region: ""
cloudwatch-template-uri: ""
dynamic-log-links: null
gcp-project: ""
kubernetes-enabled: false
kubernetes-template-uri: ""
kubernetes-url: ""
stackdriver-enabled: false
stackdriver-logresourcename: ""
stackdriver-template-uri: ""
templates: null
system (logs.LogConfig)#

Defines the log config for system logs.

Default Value:

cloudwatch-enabled: false
cloudwatch-log-group: ""
cloudwatch-region: ""
cloudwatch-template-uri: ""
dynamic-log-links: null
gcp-project: ""
kubernetes-enabled: false
kubernetes-template-uri: ""
kubernetes-url: ""
stackdriver-enabled: false
stackdriver-logresourcename: ""
stackdriver-template-uri: ""
templates: null
all-user (logs.LogConfig)#

All user logs across driver and executors.

Default Value:

cloudwatch-enabled: false
cloudwatch-log-group: ""
cloudwatch-region: ""
cloudwatch-template-uri: ""
dynamic-log-links: null
gcp-project: ""
kubernetes-enabled: false
kubernetes-template-uri: ""
kubernetes-url: ""
stackdriver-enabled: false
stackdriver-logresourcename: ""
stackdriver-template-uri: ""
templates: null

testing.Config#

sleep-duration (config.Duration)#

Indicates the amount of time before transitioning to success

Default Value:

0s

Section: propeller#

kube-config (string)#

Path to kubernetes client config file.

Default Value:

""

master (string)#

Default Value:

""

workers (int)#

Number of threads to process workflows

Default Value:

"20"

workflow-reeval-duration (config.Duration)#

Frequency of re-evaluating workflows

Default Value:

10s

downstream-eval-duration (config.Duration)#

Frequency of re-evaluating downstream tasks

Default Value:

30s

limit-namespace (string)#

Namespaces to watch for this propeller

Default Value:

all

prof-port (config.Port)#

Profiler port

Default Value:

10254

metadata-prefix (string)#

MetadataPrefix should be used if all the metadata for Flyte executions should be stored under a specific prefix in CloudStorage. If not specified, the data will be stored in the base container directly.

Default Value:

metadata/propeller

rawoutput-prefix (string)#

a fully qualified storage path of the form s3://flyte/abc/…, where all data sandboxes should be stored.

Default Value:

""

queue (config.CompositeQueueConfig)#

Workflow workqueue configuration, affects the way the work is consumed from the queue.

Default Value:

batch-size: -1
batching-interval: 1s
queue:
  base-delay: 0s
  capacity: 10000
  max-delay: 1m0s
  rate: 1000
  type: maxof
sub-queue:
  base-delay: 0s
  capacity: 10000
  max-delay: 0s
  rate: 1000
  type: bucket
type: batch

metrics-prefix (string)#

An optional prefix for all published metrics.

Default Value:

flyte

metrics-keys ([]string)#

Metrics labels applied to prometheus metrics emitted by the service.

Default Value:

- project
- domain
- wf
- task

enable-admin-launcher (bool)#

Enable remote Workflow launcher to Admin

Default Value:

"true"

max-workflow-retries (int)#

Maximum number of retries per workflow

Default Value:

"10"

max-ttl-hours (int)#

Maximum number of hours a completed workflow should be retained. Number between 1-23 hours

Default Value:

"23"

gc-interval (config.Duration)#

Run periodic GC every 30 minutes

Default Value:

30m0s

leader-election (config.LeaderElectionConfig)#

Config for leader election.

Default Value:

enabled: false
lease-duration: 15s
lock-config-map:
  Name: ""
  Namespace: ""
renew-deadline: 10s
retry-period: 2s

publish-k8s-events (bool)#

Enable events publishing to K8s events API.

Default Value:

"false"

max-output-size-bytes (int64)#

Deprecated! Use storage.limits.maxDownloadMBs instead

Default Value:

"-1"

enable-grpc-latency-metrics (bool)#

Enable grpc latency metrics. Note Histograms metrics can be expensive on Prometheus servers.

