зеркало из
1
0
Форкнуть 0
azure-quickstart-templates/elasticsearch
Trent Swanson 39b8b349f4 common cluster defaults for a large cluster 2015-04-28 17:17:24 -07:00
..
README.md finishing lb fixes for changes 2015-04-24 11:29:45 -07:00
azuredeploy-parameters.json general cleanup 2015-04-28 17:14:42 -07:00
azuredeploy.json common cluster defaults for a large cluster 2015-04-28 17:17:24 -07:00
client-nodes-resources.json general cleanup 2015-04-28 17:14:42 -07:00
data-nodes-2disk-resources.json general cleanup 2015-04-28 17:14:42 -07:00
data-nodes-4disk-resources.json general cleanup 2015-04-28 17:14:42 -07:00
data-nodes-8disk-resources.json general cleanup 2015-04-28 17:14:42 -07:00
data-nodes-16disk-resources.json general cleanup 2015-04-28 17:14:42 -07:00
elasticsearch-ubuntu-install.sh added child templates 2015-04-24 07:40:09 -07:00
empty-resources.json empty resource with client parameters 2015-04-25 22:40:00 -07:00
jumpbox-resources.json general cleanup 2015-04-28 17:14:42 -07:00
master-nodes-resources.json general cleanup 2015-04-28 17:14:42 -07:00
metadata.json updated metadata date and test parameters 2015-04-25 23:01:19 -07:00
shared-resources.json general cleanup 2015-04-28 17:14:42 -07:00

README.md

Install Elasticsearch cluster on Virtual Machines with data node storage scale units

This template deploys an Elasticsearch cluster on Virtual Machines and uses template linking to create data node scale untis. The template provisions 3 dedicated master and client nodes in separate availability sets and storage accounts. Data node scale unit count and nodes in a scale unit can be configured as parameters. A load balancer is configured with a rule to route traffic on port 9200 to the client nodes, and also includes a single SSH management rule mapped to one of the client nodes. Multiple data disks are attached (the number depends on the size of the virtual machine selected) and data is striped across them as an approach to improve disk throughput.

This template also deploys a Storage Account, Virtual Network, Availability Sets, Public IP addresses, Load Balancer, and a Network Interface.

Below are the parameters that the template expects

Name Description
storageAccountPrefix Unique DNS Name for the Storage Account and the template will use this to create at storage account for each data node scale unit. Keep this short - scale unit and vm are appended this
adminUsername Username for the Virtual Machines
adminPassword Password for the Virtual Machine
dnsNameforLBIP Unique DNS Name for the Public IP used for load balancer.
loadBalancerType Type of load balancer to place the client nodes on (internal/external)
region region where the resources will be deployed
virtualNetworkName Name of Virtual Network
OS The OS platform to deploy elasticsearch on
jumpbox Add a management box to the deployment
vmSizeMasterNodes Size of the Cluster Master Virtual Machine Instances
vmSizeClientNodes Size of the Cluster Data Virtual Machine Instances
vmSizeDataNodes Size of the Cluster Data Virtual Machine Instances (This will also affect number of data disks)
dataNodeScaleUnits Number of Elasticsearch data scale units which include a configurable number of nodes and a storage account
dataNodesPerUnit Number of Elasticsearch data nodes to provision with each scale unit
esClusterName Name of the Elasticsearch cluster (elasticsearch)
esVersion Elasticsearch version to deploy (1.5.0)
dataDiskSize The size of each data disk attached in Gb (default 100GB)

##Notes Security Warning! The configuration allows you to enabled external load balanced endpoints on a public IP. The endpoint is not secure and it's recommended that you keep these endpoints internal or secure them. Elasticsearch Sheild product should be considered.