Multi-cluster

Home

Multi-cluster

This document describes the properties that are handled in a multi-cluster deployment.

Layout of multi-cluster configurations for a component:

com.unraveldata.<component>.list = Comma-delimited list of variables representing the component instances.
com.unraveldata.<component>.<variable>.<property> = Value
The variables are for configuration purposes only and do not need to correspond to any hostnames or other properties of the components. Acceptable characters: alphanumerics, hyphens, and underscores only.

Core node configurations for components

The following components are accessed from the Core node, where Unravel is installed.

Cloudera Manager (CM)
Ambari
Hive Metastore
Kafka
Pipeline (Workflows)

The following properties must be configured on the Core node in a multi-cluster setup.

Cloudera Manager - Multi-cluster

Property/Description	Set by user	Unit	Default
com.unraveldata.cloudera.manager.list Comma-delimited list of variables to designate one or more CM instances.	Required	CSL	-
com.unraveldata.cloudera.manager.X.url Cloudera URL for variable `X`.	Required	URL	-
com.unraveldata.cloudera.manager.X.username Cloudera user name for variable `X`	Required	string	-
com.unraveldata.cloudera.manager.X.password Cloudera password for variable `X`	Required	string	-
com.unraveldata.cloudera.manager.X.unravel.cluster.ids Cloudera manager UUID and display name.	Optional	CSL

Ambari

Property/Description	Set by user	Unit	Default
com.unraveldata.ambari.manager.list Comma-delimited list of variables to designate one or more Ambari instances. This is used for configuration purposes only and does not need to correspond to Ambari hostnames or cluster ID/name.	Required	CSL	-
com.unraveldata.ambari.X.url Ambari URL for variable `X`.	Required	URL	-
com.unraveldata.ambari.X.username Ambari user name for variable `X`	Required	string	-
com.unraveldata.ambari.X.password Ambari password for variable `X`	Required	string	-

Examples

com.unraveldata.ambari.manager.list=prod,stage
com.unraveldata.ambari.manager.prod.url=https://unravel-prod:8443
com.unraveldata.ambari.manager.prod.username=admin
com.unraveldata.ambari.manager.prod.password=ENC(a3b20c0da)
com.unraveldata.ambari.manager.stage.url=http://unravel-stage:8080
com.unraveldata.ambari.manager.stage.username=admin
com.unraveldata.ambari.manager.stage.password=ENC(92f48a8cf)

Hadoop

You must set the following property when you install the core node on a server where there is no Hadoop configuration:

Property/Description	Set by user	Unit	Default
unravel.hadoop_module.platform Set this property when you install the core node on a server where there is no Hadoop configuration.	Required		unravel

Hive Metastore

Property/Description	Set by user	Unit	Default
com.unraveldata.hive.metastore.list Comma-delimited list of variables to designate one or more Hive metastore instances.	Required	CSL	-
com.unraveldata.cloudera.manager.X.unravel.cluster.ids Replace `X` in the property with one of the metastore variables listed in com.unraveldata.hive.metastore.list The value is a comma-delimited list of Unravel cluster IDs, and in the case of Impala, the Cloudera Manager (CM) cluster IDs. Refer to Obtaining CM cluster IDs	Required	CSL	-
hive.metastore.X.cluster.ids Replace `X` in the property with one of the metastore variables listed in com.unraveldata.hive.metastore.list. The value is a comma-delimited list of Unravel cluster IDs.	Required	CSL	-
javax.jdo.option.X.ConnectionURL Hive metastore URL for variable `X` listed in com.unraveldata.hive.metastore.list	Required	URL	-
javax.jdo.option.X.ConnectionDriverName Hive metastore JDBC driver name for variable `X` listed in com.unraveldata.hive.metastore.list	Required	string	-
javax.jdo.option.X.ConnectionUserName Hive metastore JDBC user name for variable `X` listed in com.unraveldata.hive.metastore.list	Required	string	-
javax.jdo.option.X.ConnectionPassword Hive metastore JDBC password for variable `X` listed in com.unraveldata.hive.metastore.list	Required	string	-

The cluster IDs of Cloudera Manager could be obtained by running the following curl command on the node with access to CM:

curl -u <CM_USER>:<CM_PASSWORD> <CM_HOST>:<PORT>/API/<API_VERSION>/clusters

This would return a list of properties. The UUID field denotes the CM cluster ID. For example:

