지금 지원 담당자와 채팅
지원 담당자와 채팅

Unified Communications Diagnostics 8.6.2 - Deployment Guide

Installing and upgrading Unified Communications Diagnostics Technical Information for UC Diagnostics for Exchange Technical Information for Office 365 Technical Information for UC Diagnostics for BlackBerry Technical Information for UC Diagnostics for OCS Technical Information for UC Diagnostics for Lync/Skype for Business Troubleshooting

Health Test Issues

The following questions and answers provide information about common health test issues.

Why do some tests, such as the Exchange Internal Server Delivery and the Exchange External Message Delivery test, not work on certain operating system versions?
For the MAPI-based tests to work on Windows Server 2008, Windows 8, Windows 8.1, Windows Server 2012, Windows Server 2012 R2, or Windows Server 2016 you must turn off User Account Controls (UAC) on the computer on which the Diagnostic Services component runs.
Why do Exchange Internal Server Delivery tests show results such as Not Complete or Complete with a Timeout Failure message?
If your environment contains Exchange 2010 (SP1 or later) servers on which Throttling Security Policy is turned on, the default throttling policy limits the number of concurrent connections a user can have to an Exchange 2010 server.
The number of MAPI calls performed during a MAPI-based Delivery test can exceed this limit, causing the Exchange Client Access Server to start rejecting calls.
To resolve this problem, create a new throttling policy that has no limit for simultaneous MAPI calls to Exchange servers. Assign this policy to the user account that is used for test execution. This account is either the Distributed Collector service account (if default credentials are specified for the test) or a user account associated with alternative credentials specified for the test.
To create a new throttling policy and assign it to a user account, logon to the Exchange 2010 Server, open the Exchange Management Shell, and type the following commands:
New-ThrottlingPolicy SOMPolicy
Set-ThrottlingPolicy SOMPolicy -RCAMaxConcurrency $null -RCAPercentTimeInAD $null -RCAPercentTimeInCAS $null -RCAPercentTimeInMailboxRPC $null
Set-Mailbox <user account> -ThrottlingPolicy SOMPolicy
Some health tests failed because Exchange Management Console (EMC) 2007 SP 1 and Update Rollup 1 was not installed. I have installed it now but my test still does not work. Why?
Ensure that you reboot the UC Diagnostics computer or restart the Spotlight Diagnostic Test Engine and Spotlight Distributed Collector services after you install EMC updates.
Why does my Exchange OWA Availability test fail?
Ensure that the test mailbox is accessible from the UC Diagnostics computer.
For Exchange 2007, try running the following PowerShell cmdlet from the Exchange Management Shell:
Test-OwaConnectivity -clientAccessServer "<server name>"
One common problem is that Internet proxy settings can interfere with DNS resolution.
Why would the Exchange OWA Availability Health test fail on the first attempt on an Exchange 2007 server?
For Exchange 2007 servers, the OWA Availability health test fails to connect on the first attempt. You see the following error message: “Failed to retrieve Outlook Web Access connectivity data from Exchange PowerShell. Mailbox logon statistics are not available.”
This error is caused by a bug in the PowerShell cmdlet. The test does connect on the second attempt.
Why does the Exchange Internal OWA test fail due to failed mailbox logon attempts? The test was executing normally until recently.
The password may have expired for a test mailbox used in this test. The user account assigned to the test may not have sufficient permissions to change the password of the test mailbox user account.
Short term solutions (which last until the mailbox password expires again):
•
Run the OWA test using the domain admin account
- OR -
•
Recreate the test mailboxes.
Long term solution:
•
Change the setting of accounts created for the OWA test to “password never expires”.
I am monitoring an Exchange 2007 organization using UC Diagnostics that is hosted on a computer in a different forest. Why does only some functionality work?
