Blog - Title

November, 2008

  • Using Network Monitor 3 to Troubleshoot a Domain Join Failure Caused by a Black Hole Router

    This is Randy again with an interesting case that I had recently. We were having problems trying to join certain workstations to the domain. We would see that every workstation in one site would join successfully and all the workstations in another site would fail with an error indicating that we could not locate a domain controller for that domain. My first hunch was either the domain controllers in the one site were broken, or there were networking issues in that problem site. The first step in troubleshooting is “Check the Event Logs!” We did not see any alarming events on any of the domain controllers in the problem site. So my next step was to take a network trace of the issue. With the help of a networking engineer here at Microsoft, Tim Quinn, we analyzed the traffic of a successful domain join and a failure. We took a simultaneous trace from the workstation and authenticating domain controller to ensure that we could see both sides of the conversation and uncover any failures across the wire.

    We used Network Monitor 3.2 to take the traces. You can find some very helpful webcasts on working with Network Monitor on the Netmon blog. Here is a snapshot of some traffic between the domain controller and the workstation that is attempting to join. This is from the Frame Summary pane and is a general overview of each frame sent on the wire.

    image

    You can see the protocol and description of each frame. We have a lot of traffic on RPC and SMB and from the frame description we see that this is communication on named pipes: Netlogon, Samr, and LSARPC. These are the connection points involved in a domain join between a workstation and a domain controller. By highlighting one of these frames in the Frame Summary pane, we can see each network layer of the frame in the Frame Details pane.

    image

    It is important to understand that data transfer uses encapsulation to transfer information from a process on one computer to a process on another computer. In the above example we see the LDAP client on a workstation talking to the LDAP server on a domain controller using the defined specifications of the LDAP protocol. This data is packaged in a TCP packet which is built using the specifications of the TCP protocol, which is packaged in an IP datagram, which is packaged on an Ethernet frame. Netmon separates each of these packages in order for you to analyze the behavior of each protocol individually.

    So now we have two traces that show every frame on the wire: one from the perspective of the workstation and one from the perspective of the domain controller. So how do we find a frame in one trace to the corresponding frame in the other trace? We have a couple of frame and packet attributes that can help; the first one we will discuss is the identification number…

    The identification number is an attribute of the IP datagram that is sent across the wire. This attribute is as simple as it sounds; it is just a random generated number to identify a specific IP datagram. If we expand the IPv4 header information from the above example, we see an attribute named Identification with a value of 3201.

    image

    If we look at either the network trace taken on the domain controller or the trace taken on the workstation, there will be a frame with an Identification number of 3201. You can filter both traces for this frame by using the filter IPv4.Identification == 3201.

    Another way to line up the conversation of two traces taken simultaneously is to compare the Sequence and Acknowledgement numbers. These attributes are at the TCP layer instead of IPv4. To view these attributes, expand the TCP header. This is from the same frame as above:

    image

    We see that the last packet sequence number sent in this frame is 4167329214, and the last packet that we received from our partner in this communication is 1946363494. These numbers can often be misleading, because a router can strip and resend at the network layer (IP layer) and all the numbering can be misleading from the IP layer up (In this case TCP.) To align to simultaneous traces, I use the Identification attribute from above, and I use the sequence and acknowledgement numbers to verify dropped and received packets. To learn more about Sequence and Acknowledgement numbers and how TCP works, check out the following KB article:

    Explanation of the Three-Way Handshake via TCP/IP
    http://support.microsoft.com/kb/172983

    In comparing these traces, we see a breakdown in the communication. From the frame summary, it would appear to be an LDAP problem. After further analysis, we see that the issue is at the TCP layer. The next two snapshots are from the simultaneous traces and an explanation under each.

    image

    This is from the trace taken on the workstation and we see at the top, frame identification number 3201 which is our LDAP request. After this we get a strange Kerberos packet. This is actually an out of sequence packet that was the last part of the LDAP response from the domain controller. Because it is an incomplete portion of the response, Netmon did not parse the frame correctly and it shows as a Kerberos packet. Beyond that, we see the workstation eventually abandon the LDAP request (frame 133).

    This is from the trace taken on the domain controller and we see at the top, frame identification number 3201 which is our LDAP request. We see that the DC does respond, but we send two frames of data, the first which never makes it to the workstation and the second frame (frame 125) that does successfully make it to the workstation as an out of sequence packet. After this, we never receive an acknowledgement from the workstation and we see the domain controller resend the missing packet (frame 127 and 128).

    image

    So what ever happened to the mysterious disappearing packet? This was caused by a router that would drop packets with a Maximum Transmission Unit (MTU) size too large to forward. This issue is known as a black hole router. We were able to change the MTU size sent and this resolved our issue.