> curl -u admin:admin "http://123.45.6.789:7180/api/v17/clusters"

{
"items" : [ {
"name" : "cluster",
"displayName" : "Sample Cluster",
"version" : "CDH5",
"fullVersion" : "5.12.2",
"maintenanceMode" : false,
"maintenanceOwners" : [ ],
"clusterUrl" : "http://xyz.unraveldata.com:7180/cmf/clusterRedirect/cluster",
"hostsUrl" : "http://xyz.unraveldata.com:7180/cmf/clusterRedirect/cluster/hosts",
"entityStatus" : "CONCERNING_HEALTH",
"uuid" : "94562003-ery8-4418-914c-f5738fc1133d"
} ]
}

Kafka

Property/Description	Set by user	Unit	Default
com.unraveldata.ext.kafka.clusters.list Comma-delimited list of variables to designate one or more Kafka instances.	Required	CSL	-
com.unraveldata.ext.kafka.X.bootstrap_servers Comma-separated list of Kafka brokers for that cluster.	Required	CSL	-
com.unraveldata.ext.kafka.<server>.jmx_servers Comma-separated list of JMX servers	Required	CSL	-
com.unraveldata.ext.kafka.<server>.jmx.kafka1.host Host name of the JMX server.	Required	CSL
com.unraveldata.ext.kafka.<server>.jmx.kafka1.port Port of JMX server.	Required	CSL	-
com.unraveldata.ext.kafka.<server>.jmx.kafka1.run_period Collection interval for JMX metrics.	Required	string	-

Pipelines

Property/Description	Unit	Default
com.unraveldata.workflow.num.past.instances The number of past workflow instance for the compare operation.	count	5
com.unraveldata.workflow.num.future.instances The number of future workflow instance for the compare operation.	count	5
com.unraveldata.analytics.max.past.samples The number of workflows to use during a SLA Analysis.	count	5
com.unravel.workflows.compareupdate.disable In a multi-cluster setup, this property must be set to true mandatorily on the Core node.	boolean

Edge node configurations for components

In a multi-cluster deployment for on-prem platforms, the following properties must be added to the Edge node server, for MR jobs to load jhist and logs, HDFS path for jhist/conf and yarn logs:

Property/Description	Default
com.unraveldata.min.job.duration.for.attempt.log Minimum duration of a successful application or which executor logs are processed (in milliseconds).	600000 (10 mins)
com.unraveldata.job.collector.log.aggregation.base HDFS path to the aggregated container logs (logs to process). Don't include the hdfs:// prefix. The log format defaults to TFile. You can specify multiple logs and log formats (TFile or IndexedFormat). Example: `TFile:/tmp/logs//logs/`, `IndexedFormat:/tmp/logs//logs-ifile/`. For HDP set this to: `IndexedFormat:/app-logs/*/logs/`.	/tmp/logs/*/logs/
com.unraveldata.job.collector.done.log.base HDFS path to "done" directory of MR logs. Do not include the `hdfs:// prefix`. For HDP set this to: `/mr-history/done.`	hdfs:///user/spark/applicationHistory/

Property/Description

Default

com.unraveldata.min.job.duration.for.attempt.log

Minimum duration of a successful application or which executor logs are processed (in milliseconds).

600000

(10 mins)

com.unraveldata.job.collector.log.aggregation.base

HDFS path to the aggregated container logs (logs to process). Don't include the hdfs:// prefix. The log format defaults to TFile. You can specify multiple logs and log formats (TFile or IndexedFormat).

Example: TFile:/tmp/logs/*/logs/, IndexedFormat:/tmp/logs/*/logs-ifile/.

For HDP set this to: IndexedFormat:/app-logs/*/logs/.

/tmp/logs/*/logs/

com.unraveldata.job.collector.done.log.base

HDFS path to "done" directory of MR logs. Do not include the hdfs:// prefix.

For HDP set this to: /mr-history/done.

hdfs:///user/spark/applicationHistory/

The following properties must be added for Tez to the Edge node server in a multi-cluster deployment for on-prem platforms.

In this section:

Would you like to provide feedback? Just click here to suggest edits.

Home

Multi-cluster

Core node configurations for components

Cloudera Manager - Multi-cluster

Ambari

Hadoop

Hive Metastore

Kafka

Pipelines

Edge node configurations for components

Search results