Default Value:

"false"

kube-client-config (config.KubeClientConfig)#

Configuration to control the Kubernetes client

Default Value:

burst: 25
qps: 100
timeout: 30s

node-config (config.NodeConfig)#

config for a workflow node

Default Value:

default-deadlines:
  node-active-deadline: 0s
  node-execution-deadline: 0s
  workflow-active-deadline: 0s
default-max-attempts: 1
enable-cr-debug-metadata: false
ignore-retry-cause: false
interruptible-failure-threshold: -1
max-node-retries-system-failures: 3

max-streak-length (int)#

Maximum number of consecutive rounds that one propeller worker can use for one workflow - >1 => turbo-mode is enabled.

Default Value:

"8"

event-config (config.EventConfig)#

Configures execution event behavior.

Default Value:

fallback-to-output-reference: false
raw-output-policy: reference

include-shard-key-label ([]string)#

Include the specified shard key label in the k8s FlyteWorkflow CRD label selector

Default Value:

[]

exclude-shard-key-label ([]string)#

Exclude the specified shard key label from the k8s FlyteWorkflow CRD label selector

Default Value:

[]

include-project-label ([]string)#

Include the specified project label in the k8s FlyteWorkflow CRD label selector

Default Value:

[]

exclude-project-label ([]string)#

Exclude the specified project label from the k8s FlyteWorkflow CRD label selector

Default Value:

[]

include-domain-label ([]string)#

Include the specified domain label in the k8s FlyteWorkflow CRD label selector

Default Value:

[]

exclude-domain-label ([]string)#

Exclude the specified domain label from the k8s FlyteWorkflow CRD label selector

Default Value:

[]

cluster-id (string)#

Unique cluster id running this flytepropeller instance with which to annotate execution events

Default Value:

propeller

create-flyteworkflow-crd (bool)#

Enable creation of the FlyteWorkflow CRD on startup

Default Value:

"false"

node-execution-worker-count (int)#

Number of workers to evaluate node executions, currently only used for array nodes

Default Value:

"8"

array-node-config (config.ArrayNodeConfig)#

Configuration for array nodes

Default Value:

default-parallelism-behavior: unlimited
event-version: 0

literal-offloading-config (config.LiteralOffloadingConfig)#

config used for literal offloading.

Default Value:

Enabled: false
max-size-in-mb-for-offloading: 1000
min-size-in-mb-for-offloading: 10
supported-sdk-versions:
  FLYTE_SDK: 1.13.5

admin-launcher (launchplan.AdminConfig)#

Default Value:

burst: 10
cacheSize: 10000
tps: 100
workers: 10

resourcemanager (config.Config (resourcemanager))#

Default Value:

redis:
  hostKey: ""
  hostPath: ""
  hostPaths: []
  maxRetries: 0
  primaryName: ""
resourceMaxQuota: 1000
type: noop

workflowstore (workflowstore.Config)#

Default Value:

policy: ResourceVersionCache

config.ArrayNodeConfig#

event-version (int)#

ArrayNode eventing version. 0 => legacy (drop-in replacement for maptask), 1 => new

Default Value:

"0"
default-parallelism-behavior (string)#

Default parallelism behavior for array nodes

Default Value:

unlimited

config.CompositeQueueConfig#

type (string)#

Type of composite queue to use for the WorkQueue

Default Value:

batch
queue (config.WorkqueueConfig)#

Workflow workqueue configuration, affects the way the work is consumed from the queue.

Default Value:

base-delay: 0s
capacity: 10000
max-delay: 1m0s
rate: 1000
type: maxof
sub-queue (config.WorkqueueConfig)#

SubQueue configuration, affects the way the nodes cause the top-level Work to be re-evaluated.