Microsoft does not support accessing Exchange 2007 servers through PowerShell from different forests. See the Heath Tests section for more information about which functionality requires PowerShell.
When I attempt to create OWA test mailboxes for Exchange 2010, why do I get the following error message?
CreateTestUser: Mailbox could not be created. Verify that OU ’Users’ exists and that password meets complexity required.
The script "new-testcasconnectivityuser.ps1" that ships with Exchange 2010 tries to create the user accounts in the built-in ’Users’ container. (The variable is defined in the script as "$OrganizationalUnit = "Users".)
If there is another organizational unit named ’Users’ anywhere in the Active Directory structure (e.g. domain.com\branchoffices\users), this error is generated because the Name "Users" would no longer be unique.
To specify the organizational unit in which to create the test users, include the ’-ou’ parameter when running the script. For example:
get-mailboxserver | .\new-TestCasConnectivityUser.ps1 -ou "cn=users,dc=domain,dc=com"
Why do my Exchange Message Delivery Health tests give an error that the test mailbox cannot be found, even after I have run the mailbox creation scripts?
Active Directory replication must occur before all the mailbox information is propagated to the global catalog to which UC Diagnostics is connected.
Why do my Exchange Message Delivery Health tests give an error that the test mailbox is not accessible, even after I have run the mailbox creation scripts?
Ensure that the credentials under which the test is running are the same as the ones specified in the mailbox creation script.
Also, sometimes Windows caches credentials lookups, which can cause access to be denied after permissions are changed. Try restarting the Spotlight Diagnostic Test Engine and Distributed Collector services.
Why do I occasionally get duplicate notifications?
If you have configured a secondary SMTP server and the primary server fails to send some notifications, all notifications are resent via the secondary server. Occasionally, the primary SMTP server tries to send the original notification later. If this occurs, you receive notifications from both servers.
How do I exclude certain event IDs?
1
Create or edit a health test.
2
Select a server and double-click <Add Event Log Entries>.
3
Select an event log type and enter a source name.
4
Enter the event IDs you would like to monitor, leaving out the event IDs you want to exclude. For example, if you want to exclude event ID 100, you would enter 0-99,101-90000.
I am connected to a second Exchange organization but discovery failed. Why?
Your service credentials may not have sufficient rights to discover the second Exchange organization. You can:
1
Stop your services.
2
Set the services log on account to another account which would have permissions to access that organization.
3
Restart your services, then connect to the organization using the new account.
4
After discovery is successful, right-click the Exchange root note and select Properties.
5
Clear Schedule Discovery to run automatically.
6
Set the services log on account to your original account, then restart your services.
7
Reconnect to both Exchange organizations.
Why does the BlackBerry Server Health test show incomplete when run against a BlackBerry Configuration database server?
All SQL-related performance counters for BlackBerry configuration database servers target SQL Server Enterprise version of SQL (unnamed instance).
The test results show a grey icon for the BlackBerry configuration database servers installed on the named instance of the SQL server. Error details specify that the object is unavailable - the performance counter is either uninstalled or disabled.
To monitor BlackBerry configuration database servers that are installed on a named instance of SQL server, change the SQL-related performance counters with corresponding performance counters for the named instance. For example, change the counter SQLServer:Access Methods\Full Scans/sec to MSSQL$MySQLInstance:Access Methods\Full Scans/sec, where MySQLInstance is the name of SQL instance.