    Even though this blog is AskDS, it is important to understand the networking components used by Directory Services. By using Network Monitor, you can avoid time spent troubleshooting the wrong component.

    - Randy Turner

  • Forced Demotion of a Windows Server 2008 Core Domain Controller

    Ned here again. Today's post is short and sweet, but when you need this one you will need it fast and we don't have this publically documented anywhere on TechNet (yet).

    Since Windows 2000 SP4, it has been possible to forcibly demote Domain Controllers using the DCPROMO /FORCEREMOVAL command. You could use this switch if:

    • There are no domain controllers currently available in the parent domain when you try to demote the last domain controller in an immediate child domain.
    • The Active Directory Installation Wizard cannot complete because there is a name resolution, authentication, replication engine, or Active Directory object dependency that you cannot resolve after you perform detailed troubleshooting.
    • A domain controller has not replicated incoming Active Directory changes in Tombstone Lifetime (Default Tombstone Lifetime is 60 days) number of days for one or more naming contexts (see also KB article 216993).
    • Time does not permit more detailed troubleshooting because you must immediately bring into service the domain controller.

    Naturally you must always follow this up with Metadata Cleanup on other DCs.

    However, if you try to run this command on a Windows Server 2008 DC running in Core (no GUI) mode, it will always return:

    An answer file or unattend installation command-line parameters must be specified.

    And even if you provide an answer file, you will continue to get various prompts and issues that prevent demotion.

    So how do we get this to work? With the following command:

    dcpromo /forceremoval /demotefsmo:yes /administratorpassword:<the new password>

    For example:

    dcpromo /forceremoval /demotefsmo:yes /administratorpassword:Password1

    So I can already hear the question "But this server did not have any FSMO roles on it - why do I need to add that switch?" The answer is because this not only forces demotion with a FSMO role in place, it also suppresses the warning prompt for FSMO when forcibly demoting.

    I tried to come up with a snappier title for this, but I figured most readers that actually needed this post would already be so busy that they wouldn't want any cutesy-pie stuff. :-)

    - Ned Pyle

  • Configuring an Authoritative Time Server with Group Policy Using WMI Filtering

    Hello everyone, Brian Singleton here. Here’s a question I often get from customer regarding Windows Time:

    “Is there a way I can configure the Windows Time settings via Group Policy and have it only apply to the domain controller that holds the PDC FSMO role?”

    This is a wonderful question to pose and there are very good reasons why this should be done:

    1. If you decide to move the PDC emulator role to another domain controller, you do not want to have to go through the trouble of making all of the registry changes again.
    2. If the PDC emulator fails, and you have to bring up a new server, you may forget to add the settings back, resulting in a time sync issue in your domain.

    We have a feature in Group Policy to help us that you may have read about on this blog, and that is WMI filtering.

    Windows Management Instrumentation (WMI) is a powerful feature in Windows that we can leverage to provide us very detailed information about computers in our environment. We can use WMI via a script to remotely manage machines, as well as gather information about machines in our domain for inventory purposes.

    The main reason why I have brought this feature up is that we can provide an additional layer of filtering for Group Policy application using WMI.

    We can configure a GPO on the domain controller OU for our W32Time settings to configure the authoritative time server, but instead of using security filtering and explicitly securing it for the domain controller that has the PDC emulator role; WMI filtering can be used instead. It is important to state here before continuing is that WMI filtering will only work with computers running Windows XP/Windows Server 2003 and later. That means that you cannot use WMI filtering with Windows 2000.

    Below is an example:

    The domain I configure this policy on is Windows Server 2003, but the same applies to Windows Server 2008 as well. I am also using Group Policy Management Console (GPMC) which can be downloaded from here. For those of you who are using Windows Vista you can get GPMC by downloading the Microsoft Remote Server Administration Tools (RSAT).

    First I will create my WMI filter:

    clip_image002

    The next part is me adding my query:

    clip_image004

    clip_image006

    In the above image I added the following query:

    Select * from Win32_ComputerSystem where DomainRole = 5

    You can use WMIC to verify the current value of the DomainRole property. This can be a helpful way to get a sanity check on the value to make sure the filtering will achieve the desired result.

    To view the DomainRole value locally:

    wmic computersystem get domainrole

    To view the DomainRole value remotely (where M1 is the remote computer):

    wmic /node:”M1” computersystem get domainrole

    In WMI we break up the various components of the OS and actual machine into classes. The Win32_ComputerSystem class is for computers running a Windows OS. Have a look at the following MSDN link for this class as well as other WMI classes:

    WMI Classes
    http://msdn.microsoft.com/en-us/library/aa394554(VS.85).aspx

    The Win32_ComputerSystem class has a lot of methods that can used in scripting as well as filtering for Group Policy but for the purposes of this post we will focus on the DomainRole:

    From the MSDN website:

    DomainRole

    Role of a computer in an assigned domain workgroup. A domain workgroup is a collection of computers on the same network. For example, a DomainRole property may show that a computer is a member workstation. This property is inherited from CIM_ManagedSystemElement.

    Value

    Meaning

    0

    Standalone Workstation

    1

    Member Workstation

    2

    Standalone Server

    3

    Member Server

    4

    Backup Domain Controller

    5

    Primary Domain Controller

    As you can see from the above chart 5 means Primary Domain Controller. So the query, Select * from Win32_ComputerSystem where DomainRole = 5, means select a machine whose DomainRole is 5, Primary Domain Controller. For those of you who would like to create a Windows Time GPO for all the other domain controllers you would just change it to DomainRole=4.

    Now I am going to link my WMI filter to my already configured Authoritative Time Server GPO:

    clip_image008

    At the end what I have just accomplished is that the Authoritative Time Server GPO will only apply to the domain controller who holds the PDC emulator FSMO role. By configuring the policy in this fashion, I can transfer the PDC role to any domain controller and the policy will follow the role. Also, if the PDC fails and I bring up a new domain controller and seize the PDC emulator role to the new domain controller, the policy will apply on the next policy refresh or by forcing a group policy refresh.

    I hope that you have learned a little more on how powerful WMI Filters are and how they can be leveraged to apply Group Policies based on a WMI Filter.

    Additional Resources

    Windows Management Instrumentation
    http://msdn.microsoft.com/en-us/library/aa394582(VS.85).aspx

    WMI Classes
    http://msdn.microsoft.com/en-us/library/aa394554(VS.85).aspx

    WMI filtering using GPMC
    http://technet.microsoft.com/en-us/library/cc779036.aspx

    Windows Management Instrumentation Command-line
    http://technet.microsoft.com/en-us/library/cc784189.aspx

    - Brian Singleton

  • So You Want to Upgrade to Windows 2008 Domain Controllers (ADPREP)

    Hi all, Rob Newhouse here and today I am talking about upgrading your domain to Windows Server 2008 and what you may see in the process, plus a couple of tips to make your transition a smooth one.

    This post will show the proper use of ADPREP and what to expect when you are running it.

    ADPREP is broken down into four stages with Windows Server 2008, instead of the two that most of you may be familiar with when you upgraded to Windows Server 2003. The four steps include Forest Preparation, Domain Preparation, Group Policy Preparation and Read-Only Domain Controller (RODC) Preparation (which you have to run if you want to add a RODC to your environment). You will use ADPREP.exe to perform all of these steps.

    clip_image002

    Preparing to Run ADPREP /forestprep

    ADPREP /forestprep makes modifications to the schema. In order to successfully run it you should:

    1. Have a good system state backup for every domain controller in your forest, or at the very least a good system state backup for one domain controller for each domain in the forest.
    2. Be logged on as a user that belongs to the Domain Admin, Schema Admin and Enterprise Admin groups in the forest root domain.
    3. Ensure that you are running Windows 2000 SP4 or later on all domain controllers in the forest.
    4. You must run ADPREP /forestprep on your schema master.
    5. If you are running Exchange 2000 in your environment refer to KB article 325379 How to upgrade Windows 2000 domain controllers to Windows Server 2003.
    6. Ensure replication is working throughout the entire forest, including that all domain controllers are up and running and that the schema master has been up long enough for a complete replication cycle to happen for the Schema partition.