Default Value:

base-delay: 0s
capacity: 10000
max-delay: 0s
rate: 1000
type: bucket
batching-interval (config.Duration)#

Duration for which downstream updates are buffered

Default Value:

1s
batch-size (int)#

Default Value:

"-1"

config.WorkqueueConfig#

type (string)#

Type of RateLimiter to use for the WorkQueue

Default Value:

maxof
base-delay (config.Duration)#

base backoff delay for failure

Default Value:

0s
max-delay (config.Duration)#

Max backoff delay for failure

Default Value:

1m0s
rate (int64)#

Bucket Refill rate per second

Default Value:

"1000"
capacity (int)#

Bucket capacity as number of items

Default Value:

"10000"

config.Config (resourcemanager)#

type (string)#

Which resource manager to use, redis or noop. Default is noop.

Default Value:

noop
resourceMaxQuota (int)#

Global limit for concurrent Qubole queries

Default Value:

"1000"
redis (config.RedisConfig)#

Config for Redis resourcemanager.

Default Value:

hostKey: ""
hostPath: ""
hostPaths: []
maxRetries: 0
primaryName: ""

config.RedisConfig#

hostPaths ([]string)#

Redis hosts locations.

Default Value:

[]
primaryName (string)#

Redis primary name, fill in only if you are connecting to a redis sentinel cluster.

Default Value:

""
hostPath (string)#

Redis host location

Default Value:

""
hostKey (string)#

Key for local Redis access

Default Value:

""
maxRetries (int)#

See Redis client options for more info

Default Value:

"0"

config.EventConfig#

raw-output-policy (string)#

How output data should be passed along in execution events.

Default Value:

reference
fallback-to-output-reference (bool)#

Whether output data should be sent by reference when it is too large to be sent inline in execution events.

Default Value:

"false"
ErrorOnAlreadyExists (bool)#

Default Value:

"false"

config.KubeClientConfig#

qps (float32)#

Default Value:

"100"
burst (int)#

Max burst rate for throttle. 0 defaults to 10

Default Value:

"25"
timeout (config.Duration)#

Max duration allowed for every request to KubeAPI before giving up. 0 implies no timeout.

Default Value:

30s

config.LeaderElectionConfig#

enabled (bool)#

Enables/Disables leader election.

Default Value:

"false"
lock-config-map (types.NamespacedName)#

ConfigMap namespace/name to use for resource lock.

Default Value:

Name: ""
Namespace: ""
lease-duration (config.Duration)#

Duration that non-leader candidates will wait to force acquire leadership. This is measured against time of last observed ack.

Default Value:

15s
renew-deadline (config.Duration)#

Duration that the acting master will retry refreshing leadership before giving up.

Default Value:

10s
retry-period (config.Duration)#

Duration the LeaderElector clients should wait between tries of actions.

Default Value:

2s

types.NamespacedName#

Namespace (string)#

Default Value:

""
Name (string)#

Default Value:

""

config.LiteralOffloadingConfig#

Enabled (bool)#

Default Value:

"false"
supported-sdk-versions (map[string]string)#

Maps flytekit and union SDK names to minimum supported version that can handle reading offloaded literals.

Default Value:

FLYTE_SDK: 1.13.5
min-size-in-mb-for-offloading (int64)#

Size of a literal at which to trigger offloading

Default Value:

"10"
max-size-in-mb-for-offloading (int64)#

Size of a literal at which to fail fast

Default Value:

"1000"

config.NodeConfig#

default-deadlines (config.DefaultDeadlines)#

Default value for timeouts

Default Value:

node-active-deadline: 0s
node-execution-deadline: 0s
workflow-active-deadline: 0s
max-node-retries-system-failures (int64)#

Maximum number of retries per node for node failure due to infra issues

Default Value:

"3"
interruptible-failure-threshold (int32)#

number of failures for a node to be still considered interruptible. Negative numbers are treated as complementary (ex. -1 means last attempt is non-interruptible).’