If your BlackBerry database name is not the default (BESMgmt), change the instance name of the perfmon counter SQLServer:Databases/Percent Log Used(BesMgmt).
When running the BlackBerry Server Health test against BlackBerry servers installed on 64-bit operating systems, why are BlackBerry Server performance counters reported as not available?
If you are using a BlackBerry Enterprise Server version earlier than 5.0.3, RIM does not support a 64-bit performance provider.
If you are using BlackBerry Enterprise Server version 5.0.3 or later, RIM supports performance counters through a WMI-based provider. Currently, UC Diagnostics supports registry-based performance providers only.
Why do the Message Delivery health test and Queue health test occasionally fail?
Exchange Server MAPI Client and Collaborative Data Objects (MAPI & CDO) can become unstable, causing the Exchange Internal Server Delivery, Exchange External Message Delivery, and BlackBerry Message Delivery Health tests to fail occasionally. The tests appear to be timed-out. If any of these tests fail, the test executes normally on the next attempt.
In some circumstances, the test can finish normally, instead of timing out. However, applications errors maybe reported in the Event log for SpotlightDiagnosticTest.exe application.
There could be as many as one error for each target in the test.
Why do I receive the following error against Exchange 2007 or 2010 mailbox servers?
<Status ServerAddress="127.0.0.1" ErrorMessage="Error code = 0&#xA; Error message = Incorrect function.&#xA; Error source = msxml6.dll&#xA; Error description = Not enough storage is available to complete this operation.&#xA;&#xA;" ManagedObjectId="148" StartDateTime="40365.2604976852"/>
This error is caused by excessive log messages that are generated when detailed logging is turned on in an Exchange Server (usually Exchange 2010, Mailbox role). To eliminate this problem, identify and exclude the informational events that occur often during a monitoring interval.
Perform the following steps:
1
In UC Diagnostics, open the Event Viewer on the Exchange server that reports the error and locate high-frequency informational events that can be filtered out.
2
Right-click the server icon and select Detect/Exchange Server Health.
3
Select either Exchange 2010 or Exchange 2007.
4
In the list, under Event Log, locate the appropriate event sources:
•
“Application\MSExchange*” (for Exchange 2010)
•
“Application\MSExchangeIS*” (for Exchange 2007).
5
Record the field values, such as Thresholds and Overrides.
6
Right-click on this item and select Delete.
7
To replace the deleted item, select “<Add Event Log Entries...>" located under the Event Log heading.
8
Leave Event Log = Application.
9
Enter “MSExchange*” or “MSExchangeIS*” in the Source field.
10
Enter the range of Event IDs, excluding those you identified in step 5, and click OK.
For example, to exclude event 1029, enter “0-1028, 1030-999999".
11
Update any field values, such as Thresholds and Overrides, needed to align with the values you recorded in step 3. and click Finish.
Why do my Exchange Server Health tests exclude results for some perfmon or service counters? I am running Exchange 2010 SP1 without SP2.
In UC Diagnostics, counters in the Exchange Server Health test for Exchange 2010 have been updated to align with Exchange 2010 SP2. Some counter names are no longer supported.
If you do not plan to upgrade to Exchange 2010 SP2 in the near future, see the following tables, which show the names of the counters in Exchange 2010 SP1 that are no longer supported in SP2. You can add these counters back into your tests.
Why do my Exchange Server Health tests report incomplete results for some perfmon or service counters? I recently applied SP2 to Exchange 2010 SP1.
Exchange Server Health tests created prior to upgrading to Exchange 2010 SP2 contain counters that are no longer supported in SP2. You can delete existing tests and create new ones. However, you lose all customizations of these tests.
The alternative is to modify the existing Exchange Server Health tests. Delete the counters that are no longer supported as shown in the following tables.
 
