Skip to content

Latest commit

 

History

History
52 lines (42 loc) · 6.51 KB

configuration.md

File metadata and controls

52 lines (42 loc) · 6.51 KB

Lighter configuration

Lighter can be configured by using environment variables. Currently, Lighter supports only one of backends on single deployment: Yarn or Kubernetes.

Global properties

Property Description Default
LIGHTER_MAX_RUNNING_JOBS Max running Batch jobs in parallel 5
LIGHTER_MAX_STARTING_JOBS Max starting Batch jobs in parallel 5
LIGHTER_SPARK_HISTORY_SERVER_URL Spark history server URL used on the Lighter UI http://localhost/spark-history/
LIGHTER_EXTERNAL_LOGS_URL_TEMPLATE Template for link to external logs (Grafana, Graylog, etc.) used on the Lighter UI. Allowed placeholders: {{id}}, {{appId}}, {{createdTs}}
LIGHTER_PY_GATEWAY_PORT Port for live Spark session communication 25333
LIGHTER_URL URL which can be used to access Lighter form Spark Job http://lighter.spark:8080
LIGHTER_SESSION_TIMEOUT_INTERVAL java.time.Duration representing session lifetime (from last statement creation). Use 0m value to disable 90m
LIGHTER_SESSION_TIMEOUT_ACTIVE Should Lighter kill sessions with waiting statements (obsolete when LIGHTER_SESSION_TIMEOUT_INTERVAL is 0m) false
LIGHTER_SESSION_SCHEDULE_INTERVAL java.time.Duration representing the interval at which a task is triggered to initiate scheduled sessions 1m
LIGHTER_SESSION_TRACK_RUNNING_INTERVAL java.time.Duration representing the interval at which a task is triggered to process and update running session state 2m
LIGHTER_STORAGE_JDBC_URL JDBC url for lighter storage jdbc:h2:mem:lighter
LIGHTER_STORAGE_JDBC_USERNAME JDBC username sa
LIGHTER_STORAGE_JDBC_PASSWORD JDBC password
LIGHTER_STORAGE_JDBC_DRIVER_CLASS_NAME JDBC driver class name org.h2.Driver
LIGHTER_BATCH_DEFAULT_CONF Default conf props for batch applications (JSON)*
LIGHTER_SESSION_DEFAULT_CONF Default conf props for session applications (JSON)
LIGHTER_SESSION_PERMANENT_SESSIONS List of configurations for permanent sessions "[]"

* default configs will be merged with configss provided in submit request, if property is defined in submit request, default will be ignored. Example of LIGHTER_BATCH_DEFAULT_CONF: {"spark.kubernetes.driverEnv.TEST1":"test1"}.

Kubernetes configuration

Property Description Default
LIGHTER_KUBERNETES_ENABLED Kubernetes enabled false
LIGHTER_KUBERNETES_MASTER Kubernetes master URL k8s://kubernetes.default.svc.cluster.local:443
LIGHTER_KUBERNETES_NAMESPACE Kubernetes namespace spark
LIGHTER_KUBERNETES_MAX_LOG_SIZE Max lines of log to store on DB 500
LIGHTER_KUBERNETES_SERVICE_ACCOUNT Kubernetes service account spark

YARN configuration

Property Description Default
LIGHTER_YARN_ENABLED Yarn enabled (Kubernetes should be disabled) false
LIGHTER_YARN_URL Yarn API URL, /ws/v1/cluster/ will be appended
HADOOP_CONF_DIR Path to core-site.xml,hdfs-site.xml and yarn-site.xml files
LIGHTER_YARN_KERBEROS_PRINCIPAL Kerberos principal used for job management*
LIGHTER_YARN_KERBEROS_KEYTAB Kerberos keytab used for job management

* Principal & Keytab provided in LIGHTER_YARN_KERBEROS_PRINCIPAL and LIGHTER_YARN_KERBEROS_KEYTAB will be used by spark job as well, if spark.kerberos.keytab is not explicitly declared in LIGHTER_BATCH_DEFAULT_CONF or provided on submit request.