Default Value:

"-1"
default-max-attempts (int32)#

Default maximum number of attempts for a node

Default Value:

"1"
ignore-retry-cause (bool)#

Ignore retry cause and count all attempts toward a node’s max attempts

Default Value:

"false"
enable-cr-debug-metadata (bool)#

Collapse node on any terminal state, not just successful terminations. This is useful to reduce the size of workflow state in etcd.

Default Value:

"false"

config.DefaultDeadlines#

node-execution-deadline (config.Duration)#

Default value of node execution timeout that includes the time spent to run the node/workflow

Default Value:

0s
node-active-deadline (config.Duration)#

Default value of node timeout that includes the time spent queued.

Default Value:

0s
workflow-active-deadline (config.Duration)#

Default value of workflow timeout that includes the time spent queued.

Default Value:

0s

config.Port#

port (int)#

Default Value:

"10254"

launchplan.AdminConfig#

tps (int64)#

The maximum number of transactions per second to flyte admin from this client.

Default Value:

"100"
burst (int)#

Maximum burst for throttle

Default Value:

"10"
cacheSize (int)#

Maximum cache in terms of number of items stored.

Default Value:

"10000"
workers (int)#

Number of parallel workers to work on the queue.

Default Value:

"10"

workflowstore.Config#

policy (string)#

Workflow Store Policy to initialize

Default Value:

ResourceVersionCache

Section: secrets#

secrets-prefix (string)#

Prefix where to look for secrets file

Default Value:

/etc/secrets

env-prefix (string)#

Prefix for environment variables

Default Value:

FLYTE_SECRET_

Section: storage#

type (string)#

Sets the type of storage to configure [s3/minio/local/mem/stow].

Default Value:

s3

connection (storage.ConnectionConfig)#

Default Value:

access-key: ""
auth-type: iam
disable-ssl: false
endpoint: ""
region: us-east-1
secret-key: ""

stow (storage.StowConfig)#

Storage config for stow backend.

Default Value:

{}

container (string)#

Initial container (in s3 a bucket) to create -if it doesn’t exist-.’

Default Value:

""

enable-multicontainer (bool)#

If this is true, then the container argument is overlooked and redundant. This config will automatically open new connections to new containers/buckets as they are encountered

Default Value:

"false"

cache (storage.CachingConfig)#

Default Value:

max_size_mbs: 0
target_gc_percent: 0

limits (storage.LimitsConfig)#

Sets limits for stores.

Default Value:

maxDownloadMBs: 2

defaultHttpClient (storage.HTTPClientConfig)#

Sets the default http client config.

Default Value:

headers: null
timeout: 0s

signedUrl (storage.SignedURLConfig)#

Sets config for SignedURL.

Default Value:

{}

storage.CachingConfig#

max_size_mbs (int)#

Maximum size of the cache where the Blob store data is cached in-memory. If not specified or set to 0, cache is not used

Default Value:

"0"
target_gc_percent (int)#

Sets the garbage collection target percentage.

Default Value:

"0"

storage.ConnectionConfig#

endpoint (config.URL)#

URL for storage client to connect to.

Default Value:

""
auth-type (string)#

Auth Type to use [iam,accesskey].

Default Value:

iam
access-key (string)#

Access key to use. Only required when authtype is set to accesskey.

Default Value:

""
secret-key (string)#

Secret to use when accesskey is set.

Default Value:

""
region (string)#

Region to connect to.

Default Value:

us-east-1
disable-ssl (bool)#

Disables SSL connection. Should only be used for development.

Default Value:

"false"

storage.HTTPClientConfig#

headers (map[string][]string)#

Default Value:

null
timeout (config.Duration)#

Sets time out on the http client.

Default Value:

0s

storage.LimitsConfig#

maxDownloadMBs (int64)#

Maximum allowed download size (in MBs) per call.

