Skip to content

CRD reference

This page describes every field of the custom resources provided by the Stroom K8s Operator, and every type nested within them. It is generated from the Go type definitions in api/v1, which are also the source of the CRD schemas installed into your cluster.

Generated page

Do not edit this page by hand. Change the doc comments in api/v1, then run hack/gen-crd-docs.sh. See Contributing to the docs.

You can also query the schema of any field from your cluster, for example:

kubectl explain stroomcluster.spec.nodeSets.livenessProbeTimings

Fields of standard Kubernetes types (such as PersistentVolumeClaimSpec or ResourceRequirements) link to the Kubernetes API reference.

Resource types

All resources belong to the API group and version stroom.gchq.github.io/v1.

Kind Description
DatabaseBackup A scheduled mysqldump of a DatabaseServer or external MySQL database, run as a CronJob.
DatabaseServer An operator-managed, single-instance MySQL server for a StroomCluster.
StroomCluster A Stroom cluster: one StatefulSet of Stroom nodes per NodeSet, plus the Services, Ingresses and configuration they need.
StroomTaskAutoscaler Adjusts the task limit of a Stroom job on each processing node, based on Pod CPU usage.

DatabaseBackup

DatabaseBackup backs up the databases on a database server on a schedule, with mysqldump

Field Description Default Validation
apiVersion string stroom.gchq.github.io/v1
kind string DatabaseBackup
metadata ObjectMeta Standard Kubernetes object metadata.
spec DatabaseBackupSpec
status DatabaseBackupStatus

DatabaseServer

DatabaseServer is a MySQL database server for a StroomCluster, managed by the operator

Field Description Default Validation
apiVersion string stroom.gchq.github.io/v1
kind string DatabaseServer
metadata ObjectMeta Standard Kubernetes object metadata.
spec DatabaseServerSpec
status DatabaseServerStatus
stroomClusterRef ResourceRef Set by the controller when a StroomCluster binds to the DatabaseServer.
This is used to prevent the DatabaseServer from being deleted while its paired StroomCluster still exists.
Optional

StroomCluster

StroomCluster is a cluster of Stroom nodes, in one or more NodeSets

Field Description Default Validation
apiVersion string stroom.gchq.github.io/v1
kind string StroomCluster
metadata ObjectMeta Standard Kubernetes object metadata.
spec StroomClusterSpec
status StroomClusterStatus

StroomTaskAutoscaler

StroomTaskAutoscaler adjusts the task limit of a Stroom job on a StroomCluster's nodes, according to their CPU use

Field Description Default Validation
apiVersion string stroom.gchq.github.io/v1
kind string StroomTaskAutoscaler
metadata ObjectMeta Standard Kubernetes object metadata.
spec StroomTaskAutoscalerSpec
status StroomTaskAutoscalerStatus

Supporting types

Types used within the resources above, in alphabetical order.

ConfigMapRef

Appears in: StroomClusterSpec

Field Description Default Validation
name string Name of the ConfigMap
itemName string ConfigMap key containing the Stroom configuration data

DatabaseBackupSpec

DatabaseBackupSpec defines the desired state of DatabaseBackup

Appears in: DatabaseBackup

Field Description Default Validation
image Image Image the backup runs in, which must have bash, mysqldump and gzip, such as the official mysql:8.4
image. The mysql/mysql-server images have no gzip.
Required
imagePullPolicy PullPolicy
databaseServerRef DatabaseServerRef DatabaseServerRef contains either the details of a DatabaseServer resource, or the TCP connection details of
an external MySQL database
Required
databaseNames string array Backup the specified database names. If unspecified, all user databases are backed up items:Pattern: ^[A-Za-z0-9_]{1,64}$
volume VolumeSource File system location to store the backup files
schedule string Cron schedule that determines how often backups are to be performed

DatabaseBackupStatus

DatabaseBackupStatus defines the observed state of DatabaseBackup

Appears in: DatabaseBackup

DatabaseServerRef

Appears in: DatabaseBackupSpec, StroomClusterSpec

Field Description Default Validation
serverRef ResourceRef If specified, point to an operator-managed DatabaseServer object Optional
serverAddress ServerAddress Alternatively, if the following parameters are provided, point directly to a DB by its TCP address.
This allows external database instances to be used in place of an operator-managed one.
userName string UserName is the name of the Stroom database user to use when connecting to the server.
A corresponding key should exist in the provided Secret referenced by SecretName.
stroomuser

