Chapter 17 – understanding Sharepoint 2010 Search

Wrox Real World Sharepoint 2010 (p. 605)

FAST Search server features:
Thumbnails and previews of documents
View in Browser
Visual best Bets
Deep refiners with counts
User Context from User Profile
Sorting on any property
Similar search
Broader, better language support
Richer query language

Deploying Sharepoint 2010 Search —
The search service application can be created in Manage Service Application.

Trouble shooting:

Problem: Sharepoint Search won’t start:

and Windows log:

System log:

The error was caused by lack of permission in the account that runs the Search service.

Solution:
1. All Programs
Administrative Tools Component Service.

2. Right click Osearch14.

Right click Identity tab, and enter a user with enough permission.

Also, make sure the the user account/password in Managed Account is correct. If outdated, use this to refresh/enter the current cknown password:
Central Administration
Security Configure Managed Account Click on the account that you want ot manage, and enter the curren tpassword (won’t send to AD) (p.613)

Deploying a New Search Service Application —
1. Create a new Search Service Application from Central Admin
à Manage Service Application
2. Central Admin
à Application Management à
Configure service Application Associations.
3. Click a web application to set the Service Application/Web Application associations.

4. Change the group down to [custom] (if you don’t like the default) and check the service applications that you want to associate with your current web application.

5. In the end each web application has their own service application group.

Crawling and Indexing —
Crawler – the component that browses the contents automatically on a regular basis to provide up-to-date data for indexers.
Indexer – collecting and storing relevant data of crawled contents, in order to make available fast and precise information retrieval during the queries.
Query – The query component provides the UI for entering the user queries, and presents the result set to end users. It communicate with indexer directly to get the results of query.

Configuring Crawl/Index/Query

Crawling —

Content Sources – defined additional location to crawl/index, such as other Sharepoint site, external site, file shares, LOB data etc. Also Full/Incremental crawl and its schedule.
Crawl Rules – Include or exclude crawl path/URL. Or set authentication account for a specific URL.
Crawl Log – For each content source, display crawl log on screen. Succeed/Warning/Errors.
Add Server Name Mapping – if the URL in the indexer is different from the public URL, use this feature to replace the pattern.
Host Distribution Rules – If you have more than one crawl database, use this to associate a host with a specific database.
File Types – the file types that are indexed by Sharepoint 2010.
Index Reset – reset index – rebuild from scratch.
Crawler Impact Rules – define how often you should crawl a site —
(1) X requests max. at the same time, or
(2) 1 request at a time, and interval X seconds.

Responsibilities of the Indexer —

(1) Processing the crawls
(2) Making the indices available to query servers
(3) Managing content source properties such as location and scheduling
(4) Creating and maintaining an index database

Queries and Results —

Authoritative pages – Enter URL of Authoritative pages and Non-Authoritative page for ranking purpose.
Federated Location – usually remote search engines that provide results to the queries initiated in Sharepoint 2010, but also can be local if you want to run simultaneous searches on the same contents.
Metadata Properties – properties of crawled content that user can use in the search.
Scopes – define different search scope.
e.g., for “People Search”, the scope rule/definition is:

for “all sites search”, the rule/definition is:

Search result removal – One big text box. Enter the URL that you would like to remove. Crawl rule will be automatically created to exclude these URL’s.

Reports —

Administrative Report —
Web Analytics Report – Similar to Webtrends (P.617)

Deploying Fast Search Server —
FAST can be installed on a standalone server, but normally install on one or more different servers. When you deploy FAST on two or more servers, one must be deployed as a FAST ADMIN SERVER. , where administrative services are installed and run.

First you need a FAST Search Admin domain user., who will perform admin work on FAST servers. Must be a member of FASTSearchAdministrator group, as well as DBCreator SQL role. Then start instll wizard.

Two kinds of FAST search service applications —
(1) FAST search content
service application —
(2) FAST query service application —

FAST Search Content Service Application – (FAST Crawler!!!!!!!)
Crawls content and feeds into the FAST search server 2010 .

FAST Query Service Application —
Receives query and provides query from the content that is crawled by FAST search content service application.

again, here we’re re-visiting some of the important search features —

Defining Content Sources —
1. Local & Remote Sharepoint sites
2. Web Sites
3. File Shares
4. Exchange public folders
5. Business data (LOB).

Using Scopes —
Contents from all sources must be organized into Search Scopes to refine the queries and allow access to the content from the search UI.
Define rule:

Scheduling Crawls —

Full crawls – whole content crawled
Incremental crawls – only send new content created since the last crawl.

Using Search Federation —
advantage —
Remote sites that block sharepoint crawler in their robots.txt files can also be crawled by a federated search engine.
Include specific keywords and pattern in the query
content that changes very often can immediately be crawled.
You can query content under a different security context
You don’t have to crawl infrequently queries content.

Federated Location connections
can be saved in .FLD (Federated File Location) and you can import the file to your search service application. All search engines that are compatible with OpenSearch 1.0/1.1
can be used. Once done, simply insert a Federated Results Web Part. If your target site does not support OpenSearch 1.0/1.1, you still can use a third part site that can search your target site, such as BING.com.

Understanding Keywords and Best Bets —

With keywords, you can mark specific items as more relevant in your Sharepoint search results. For example, if you have a financial, the keywords can be “money”, “stock” and “finance”.

Click “Add Keyword” to add a keyword, and define its synonyms, and add best bets (URL).

User Content in FAST Search —
user context (what division you’re in, where do you live etc.) can be defined in FAST and can be associated with various search settings, including Best Bets.

Using People Search —
To enable People Search, you must crawl for people by modifying your “Local Sharepoint Sites” content source, or by creating a new one – sps3://SP2010

Customizing User Interface —
Two primary kinds of Search UIs —
(1) Centralized (dedicated) – a centralized UI is a site where the only function is Search. Typically, this is Search Center template.
(2) De-Centralized — E.g., create a Document site and insert a search box etc.

Using Search Centers —

In Sharepoint 2010, 3 Search Centers are available as site template.
(1) Basic Search Center
— search; advanced search; page result.
(2) Enterprise Search Center
– welcome page with two tabs – general search & people search. Search result; advanced search. You can add new pages and new tabs on the welcome page.
(3) FAST Search Center
– similar to the above.

***** MANY MANY Search Web Parts Available for USE

Customizing the Refinement Panel —
Sharepoint 2010 (and FAST) Search’s Result Page has
Refinement Panel. And can be insert to any page because it’s a web part.

Integrating the Client —

If you have WorkSpace, you can use client side to search list even though it’s offline.
If you have Windows 7, you can search Sharepoint from Windows Explorer. (on Windows 7, define the Sharepoint site as a federated search connector)

Using Managed Metadata – 02/03/2011, 11:18PM


Advertisements
Post a comment or leave a trackback: Trackback URL.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: