Troubleshoot your Cloud Connector deployment

Skype for Business Server 2015
 

마지막으로 수정된 항목: 2017-07-18

This topic describes solutions to common issues with Cloud Connector Edition deployments. If you are experiencing issues with calls to and from the Public Switched Telephone Network (PSTN), you can investigate by following the solutions described in this topic.

UNRESOLVED_TOKEN_VAL(CloudConnector_shortest) provides built-in mechanisms for automatically resolving some issues. An automatic detection process looks for potential issues with the UNRESOLVED_TOKEN_VAL(CloudConnector_shortest) deployment, and, if possible, takes corrective action to resolve those issues without the need for administrator intervention. The detection process works as follows:

  • Detection sequence: The process for detecting whether an appliance is down runs every 60 seconds.

  • Monitoring: The following services are monitored:

    • Mediation Server: RTCSRV and MEDSVC

    • Edge Server: RTCSRV

  • Recovery process: If any monitored services are down:

    • Three attempts are made to restart the failed service.

    • If the service restart is unsuccessful, then the process to restart all UNRESOLVED_TOKEN_VAL(CloudConnector_shortest) VMs will begin.

Following are solutions to commonly encountered issues:

  • Issue: Deployment fails or stops responding when running the deployment scripts. After logging on to each VM, the IP address is missing or incorrect for the Management/Internal/External NIC.

    Resolution: This issue cannot be resolved automatically. NICs cannot be added to VMs while they are running. Please shut down and remove these VMs in hyper-v manager, then run the following cmdlets:

    Uninstall-CcAppliance
    
    Install-CcAppliance
    
  • Issue: After the Active Directory Server and forest are installed, the CMS Server and/or Mediation Server did not join the domain correctly.

    Resolution: To resolve this issue, take the following steps:

    • Log on to the Active Directory Server and verify that the domain was created correctly.

    • Log on to the CMS/Mediation Server and verify that on the corpnet NIC a valid IP address is assigned, and that valid static IP and DNS is configured as the AD server's IP address.

    • Log on to the CMS/Mediation Server, and open a command prompt. Make sure you can ping the FQDN and IP address of the Active Directory Server. If you cannot, there may be an IP address conflict. Please try to assign a new IP for Active Directory and update DNS on the CMS/Mediation server accordingly.

  • Issue: You receive the following error message, “Remove-VMSwitch : Failed while removing virtual Ethernet switch. The virtual switch 'Cloud Connector Management Switch' cannot be deleted because it is being used by running virtual machines or assigned to child pools.“

    Resolution: The “Cloud Connector Management Switch” was not deleted after the deployment. If you hit this error, please go to Hyper-v manager and verify that there is not still a virtual machine still connected to it. If there are virtual machines still connected, disconnect them and delete the management switch. If the management switch still cannot be deleted, restart the host server and try again.

  • Issue: You receive the following error message, "Service RTCMRAUTH failed to start. Check to make sure the service is not disabled."

    note참고:
    This issue only applies to UNRESOLVED_TOKEN_VAL(CloudConnector_shortest) versions earlier than 1.4.2.

    The failure to start could also be because this Front End server was previously failed over (using computer fail over). If that's the case, please invoke fail back (using computer fail back).

    Resolution: This issue happens on an Edge Server when the root CA certificate or intermediate CA certificate is not trusted by the Edge Server. Even if the external certificate can be imported but the certificate chain is broken. Under this condition, the RTCMRAUTH and/or RTCSRV service can not start.

    Please import the root CA certificate and all intermediate CA certificates of your external certificate manually into the Edge Server and then restart the Edge Server. After you see the RTCMRAUTH and RTCSRV services started on the Edge Server, go back to your host server, launch a PowerShell console as administrator, and run following cmdlet to switch to the new deployment:

    Switch-CcVersion
    
  • Issue: The host server was restarted when Windows updates were applied, and calls serviced by the server are failing.

    Resolution: If you deployed a high availability environment, Microsoft provides a cmdlet to help move one host machine (deployment instance) into or out of the current topology when you check and install Windows update manually. Use the following steps to accomplish this:

    1. On the host server, start a PowerShell console as administrator, then run:

      Enter-CcUpdate
      
    2. Check for updates and install any that are available.

    3. In the PowerShell console, run the following cmdlet:

      Exit-CcUpdate
      
  • Issue: When a call is made from a Skype for Business client using a PSTN number, the call cannot be escalated to a conference by inviting another PSTN number.

    Resolution: To resolve this issue, see Configure online hybrid Mediation Server Settings.

  • Issue: A warning message is displayed about Windows Update when you are installing Active Directory server - "Windows automatic updating is not enabled. To ensure that your newly-installed role or feature is automatically updated, turn on Windows Update."

    Resolution: Launch a Tenant Remote PowerShell session using 비즈니스용 Skype tenant admin credentials, then run the following cmdlet to check the EnableAutoUpdate configuration of your site:

    Get-CsHybridPSTNSite
    

    If EnableAutoUpdate is set to True, you can safely ignore this warning message because the CCEManagement service will handle downloading and installing Windows updates for both virtual machines and the host server. If EnableAutoUpdate is set to False, run following cmdlet to set it to True.

    Set-CsHybridPSTNSite -EnableAutoUpdate $true
    

    Alternatively, you can manually check for and install updates. See the next section.

  • You receive an error message: Cannot register appliance because your current input/configuration <SiteName> or <ApplianceName> or <Mediation Server FQDN> or <Mediation Server IP Address> conflicts with existing appliance(s). Remove conflict appliance or update your input/configuration information then register again. ' when run Register-CcAppliance to register current appliance to online.

    Resolution: Values for the <ApplianceName>, <Mediation Server FQDN> and <Mediation Server IP Address> must be unique and only used for one appliance registration. By default, <ApplianceName> comes from the host name. <Mediation Server FQDN> and <Mediation Server IP Address> defined in configuration ini file.

    For example, using (ApplianceName= MyserverNew, Mediation Server FQDN=ms.contoso.com, Mediation Server IP Address=10.10.10.10) to register to SiteName=MySite, but if there is a registered appliance (ApplianceName= Myserver, Mediation Server FQDN=ms.contoso.com, Mediation Server IP Address=10.10.10.10), you will have the conflict.

    First, please check your CloudConnector.ini file in ApplianceRoot directory section. You will get <SiteName>, <Mediation Server FQDN> and <Mediation Server IP Address> values in the file. <ApplianceName> is your host server name.

    Second, Launch Tenant Remote PowerShell using your 비즈니스용 Skype tenant admin credentials, then run the following cmdlet to check the registered appliance(s).

    Get-CsHybridPSTNAppliance
    

    After identifying any conflicts, you can either update your CloudConnector.ini file with information that matches the registered appliance, or unregister the existing appliance to resolve the conflicts.

    Unregister-CsHybridPSTNAppliance -Force
    
  • The Get-CcRunningVersion cmdlet returns an empty value if there is a deployed appliance running on the host.

    Resolution: This can happen when you upgrade from 1.3.4 or 1.3.8 to 1.4.1. After you install version 1.4.1 with the .msi, you must run Register-CcAppliance before you run any other cmdlet. Register-CcAppliance will migrate the module.ini file from %UserProfile%\CloudConnector to %ProgramData%\CloudConnector. If you missed it, a new module.ini will be created in %ProgramData%\CloudConnector folder and replace the running/backup version information for 1.3.4 or 1.3.8.

    Compare module.ini files in %UserProfile%\CloudConnector and %ProgramData%\CloudConnector folder. If there are differences, delete the module.ini file in %ProgramData%\CloudConnector and rerun Register-CcAppliance. You could also modify the file manually to the correct running and backup version.

  • After you run the Switch-CcVersion cmdlet to switch to an old version which is different from current script version, there is no High Availability support for this old version.

    Resolution: For example, you have upgraded from 1.4.1 to 1.4.2. Your current script version, which can be determined by running Get-CcVersion, and your running version, which can be determined by running Get-CcRunningVersion are both 1.4.2. At this time, if you run Switch-CcVersion to switch the running version back to 1.4.1, then there will be no High Availability support for this old version.

    To get full High Availability support, please switch back to 1.4.2, so the running version and script version are the same. If you are experiencing problems with your 1.4.2 deployment, please uninstall and reinstall it as soon as possible.