DatabaseServerSpec

DatabaseServerSpec defines the desired state of DatabaseServer

Appears in: DatabaseServer

Field Description Default Validation
image Image
imagePullPolicy PullPolicy
databaseNames string array Names of the databases to be created on startup (if they don't already exist) items:Pattern: ^[A-Za-z0-9_]{1,64}$
additionalConfig string array Any additional configuration lines to append to the [mysqld] section of the MySQL server configuration file
/etc/my.cnf, each a single option, such as max_connections=500. Options that the operator sets, or that
would give access to the server's files or bypass authentication, aren't allowed.
MaxItems: 200
items:MaxLength: 1024
items:Pattern: ^[A-Za-z][A-Za-z0-9_-]*(\s*=.*)?$
resources ResourceRequirements
volumeClaim PersistentVolumeClaimSpec
readinessProbeTimings ProbeTimings
livenessProbeTimings ProbeTimings
podAnnotations object (keys:string, values:string)
podSecurityContext PodSecurityContext
securityContext SecurityContext
nodeSelector object (keys:string, values:string)
tolerations Toleration array
affinity Affinity

DatabaseServerStatus

DatabaseServerStatus defines the observed state of DatabaseServer

Appears in: DatabaseServer

Field Description Default Validation
state string
address string
port integer

HttpsSettings

Appears in: StroomClusterSpec

Field Description Default Validation
tlsSecretName string Name of the TLS Secret containing the nodes' certificate (tls.crt), its private key (tls.key) and the CA
certificate (ca.crt), from which each node's key store and trust store are created
tlsKeystorePasswordSecret SecretItem Password of the keystore and truststore

Image

Appears in: DatabaseBackupSpec, DatabaseServerSpec, LogSenderSettings, StroomClusterSpec

Field Description Default Validation
repository string MinLength: 1
tag string

IngressSettings

Appears in: StroomClusterSpec

Field Description Default Validation
hostName string DNS name at which the application will be reached (e.g. stroom.example.com)
secretName string Name of the TLS Secret containing the private key and server certificate for the Ingress. If omitted, the
Ingresses have no TLS section.
Optional
className string Ingress class name (e.g. nginx). If omitted, the cluster's default IngressClass is used.
pathTypeOverride boolean Override path type for all ingress resources as ImplementationSpecific

JvmMemoryOptions

Appears in: NodeSet

Field Description Default Validation
initialPercentage integer Maximum: 100
Minimum: 1
maxPercentage integer Maximum: 100
Minimum: 1

LivenessProbeTimings

LivenessProbeTimings determine the parameters for the Stroom node liveness probe, which queries the Stroom filtered health check endpoint. Allow and Deny filter which Stroom health checks are run. Each name must be a health check registered by Stroom, otherwise the probe will fail.

Appears in: NodeSet

Field Description Default Validation
initialDelaySeconds integer 5
timeoutSeconds integer 5
periodSeconds integer 5
successThreshold integer 1
failureThreshold integer 10
allow string Allow lists the health checks to run. If empty, all health checks are run, except those listed in Deny.
Defaults to the JVM thread deadlock check, so nodes are not restarted when an external dependency is down.
Set to an empty list to run all health checks.
[deadlocks] items:Pattern: ^[^,]+$
deny string Deny lists the health checks to exclude. items:Pattern: ^[^,]+$

LogSenderSettings

Appears in: StroomClusterSpec

Field Description Default Validation
enabled boolean If true, Stroom internal audit and application logs are sent to the Stroom Ingress for ingestion false
image Image Required
imagePullPolicy PullPolicy
securityContext SecurityContext Override the container security context
destinationUrl string Configure an alternate destination for events to be shipped to. If omitted, events are posted to the local cluster.
environmentName string Name of the Environment to set in feed metadata. If omitted, the cluster name is used (converted to UPPERCASE).
systemName string Name of the System to set in feed metadata
resources ResourceRequirements
tls TlsSettings

NodeRole

Underlying type: string

Validation: Enum: [Processing Frontend]

Appears in: NodeSet

Value Description
Processing ProcessingNodeRole applies to a NodeSet that is dedicated to receiving and processing data and does not serve
web front-end (UI) requests
Frontend FrontendNodeRole applies to a NodeSet that is dedicated to serving web front-end (UI) requests only