Table 1. Perfmon counters no longer supported in Exchange 2010 SP2
PerfMon Counter Name - Exchange 2010 SP1
PerfMon Counter Name - Exchange 2010 SP2

MSExchangeAvailability\Queue OCS User Event Notifications

not supported

Process\%Processor Time (SpeechService)

not supported

Table 2. Service counters no longer supported in Exchange 2010 SP2
Service Counter Name - Exchange 2010 SP1
Service Counter Name - Exchange 2010 SP2

Service: Microsoft Exchange Speech Engine service (MSSpeechService)

not supported

Why are my test results very slow to load when I am switching between servers?
This can happen if you have test results history data retention set to a number higher than 7 days in the Global Settings. Decrease the data retention settings number.
My Internal Server Delivery Health test returns an incomplete result for Exchange servers that host only recovery databases?
When setting up an Internal Delivery Health test against a whole organization, exclude Exchange servers that are hosting only recovery databases from the test targets.
Why do some tests, such as the Exchange Internal Server Delivery, Exchange External Message Delivery, and Blackberry Message Delivery tests, not work in my Exchange 2007 and later environment with no public folders configured?
If you do not have a public folder configured in your Exchange 2007/2010 environment, all MAPI-based tests may fail to complete. The following may be displayed in the test results:
Summary:
Failed to execute the MAPI request on the requested Exchange server. Test results are not available.
Details:
The following error occurred while trying to logon to a mailbox on server <SERVERNAME>: Failed to open the default message store.
Make sure the account used for the test execution has rights to access the source mailbox on the Exchange server.
As a workaround, create a registry key on the computer on which your Diagnostic Service is located:
To create a MAPI settings registry key
1
Open registry editor and create the following registry key:
HKEY_LOCAL_MACHINE\SOFTWARE\Quest\UC Diagnostics\MapiSettings (on 32-bit computer)
- OR -
HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432Node\Quest\UC Diagnostics\MapiSettings (on 64-bit computer)
2
Add a new DWORD value named IgnoreNoPublicFoldersError under the created key. You do not have to provide a value.

Diagnostic Console Issues

The following questions and answers provide information about common diagnostic console issues.

Why do some of my data sources stop working after my Exchange cluster failed?
Some of the data sources may not reconnect properly on the new active node. Close and reopen the Diagnostic Console to force the data sources to reconnect properly.
Windows clusters are not supported as a separate plug-in.
Why does Last AB Update show “Data Not Available”?
If the target server is a standard edition server, the Address Book Share UNC path is not accessible externally, and the console cannot retrieve the file list to determine when the address book files were last updated.
What does it mean when my certificate reads RemoteCertificateChainErrors?
If your certificate reads RemoteCertificateChainErrors, the computer on which the console is running does not trust the Certificate Authority that was used for this server.
Why is the database information not available for my OCS connection?
The console attempts to gather data directly from the Back-End database. The user account used to run the Diagnostic Console must have rights on the Database server to gather perfmon information.
Why are Connected Endpoints failing to retrieve data?
The Connected Endpoints feature requires the ability to run SQL queries against the Back-End database. You must be a member of RTCUniversalServerAdmins to have the necessary rights.
When I go into the Archiving or Database drilldown, some tables read “No Data”. Why?
This happens if information has to be gathered from a computer beyond the initial target computer. If the Console cannot connect to the additional computer, an error message is not displayed. Instead the tables display a “No Data” message. This situation can occur if you have no rights, cannot resolve the NetBIOS name, or if the databases being queried do not allow external connections to the Backend database, Archiving server, or Archiving Database server.
Why are certificates not validating in the OCS and OCS R2 Diagnostic Console?
Certificates may not validate if you have proxy authentication configured but you have not yet entered the credentials for the proxy. To do so, go to Internet Explorer and enter your proxy credentials. For more information on proxy authentication, go to Microsoft's web site.
Why does the Diagnostic Console fail to launch sometimes?
The installer does not respect non-default configuration directories for the Diagnostic Console. You must uninstall and re-install the Diagnostic Console to resolve the issue.
On the Exchange 2007 Diagnostic Console, if you re-order the columns or hide the columns in the Storage Group drilldown, an error message appears. How do I resolve the error?
Right-click the column header and select Restore Default Settings.

Web Reports Issues

The following questions and answers provide information about common issues with the UC Diagnostics reports web site.