  • Issue: Certificate authority certificates or internal certificates issued to Central Management Store, Mediation Server, and Edge Server are near expiration or are compromised.

    Resolution: Skype for Business certification authority certificates are valid for five years. Internal certificates issued to the Central Management Store, Mediation Server, and Edge Server are valid for two years.

    note참고:
    As of CCE 2.0, the Renew-CcServerCertificate cmdlet has changed to Update-CcServerCertificate and the Renew-CcCACertificate cmdlet has changed to Update-CcCACertificate.

    If internal certificates issued to the Central Management Store, Mediation Server, and Edge Server are near expiration or compromised, run the Renew-CcServerCertificate or Update-CcServerCertificate cmdlet to renew your certificates.

    If certification authority certificates are near expiration, run the Renew-CcCACertificate or Update-CcCACertificate cmdlet to renew your certificates.

    If certification authority certificates are compromised and there is only one appliance in the site, perform the following steps:

    1. Run the Enter-CcUpdate cmdlet to drain services and put the appliance in maintenance mode.

    2. Run the following cmdlets to reset and create new certification authority certificates and all internal server certificates:

      For CCE releases before 2.0

      Reset-CcCACertificate 
      Renew-CcServerCertificate 
      Remove-CcLegacyServerCertificate 
      
      

      or for CCE release 2.0 and later

      Reset-CcCACertificate 
      Update-CcServerCertificate 
      Remove-CcLegacyServerCertificate 
      
      
    3. Run the Exit-CcUpdate cmdlet to start services and and exit maintenance mode.

    4. Run the Export-CcRootCertificate cmdlet on the local file in the appliance, and then copy and install the exported certificate to your PSTN gateways.

    If certification authority certificates are compromised and there are multiple appliances in the site, perform the following steps:

    • On each appliance in the site run the Enter-CcUpdate cmdlet to drain services and put each appliance in maintenance mode.

    • On each appliance in the site, one by one, run the following cmdlets to reset and create new certification authority certificates and all internal server certificates:

      For CCE releases before 2.0

      Reset-CcCACertificate
      Renew-CcServerCertificate
      Remove-CcLegacyServerCertificate 
      
      

      or for CCE release 2.0 and later

      Reset-CcCACertificate
      Update-CcServerCertificate
      Remove-CcLegacyServerCertificate 
      
      
    • On each appliance in the site, run the Exit-CcUpdate cmdlet to start services and exit maintenace mode.

    • Run the Export-CcRootCertificate cmdlet on the local file in any appliance, then copy and install the exported certificate to your PSTN gateways.

  • Issue: You need to check for and install Windows updates manually on the host server or virtual machines.

    Resolution: We recommend that you take advantage of automated OS updates provided by Skype for Business Cloud Connector Edition. After an appliance is registered for online management and auto OS update is enabled, the host server and virtual machines will check and install Windows Updates automatically according to the OS update time window settings.

    If you need to check for and install Windows Updates manually, follow the steps in this section that apply to your type of deployment. You should plan on updating both the host server and the virtual machines running on it at the same time to minimize the amount of down time necessary for the updates.

    If you prefer, you can use a Windows Server Update Services (WSUS) server to provide updates to Cloud Connector servers. Just be sure to configure the Windows Update to not install automatically.

    For information about how to manually update your Cloud Connector deployment, see the following section.

  • Issue: You receive the following error message in the Cloud Connector Management Service Log, "C:\Program Files\Skype for Business Cloud Connector Edition\ManagementService\CceManagementService.log”: CceService Error: 0 : Unexpected exception when reporting status to online: System.Management.Automation.CmdletInvocationException: Logon failed for the user <Global Tenant Admin>. Please create a new credential object, making sure that you have used the correct user name and password. --->

    Resolution: The Office 365 global tenant admin credentials have been changed since the UNRESOLVED_TOKEN_VAL(CloudConnector_shortest) appliance was registered. To update the locally stored credentials on the UNRESOLVED_TOKEN_VAL(CloudConnector_shortest) appliance, run the following from Administrator PowerShell on the host appliance:

    Set-CcCredential -AccountType TenantAdmin
    
  • Issue: After you change the password for the host server account you used for the deployment, you receive the following error message: “ConvertTo-SecureString : Key not valid for use in specified state.” in %ProgramFiles%\Skype for Business Cloud Connector Edition\ManagementService\CceManagementService.log or while running the Get-CcCredential cmdlet.