NodeSet

Appears in: StroomClusterSpec

Field Description Default Validation
name string Name uniquely identifies the NodeSet within a StroomCluster. Should be a short name like "prod".
The NodeSet name determines the name of the Stroom node, which follows the format:
stroom-<cluster name>-node-<nodeset name>-<pod ordinal>.
For example, the first Stroom node in StroomCluster dev, NodeSet data will be named:
stroom-dev-node-data-0.
count integer Number of replicas (discrete Stroom nodes) to deploy in the NodeSet Minimum: 1
role NodeRole Role of each node in the NodeSet. Processing nodes receive and process data, while Frontend nodes serve
the Stroom user interface and API.
If this property is omitted, nodes in the NodeSet will assume both Processing and Frontend roles.
Enum: [Processing Frontend]
managedJobs string array Automatically enable the specified job(s) on node startup and disable them prior to node shutdown.
If omitted, nodes with the Processing role or no role enable and disable all jobs, while Frontend nodes
only manage the job Reindex Content.
Jobs that aren't managed may be enabled or disabled via the Stroom Jobs pane.
localDataVolumeClaim PersistentVolumeClaimSpec LocalDataVolumeClaim provides persistent storage for each Stroom node's data
extraVolumeMounts VolumeMount array Extra volume mounts to add to each NodeSet pod, in addition to volume mounts defined at the StroomCluster level.
resources ResourceRequirements Resources determine how much CPU and memory each individual Stroom node Pod within the NodeSet requests and is
limited to.
memoryOptions JvmMemoryOptions MemoryOptions define JVM memory parameters
ingressEnabled boolean IngressEnabled determines whether this node receives requests via the created Kubernetes Ingresses. Usually this
should be true, unless there is a need for a NodeSet to be pure processing-only nodes, which cannot receive data.
true
ingressAnnotations object (keys:string, values:string) IngressAnnotations is an optional map of annotations to apply to the NodeSet's Ingress. These override any
default annotations provided by the controller.
ingressLabels object (keys:string, values:string) IngressLabels is an optional map of labels to apply to the NodeSet's Ingress. These take precedence over any
controller provided labels.
startupProbeTimings StartupProbeTimings StartupProbeTimings specify parameters for initial Pod startup. These should be set according to how long a node
typically takes to start up and respond to health checks.
{ }
readinessProbeTimings ReadinessProbeTimings ReadinessProbeTimings determine the parameters for readiness probes. These help to protect nodes from overload,
removing them from service while they are non-responsive to health checks.
{ }
livenessProbeTimings LivenessProbeTimings LivenessProbeTimings determine the parameters for liveness probes. If a Pod fails successive probes, it is
restarted.
{ }
podAnnotations object (keys:string, values:string) PodAnnotations are additional annotations to set for each NodeSet Pod
podSecurityContext PodSecurityContext PodSecurityContext applies to each NodeSet Pod
securityContext SecurityContext SecurityContext overrides the PodSecurityContext at the container level
nodeSelector object (keys:string, values:string) NodeSelector allows for NodeSet Pods to be deployed to a particular node, or set of nodes, by the specified labels
tolerations Toleration array
affinity Affinity

OpenIdConfiguration

Appears in: StroomClusterSpec

Field Description Default Validation
clientId string Name of the OpenID client
clientSecret SecretItem Details of the Secret containing the OpenID client secret

ProbeTimings

Appears in: DatabaseServerSpec, LivenessProbeTimings, ReadinessProbeTimings

Field Description Default Validation
initialDelaySeconds integer 5
timeoutSeconds integer 5
periodSeconds integer 5
successThreshold integer 1
failureThreshold integer 10

ReadinessProbeTimings

ReadinessProbeTimings determine the parameters for the Stroom node readiness probe, which queries the Stroom filtered health check endpoint. Allow and Deny filter which Stroom health checks are run. Each name must be a health check registered by Stroom, otherwise the probe will fail.

Appears in: NodeSet

Field Description Default Validation
initialDelaySeconds integer 5
timeoutSeconds integer 5
periodSeconds integer 5
successThreshold integer 1
failureThreshold integer 10
allow string Allow lists the health checks to run. If empty, all health checks are run, except those listed in Deny. items:Pattern: ^[^,]+$
deny string Deny lists the health checks to exclude. Defaults to the external identity provider check, so nodes keep
receiving data while the identity provider is unavailable. Set to an empty list to exclude no health checks.
[stroom.security.common.impl.ExternalIdpConfigurationProvider] items:Pattern: ^[^,]+$

ResourceRef

Appears in: DatabaseServer, DatabaseServerRef, StroomTaskAutoscalerSpec

Field Description Default Validation
name string
namespace string

SecretItem

Appears in: HttpsSettings, OpenIdConfiguration

Field Description Default Validation
secretName string
key string

ServerAddress

Appears in: DatabaseServerRef

Field Description Default Validation
host string Host is the hostname or IP of the database server
port integer Port number the database server is listening on 3306
secretName string SecretName is the name of the Secret containing the database user's password, under a key named after the
user (userName, by default stroomuser)

StartupProbeTimings

StartupProbeTimings determine how long a Stroom node is given to start up before it is restarted.

