3.1.4.40 StartCrawl

  • Article

This operation is used to start a crawl of a content source of the crawler application.

 <wsdl:operation name="StartCrawl">
   <wsdl:input wsam:Action="http://tempuri.org/ISearchApplicationAdminWebService/StartCrawl" message="tns:ISearchApplicationAdminWebService_StartCrawl_InputMessage"/>
   <wsdl:output wsam:Action="http://tempuri.org/ISearchApplicationAdminWebService/StartCrawlResponse" message="tns:ISearchApplicationAdminWebService_StartCrawl_OutputMessage"/>
 </wsdl:operation>

The protocol client sends an ISearchApplicationAdminWebService_StartCrawl_InputMessage request message and the protocol server responds with an ISearchApplicationAdminWebService_StartCrawl_OutputMessage response message, as follows:

  • On receipt the protocol server validates the versionIn value from the request message. If it doesn’t match the current configuration version of the crawler application, the protocol server MUST throw a FaultException<ExceptionDetail> message.

  • If there is no content source with the identifier specified in the contentSource element of the request message, the protocol server MUST throw a FaultException<ExceptionDetail> message.

  • If a full crawl is specified in the request message’s type element, the protocol server MUST attempt to start a full crawl for the specified content source.

  • If an incremental crawl is specified in the request message’s type element, and the specified content source has not been crawled before, or the last crawl of the content source was stopped before it completed, the protocol server MUST attempt to start a full crawl for the content source.

  • If an incremental crawl is specified in the request message’s type element, and the specified content source has been crawled before, and the last crawl was not stopped before it completed, the protocol server MUST attempt to start an incremental crawl for the content source.

  • If the crawl is in a state where it cannot be started or is already started, the protocol server MUST ignore the start request and return the current crawl status in the StartCrawlResult element the response message.

  • On success, the protocol server MUST return the current crawl status in the StartCrawlResult element of the response message.

  • For all other errors, the protocol server MUST send a FaultException<ExceptionDetail> message to the protocol client.