Default Value:

"2"

storage.SignedURLConfig#

stowConfigOverride (map[string]string)#

Default Value:

null

storage.StowConfig#

kind (string)#

Kind of Stow backend to use. Refer to github/flyteorg/stow

Default Value:

""
config (map[string]string)#

Configuration for stow backend. Refer to github/flyteorg/stow

Default Value:

{}

Section: tasks#

task-plugins (config.TaskPluginConfig)#

Task plugin configuration

Default Value:

default-for-task-types: {}
enabled-plugins: []

max-plugin-phase-versions (int32)#

Maximum number of plugin phase versions allowed for one phase.

Default Value:

"100000"

backoff (config.BackOffConfig)#

Config for Exponential BackOff implementation

Default Value:

base-second: 2
max-duration: 20s

maxLogMessageLength (int)#

Deprecated!!! Max length of error message.

Default Value:

"0"

config.BackOffConfig#

base-second (int)#

The number of seconds representing the base duration of the exponential backoff

Default Value:

"2"
max-duration (config.Duration)#

The cap of the backoff duration

Default Value:

20s

config.TaskPluginConfig#

enabled-plugins ([]string)#

Plugins enabled currently

Default Value:

[]
default-for-task-types (map[string]string)#

Default Value:

{}

Section: webhook#

metrics-prefix (string)#

An optional prefix for all published metrics.

Default Value:

'flyte:'

certDir (string)#

Certificate directory to use to write generated certs. Defaults to /etc/webhook/certs/

Default Value:

/etc/webhook/certs

localCert (bool)#

write certs locally. Defaults to false

Default Value:

"false"

listenPort (int)#

The port to use to listen to webhook calls. Defaults to 9443

Default Value:

"9443"

serviceName (string)#

The name of the webhook service.

Default Value:

flyte-pod-webhook

servicePort (int32)#

The port on the service that hosting webhook.

Default Value:

"443"

secretName (string)#

Secret name to write generated certs to.

Default Value:

flyte-pod-webhook

secretManagerType (int)#

Default Value:

K8s

awsSecretManager (config.AWSSecretManagerConfig)#

AWS Secret Manager config.

Default Value:

resources:
  limits:
    cpu: 200m
    memory: 500Mi
  requests:
    cpu: 200m
    memory: 500Mi
sidecarImage: docker.io/amazon/aws-secrets-manager-secret-sidecar:v0.1.4

gcpSecretManager (config.GCPSecretManagerConfig)#

GCP Secret Manager config.

Default Value:

resources:
  limits:
    cpu: 200m
    memory: 500Mi
  requests:
    cpu: 200m
    memory: 500Mi
sidecarImage: gcr.io/google.com/cloudsdktool/cloud-sdk:alpine

vaultSecretManager (config.VaultSecretManagerConfig)#

Vault Secret Manager config.

Default Value:

annotations: null
kvVersion: "2"
role: flyte

config.AWSSecretManagerConfig#

sidecarImage (string)#

Specifies the sidecar docker image to use

Default Value:

docker.io/amazon/aws-secrets-manager-secret-sidecar:v0.1.4
resources (v1.ResourceRequirements)#

Default Value:

limits:
  cpu: 200m
  memory: 500Mi
requests:
  cpu: 200m
  memory: 500Mi

v1.ResourceRequirements#

limits (v1.ResourceList)#

Default Value:

cpu: 200m
memory: 500Mi
requests (v1.ResourceList)#

Default Value:

cpu: 200m
memory: 500Mi
claims ([]v1.ResourceClaim)#

Default Value:

null

config.GCPSecretManagerConfig#

sidecarImage (string)#

Specifies the sidecar docker image to use

Default Value:

gcr.io/google.com/cloudsdktool/cloud-sdk:alpine
resources (v1.ResourceRequirements)#

Default Value:

limits:
  cpu: 200m
  memory: 500Mi
requests:
  cpu: 200m
  memory: 500Mi

config.VaultSecretManagerConfig#

role (string)#

Specifies the vault role to use

Default Value:

flyte
kvVersion (int)#

Default Value:

"2"
annotations (map[string]string)#

Default Value:

null