    Resolution: All Cloud Connector credentials are stored in the following file: “%SystemDrive%\Programdata\Cloudconnector\credentials.<CurrentUser>.xml”. When the password on the host server changes, you will need to update the locally stored credentials.

    If you are running Cloud Connector version 1.4.2, regenerate all Cloud Connector passwords by following these steps:

    1. Restart the host server.

    2. Delete the following file: “%SystemDrive%\Programdata\Cloudconnector\credentials.<CurrentUser>.xml”.

    3. Launch a PowerShell console as administrator, and then run “Register-CcAppliance -Local” to re-enter the passwords following the description. Enter the same passwords you entered before for the Cloud Connector deployment.

    If you are running Cloud Connector version 2.0, regenerate all Cloud Connector passwords by following these steps:

    1. Restart the host server.

    2. Delete the following file: “%SystemDrive%\Programdata\Cloudconnector\credentials.<CurrentUser>.xml” .

    3. Launch a PowerShell console as administrator, and then run “Register-CcAppliance -Local” to re-enter the passwords following the description.

    If the cached password file was generated with Cloud Connector version 1.4.2, use the VMAdmin password for the CceService password when prompted. Enter the same password you entered before for the Cloud Connector deployment for all other accounts.

    If the cached password file was generated with Cloud Connector version 1.4.2 and the DomainAdmin and VMAdmin passwords are different, you must perform the following steps:

    1. Run Set-CcCredential -AccountType DomainAdmin as follows:

      1. When prompted for the old account credential, enter the credential you used for the CceService password.

      2. When prompted for the new account credential, enter the password for the DomainAdmin password you used before.

    If the cached password file was generated with Cloud Connector version 2.0, by default, VmAdmin and DomainAdmin use the same password as CceService. If you changed the DomainAdmin and VMAdmin passwords, you must perform the following steps:

    1. Run Set-CcCredential -AccountType DomainAdmin as follows:

      1. When prompted for the old account credential, enter the credential you used for the CceService password

      2. When prompted for the new account credential, enter the password for the DomainAdmin password you used before.

    2. Run Set-CcCredential -AccountType VmAdmin as follows:

      1. When prompted for the old account credential, enter the credential you used for the CceService password

      2. When prompted for the new account credential, enter the password for the VmAdmin password you used before.

  • Issue: You receive the following error message “Dismount-WindowsImage : Dismount-WindowsImage failed. Error code = 0xc1550115” when installing or upgrading Cloud Connector Edition.

    Resolution: Launch a PowerShell console as administrator, run “DISM -Cleanup-Wim’”. This will clean up all troubled images. Run Install-CcAppliance again or wait for the bits to automatically upgrade.

If you do not want to use automatic updates in your environment, follow these steps to manually check for and apply Windows updates. Checking for and installing Windows updates may require a server restart. When a host server is restarting, users will not be able to use UNRESOLVED_TOKEN_VAL(CloudConnector_shortest) to place or receive calls. You can manually check for and install updates to control when the updates take place, and then restart the machines as needed during times you choose to avoid disruption of services.

To manually check for updates, connect to each host server and open the Control Panel. Select System and Security >Windows Update, and then manage the updates and server restarts as appropriate for your environment.

  • If there is only one appliance in the site, connect to each virtual machine and open the Control Panel. Select System and Security >Windows Update, and then configure the updates and server restarts as appropriate.

  • If there are more than one appliance in the site, the instance being updated and restarted cannot be accessed by users during the updates. Users will connect to other instances in the deployment until all virtual machines and all Skype for Business services start on the virtual machines after the updates are completed. To avoid any potential disruption to service, you can remove the instance from HA while you apply the updates, and then restore it when complete. To do so:

    1. On each host server, open a PowerShell console as administrator.

    2. Remove the instance from HA with the following cmdlet:

      Enter-CcUpdate
      
    3. Follow the steps for a single instance to manually apply the updates and restart the virtual machine.

    4. Set the instance back to HA with the following cmdlet:

      Exit-CcUpdate
      

For multi-site deployments, follow the steps for a single site for each site in the deployment, applying updates to one site at a time.

 
표시: