次の方法で共有


Search Health Reports (SRx) - Digging in further with PowerShell

After introducing the Search Health Reports (SRx) last year, we continued to extend the battery of PowerShell tests for analyzing and troubleshooting a SharePoint 2013 and SharePoint 2016 on-premises search farm. Because most of these efforts resulted in new or improved tests, we largely suggested and recommended the RunAllTests report and Indexer Disk report to leverage the SRx in the vast majority of troubleshooting scenarios. However, sometimes you need (or simply want) to dig in a bit further and expose more detail about the Search system.

In this post, I want to share some of the most common building blocks - the lower level custom functionality - used within the tests and help uncover the richness built into the SRx. I encourage you to [download the SRx Core,] initialize the shell by running the .\initSRx.ps1 script, and try out these in your environment... and let us know what you think.

 #Options when initializing SRx 
 .\initSRx.ps1 -SSA "Name-of-an-SSA"    # > For handling multiple SSAs in a farm
 .\initSRx.ps1 -Verbose                 # > Enables verbose logging in the console
 .\initSRx.ps1 -RebuildSRx              # > Re-initializes the SRx shell, which
                                        #   rebuilds the $SRxEnv and $xSSA objects
                                        #   (e.g. useful after a topology change)
 #Run all tests with detailed output: 
New-SRxReport -RunAllTests -Details 

#Run a specific test (in this case, "OSPingSearchServer") 
New-SRxReport -Test OSPingSearchServer  
    #Hint: After the  -Test parameter, use <tab> key to iterate 
    #      through each test names or auto-complete a test name

#New-SRxReport acts as a wrapper for Test-SRx, which invokes a 
#PowerShell script with a corresponding name. For example: 
Test-SRx -Name OSPingSearchServer 
#Invokes: <SRxPath>\lib\tests\core\Test-OSPingSearchServer.ps1  

    # > Test-SRx returns a "standardized" object, which can be 
    #   pipelined to other custom functionality 
    # > New-SRxReport provides standardized formatting of the 
    #   output object from the test 

#To run all tests and generate an array of test result objects: 
Test-SRx -RunAllTests 
 #Detailed indexer reports...
Get-SRxIndexReports -DiskReport
 #The extended and customized SSA Object ($xSSA)
$xSSA | gm | ? {$_.Name -like "_*"}    #In SRx, custom properties and methods appended 
                                       #to any out-of-the-box objects are named with an
                                       #underscore "_" as their prefix to differentiate
 #To get Search Servers details...  
 $xSSA._Servers                        # > An array of all search servers discovered
                                       #   during initialization of SRx, where each
                                       #   server is represented as a custom object
$xSSA._GetServer("-Name-of-Server-")   # > Gets a specific server by name
$xSSA._GetServerEx("-Name-of-Server-") # > Similar to above, but returns an extended
                                       #   server object by fetching the applicable
                                       #   registry keys and system properties (from
                                       #   Get-WMI for various classes)

#-------------------------------- 
#Example of tools available for a specific server... 
 $server = $xSSA._GetServer("-Name-of-Server")  
 $server.canPing()           # > Wrapper: Test-Connection <severname>
 $server.GetProcesses()      # > Under the covers, runs "Get-Process" with a list of 
                             #   applicable Search processes (e.g. noderunner, mssearch)
 
 $server = $xSSA._GetServerEx("-Name-of-Server")  
 $server | gm                #View the other extended properties and methods for this
 #Topology Visualization...
 $xSSA._ShowTopologyReport()
 #Component fun...
 $xSSA._GetCC()       #Get the list of Crawl Components 
 $xSSA._GetCPC()      #Get the list of Content Processing Components 
 $xSSA._GetAPC()      #Get the list of Analytics Processing Components 
 $xSSA._GetQPC()      #Get the list of Query Processing Components 
 $xSSA._GetIndexer()  #Get the list of Index Components     

#Assuming you have an component named "CrawlComponent2" or "IndexComponent4", run: 
 $xSSA._GetCC(2)      #Gets "CrawlComponent2" from the Active topology
 $xSSA._GetIndexer(4) #Gets "IndexComponent4" from the Active topology

#Each component object also has extended methods and properties... 
 $i4 = $xSSA._GetIndexer(4) 
 $i4._GetProcess()                 # > Wrapper: Get-Process noderunner -computer <name>
 $i4._GetHealthReport()            # > Wrapper: Get-SPEnterpriseSearchStatus -SSA <SSA>
                                   #              -Component IndexComponent4 -HealthReport
 $i4._BuildDiskReportData()        # > Runs underlying processing for an index disk report
 $i4._CellPath                     # > An extended property with the index cell path
 $i4 | gm | ? {$_.Name -like "_*"} # > View the list of properties and methods for this 
 #To get more Content Sources details, which aggregates data from the following:
#  - The out-of-the-box content source object
#  - MSSCrawlHistory
#  - MSSCrawlComponentsState
#  - Web Application info [if in the same farm as the SSA]

 $xSSA._GetContentSource()       # > Wrapper: Get-SPEnterpriseSearchCrawlContentSource
 $xSSA._GetContentSourceEx()     # > Returns an extended content source object
                                 #   Hint: To view the extended properties, pipe the 
                                 #         output object to: | SELECT *
 $xSSA._GetContentSourceReport() #Equivalent to: $xSSA._GetContentSourceEx() | SELECT *

#Get a specific content source (either by Content Source ID or by name) 
 $cs1 = $xSSA._GetContentSource("Local SharePoint Sites") 
#And this is the equivalent (assuming "Local SharePoint Sites" has an ID of 1) 
 $cs1 = $xSSA._GetContentSource(1)

#Working with a specific content source 
 $xCS1 = $xSSA._GetContentSourceEx(1)
 $xCS1.StartAddresses.AbsoluteUri
 $xCS1.StartIncrementalCrawl()
 $xCS1._CrawlStatusDetailed

#Note: Fulfilling the extended properties can incur higher overhead for environments 
#      with many content sources. Therefore, when running GetContentSourceEx for all
#      content sources (e.g. by not specifying a parameter value () or by using "*"), 
#      this method implements an in-memory caching mechanism for 20 minutes. 
#       - However, the cache is bypassed when requesting a specific content source 
 #Crawl Visualization...
 $xSSA._ShowRecentCrawlVisualization(2,1)  #Displays a visualization of crawl activity
                                           #for the last 2 hours where each character
                                           #represents a 1 minute block of time

 

Other key blog posts about this tool:

----------------------

I hope this helps...