I am using Microsoft Windows 7. Why do the reports not display?
In Microsoft Windows 7, you must enable Default Document and Static Content in IIS.
To enable Default Document and Static Content
1
Select Start | Control Panel | Programs and Features | Turn Windows features on or off.
2
Double-click Internet Information Services | World Wide Web Services | Common HTTP Features.
3
Select the Common HTTP Features check box.
Default Document and Static Content should be automatically selected. If not, select their check boxes.
Why does the Storage Group Disk Utilization Web Report not display data when using Exchange 2010 or later servers?
For Exchange 2010 and later, the Exchange servers use DAGs instead of Storage Groups. Only Exchange 2007 servers show data for this report.
Why am I unable to browse to Web Reports using Internet Explorer?
User Account Controls (UAC) must be turned off on the Windows Server computer on which the Web Reports component is running. This affects the following operating systems: Windows Server 2008, Windows Server 2008 R2, Windows 7, Windows 8, Windows 8.1, Windows Server 2012, Windows Server 2012 R2, or Windows Server 2016 computers.
Why do the Web Reports not display information properly?

Your Web Reports may not display information properly for the following reasons:

•
If .NET is not configured properly
•
You have more than one version of .NET installed
If .NET is not configured properly

In some Microsoft environments, you may encounter a configuration issue with Microsoft .NET which results in the Web Reports not working properly. The error indicates that Microsoft .NET Framework version 2.0 must be installed on the IIS server that hosts Web Reports, and that Microsoft ASP.NET must be enabled on the virtual directory that contains the Web Reports. However, if you are sure that the Microsoft .NET Framework version 2.0 and ASP.NET are installed, you may need to repair the IIS mappings for ASP.NET.

If you encounter this issue, the following procedure explains how to properly configure Microsoft ASP.NET. To fix configuration mappings for ASP.NET.

1
Run the Aspnet_regiis.exe utility:
a
Click Start and click Run.
b
In the Open text box, type cmd, then click OK.
c
At the command prompt, type the following: "%windir%\Microsoft.NET\Framework\v2.0.50727\aspnet_regiis.exe" -i
d
Click OK.
2
Register the Aspnet_ispi.dll:
a
Click Start and click Run.
b
In the Open text box, type the following:
regsvr32 "%windir%\Microsoft.NET\Framework\v2.0.50727\aspnet_isapi.dll"
Regsvr32 returns the results of the registration.
c
Click OK.
3
Restart the IIS Service:
a
Click Start and click Run.
b
In the Open text box, type iisreset.
c
Click OK.
If Repairing the IIS Mappings for ASP.NET Fails

If you follow the procedure to repair the IIS mappings and you still see the error, the cause might be that the Network Service is not granted the correct permission to write to the IIS metabase.

To grant permissions to the Network Service to write to the IIS metabase
1
Click Start and select Run.
2
In the Open text box, type cmd and click OK.
3
Navigate to the .NET Framework directory. (%windir%\Microsoft.NET\Framework\version\)
In this path, the version represents the .NET Framework directory version that is installed on the server. Replace the placeholder in the example with an actual version number.
For example, at the command prompt, type the following:
CD /D "%windir%\Microsoft.NET\Framework\v2.0.50727"
4
Type the following command:
aspnet_regiis.exe -ga “nt authority\network service”

The “-ga” option grants the specified user or group access to the IIS metabase and other directories that are used by ASP.NET.

You have more than one version of .NET installed

If you have more than one version of Microsoft .NET Framework installed on your local computer, Internet Information Server (IIS) automatically uses the most recent .NET version. Currently, UC Diagnostics only supports Microsoft .NET 2.0, so you may need to change the version used by IIS.

To change the Microsoft .NET Framework version used by IIS
1
Click Start | Control Panel and select Administrative Tools.
2
Select the Internet Information Services (IIS) Manager.
3
Expand the Local computer | Web Sites | Default Web Site.
4
Right-click on the UCDiagnostics web application and click Properties.
5
Select the ASP.NET tab, and select .NET 2.0.50727 from the ASP.NET Version box.
6
Click OK.
관련 문서

The document was helpful.

평가 결과 선택

I easily found the information I needed.

평가 결과 선택