Appears in: NodeSet

Field Description Default Validation
initialDelaySeconds integer 5
timeoutSeconds integer 5
periodSeconds integer 5
successThreshold integer 1
failureThreshold integer FailureThreshold multiplied by PeriodSeconds is the base time allowed for a node to start up. 24
lockWaitAllowanceSeconds integer LockWaitAllowanceSeconds is added to the startup time allowed for each additional node in the NodeSet, when
the StroomCluster PodManagementPolicy is Parallel. This gives nodes time to wait on the Stroom bootstrap lock,
which is held while another node upgrades the database.
30 Minimum: 0

StroomClusterSpec

StroomClusterSpec defines the desired state of StroomCluster

Appears in: StroomCluster

Field Description Default Validation
image Image Required
imagePullPolicy PullPolicy
databaseServerRef DatabaseServerRef DatabaseServerRef contains either the details of a DatabaseServer resource, or the TCP connection details of
an external MySQL database
Required
appDatabaseName string Name of the main Stroom database, usually stroom stroom Pattern: ^[A-Za-z0-9_]{1,64}$
statsDatabaseName string Name of the statistics database, usually stats stats Pattern: ^[A-Za-z0-9_]{1,64}$
configMapRef ConfigMapRef Override the Stroom configuration provided to each node, by providing the name of an existing ConfigMap
in the same namespace as the StroomCluster
openId OpenIdConfiguration Configures OpenID to enable operator components to query the Stroom API
https HttpsSettings HTTPS settings. Omit to use plain-text (HTTP) Optional
ingress IngressSettings Required
podManagementPolicy PodManagementPolicyType Pod management policy to use when deploying or scaling the StroomCluster
nodeTerminationPeriodSecs integer Amount of time granted to nodes to drain their active tasks before being terminated 60
volumeClaimDeletePolicy VolumeClaimDeletePolicy Delete Stroom node PersistentVolumeClaims in accordance with this policy Enum: [DeleteOnScaledownAndClusterDeletion DeleteOnScaledownOnly]
nodeSets NodeSet array Each NodeSet is a functional grouping of Stroom nodes with a particular role, within the cluster.
It is recommended two NodeSets should be provided: one for storing and processing data and a separate one for
serving the Stroom front-end.
MinItems: 1
extraJvmOpts string array Additional Java Virtual Machine (JVM) options to use. Example of a valid entry: -Xms1g
extraEnv EnvVar array Additional environment variables provided to each NodeSet pod
extraVolumes Volume array Additional volumes to be mounted in each NodeSet pod
extraVolumeMounts VolumeMount array
logSender LogSenderSettings Configures the mechanism that posts internal audit and logging to Stroom

StroomClusterStatus

StroomClusterStatus defines the observed state of StroomCluster

Appears in: StroomCluster

Field Description Default Validation
nodes string array

StroomTaskAutoscalerSpec

StroomTaskAutoscalerSpec defines the desired state of StroomTaskAutoscaler

Appears in: StroomTaskAutoscaler

Field Description Default Validation
stroomClusterRef ResourceRef The target StroomCluster to apply autoscaling to
taskName string Name of the Stroom node task to auto-scale. Usually "Data Processor". Required
adjustmentIntervalMins integer How often (in minutes) adjustments are made to the number of Stroom node tasks 1 Minimum: 1
metricsSlidingWindowMins integer Sliding window (in minutes) over which to calculate CPU usage vs. the threshold parameters 1 Minimum: 1
minCpuPercent integer Minimum CPU usage threshold before the number of tasks is adjusted upwards 50
maxCpuPercent integer Maximum CPU usage threshold before the number of tasks is adjusted downwards 90
minTaskLimit integer Minimum number of tasks auto-scaler may set the node limit to 1
maxTaskLimit integer Maximum number of tasks auto-scaler may set the node limit to 20
stepAmount integer Number of tasks to add/subtract each adjustment interval, based on usage 1 Minimum: 1

StroomTaskAutoscalerStatus

StroomTaskAutoscalerStatus defines the observed state of StroomTaskAutoscaler

Appears in: StroomTaskAutoscaler

TlsSettings

Appears in: LogSenderSettings

Field Description Default Validation
secretName string Name of the TLS Secret containing a CA certificate (ca.crt) and client certificate/key pair (tls.crt and tls.key)

VolumeClaimDeletePolicy

Underlying type: string

VolumeClaimDeletePolicy determines when Stroom node PersistentVolumeClaims are deleted. If unset, they're never deleted by the operator.

Validation: Enum: [DeleteOnScaledownAndClusterDeletion DeleteOnScaledownOnly]

Appears in: StroomClusterSpec

Value Description
DeleteOnScaledownAndClusterDeletion DeleteOnScaledownAndClusterDeletionPolicy deletes a node's PersistentVolumeClaim when its NodeSet is scaled
down, and every node's when the StroomCluster is deleted
DeleteOnScaledownOnly DeleteOnScaledownOnlyPolicy deletes a node's PersistentVolumeClaim when its NodeSet is scaled down