Configuration Guide

10.3 HotFix 2

Back Next

Step 2. Configure a Process Server as a Solr Server

In a multiple-host deployment, after you configure a Process Server as the ZooKeeper and Solr servers, you can configure other Process Servers to function as the Solr servers.

In the Hub Console of a node, start the Process Server tool.

To acquire a write lock, on the

Write Lock

menu, click

Acquire Lock

In the right pane of the Process Server tool, click the

Add Process Server

button.

The

Add/Edit Process Server

dialog box appears.

Set the following properties of a Process Server for search:

Property	Description
Server	IP address or the fully qualified host name of the application server on which you want to deploy the Process Server. The name is case sensitive. Do not use localhost as the host name.
Port	Port of the application server on which you want to deploy the Process Server. If you want to use the HTTPS protocol to communicate with the Process Server, specify the HTTPS port of the application server.
Enable Search Processing	Indicates whether to use the Process Server to perform search. Select the check box to enable search, or clear the check box to disable search. Informatica recommends to enable search on a single Process Server.
Enable Secured Connection (HTTPS)	Indicates whether to use the HTTPS protocol to communicate with the Process Server. Select the check box to use the HTTPS protocol, or clear the check box to use the HTTP protocol. If you select to use the HTTPS protocol, you must specify the HTTPS port of the application server.

Do not enable ZooKeeper on more than one Process Server.

Click

Click the

Save

button.

Use a text editor to open the following file:

<MDM Hub Installation Directory>\hub\cleanse\resources\cmxcleanse.properties

Configure the following properties for search:

cmx.ss.enabled: Indicates whether to enable search. In a new installation, the default is
true
. When upgrading, if this property is set, the value remains set to the pre-upgrade value. If this property is not set, the default is
false
.; JBoss 6.4.0 only. When you enable search in an environment that uses JBoss 6.4.0, you must set
cmx.server.match.file_load
to
false
. This setting forces the Process Server to use the JDBC uploader instead of native database utilities for matches.
solr.allowAdminConsole: Manually add the property. Indicates whether you can access the Solr Administration User Interface (Admin UI). Use the Solr Admin UI to view the configuration details, run queries, and analyze search data. Set to true to enable access to the Solr Admin UI. Set to false or remove the property to disable access to the Solr Admin UI.
zookeeper.tickTime: Optional. Length of a single tick, in milliseconds. ZooKeeper uses tick as the basic time unit to regulate timeouts. Default is 2000. Increase the time based on your requirement and the performance of the machine on which ZooKeeper is enabled.
search.provisioning.maxshards: Optional. Maximum number of shards to create for each Solr collection on a single node when you index the search data. For optimal performance, create one shard for each collection on each node. Default is 1.
search.provisioning.numshards: Optional. Number of shards to create on your Solr environment. The value depends on the maximum number of shards and the total number of nodes. For example, if the maximum number of shards is 1 and the number of nodes is 3, you can create 3 shards. Default is the total number of Process Servers on which you enable search.
search.provisioning.numreplicas: Optional. Number of copies of the search engine (Solr or Elasticsearch) documents that you want to create on different nodes. Use the replication factor to create multiple copies of the documents in the shards of different nodes. You require multiple copies of the documents to achieve high availability if one or more nodes shut down unexpectedly. For example, if the replication factor is 2, you get two copies of the documents in two nodes. For Solr, the default is 1. For Elasticsearch, the default is 0.
facet.infacount: Optional. Applicable to the Solr search engine only. Number of search results for each shard that the search functionality uses to facet the results. The search functionality uses the results based on the rank of the results in the descending order. For example, if you have 3 shards and specify 2000 as the count, the search functionality facets 6000 search results. If you specify 0 or do not specify any value, the search functionality facets all the results. Default is 1000.
MAX_INITIAL_RESULT_SIZE_TO_CONSIDER: Optional. Manually add the property. Total number of search results to display in the
Data Director
application. The recommended maximum value is 250. Default is 130. Any value higher than 130 affects the performance of the
Data Director
application.
ssl.keyStore: Required if you use the HTTPS port of the application server to configure the Process Server. Manually add the property. Absolute path and file name of the keystore file.
ssl.keyStore.password: Required if you use the HTTPS port of the application server to configure the Process Server. Manually add the property. Plain text password for the keystore file.
ssl.trustStore: Required if you use the HTTPS port of the application server to configure the Process Server. Manually add the property. Absolute path and file name of the truststore file.
ssl.trustStore.password: Required if you use the HTTPS port of the application server to configure the Process Server. Manually add the property. Plain text password for the truststore file.
cmx.websphere.security.ssl.config.url: Required if you use the HTTPS port of the application server to configure the Process Server. For WebSphere only. Manually add the property. Absolute path of the
ssl.client.props
file with the file name.
pingSolrOnStartup: Required if you use WebSphere and not applicable for JBoss and WebLogic. Must be added manually. Indicates whether to create Solr collections when you start the application server after you configure search. Set to true to create collections.
mdm.smartsearch.cache.ttl: Optional. Manually add the property. Number of milliseconds for the cached search results of a Search Business Entity web service request to survive before the cached results expire. Default is 60000.