    So let’s go through all these preparatory steps in detail

    1. First you should perform a system state backup on all of your domain controllers using either Windows Backup (NTBackup) or a third-party backup tool. This step is necessary if you find that your schema is incompatible and you need to roll back to a previous state.
    2. Next, check to see if your account has the appropriate group memberships. Open Active Directory Users and Computers, right-click the account you are using to do the upgrade and choose Properties. Select the Member Of tab. If you do not see Domain Admins, Enterprise Admins and Schema Admins, add the ones you are missing. Log off and back on, then run whoami /groups in a command prompt to verify the groups are in your security token.

      clip_image004
    3. ADPREP /forestprep will check to see if you are running at least Windows 2000 SP4. If you have Windows 2000 domain controllers in your environment you should upgrade them all to SP4. You can download SP4 from here - Windows 2000 Service Pack 4 for IT professionals.
    4. Next, check to see if you are logged on to your schema master. There are two ways to accomplish this. One is to run regsvr32 schmmgmt.dll so you can load the Active Directory Schema snap-in. Open a new MMC and add Active Directory Schema. Right-click on the words Active Directory Schema and choose Operations Master.

      clip_image006

      clip_image008

      clip_image010

      Another alternative is to run netdom query fsmo from a command prompt. Netdom is part of the Windows Server 2003 Support Tools.

      clip_image012
    5. There are known issues with upgrading a Windows 2000 domain with Exchange 2000 running in the environment. There are different scenarios with different steps in KB article 325379 to address problems that have been encountered in the upgrade process. You will be performing one of the scenarios regardless. It is just a matter of which scenario you will have to perform.
    6. The final verification is to check and make sure replication is working. To do this install the Windows Server 2003 Support Tools if you do not have them already installed. Run repadmin /showreps from a command prompt.

      clip_image014

      You are looking for Last attempt @ date\time was successful. Any errors should be addressed before attempting to run ADPREP /forestprep.

      NOTE:
      ADPREP /forestprep will only check to see if replication is working on your schema master. It will not check the replication status of all DCs in your environment. Repadmin /showreps will only check the DC that you focus it on. In order to check the entire environment you will want to run repadmin /replsum. This command will give you the status of your entire environment. You will want to fix any errors you have with replication prior to running ADPREP /forestprep.

    Running ADPREP /forestprep

    1. Now you are ready to prepare your forest. This procedure takes a while depending on the speed of your computer so do not interrupt it. Insert your Windows Server 2008 DVD into the DVD drive on the schema master.
    2. Open a command prompt.
    3. Change your drive letter to the DVD drive. If you do not have a DVD drive on your schema master you can copy the Sources\Adprep folder to your local drive and run it from the copy.
    4. Change into the Sources\Adprep directory.
    5. Run ADPREP /forestprep.
    6. You will get a warning that you need to be running Windows 2000 SP4 or later.
    7. Type C and press Enter.
    8. You will see a series of updates from LDF files.
    9. If all goes well, you will see ADPREP successfully updated the forest-wide information.

    clip_image016

    clip_image018

    Preparing to Run ADPREP /domainprep

    After a successful completion of ADPREP /forestprep, you will be ready to run ADPREP /domainprep. ADPREP /domainprep must be run against each domain that you wish to upgrade.

    Prerequisites

    In order to run ADPREP /domainprep you should:

    1. Have successfully completed ADPREP /forestprep.
    2. Be a domain admin for the domain you are running it on.
    3. Be at Windows 2000 Native Mode Domain Functional level.
    4. Have access to the Infrastructure Master.
    5. Wait for the schema changes to replicate throughout the environment, or at least the Infrastructure Master must have the schema updates replicated to it.

    Note: Upgrading from Windows 2000 is not supported. For more information see Guide for Upgrading to Windows 2008.

    Running ADPREP /Domainprep

    1. Insert the Windows Server 2008 DVD.
      Open a command prompt.
    2. Change your drive letter to the DVD drive.
    3. Change your directory to Sources\Adprep.
    4. Run ADPREP /domainprep.

    clip_image020

    For a better understanding of what will occur running the ADPREP /Domainprep command, I have referenced the KB article Enhancements to ADPREP.exe in Windows Server 2003 Service Pack 1(Q324392). The More Information section describes the functionality post-Windows 2003 SP1, including the Windows 2008 ADPREP.

    Preparing to Run ADPREP /domainprep /gpprep

    ADPREP /domainPrep /gpprep only adds the inheritable access control entries on Group Policy objects in the Sysvol share. If you run it prior to running adprep /domainprep it will run both functions, first the domain prep and then the GP prep.

    Prerequisites

    In order to run ADPREP /domainprep /gpprep you should:

    1. Have completed the prerequisites for ADPREP /domainprep.
    2. Have Sysvol\Sysvol\Policies\{Default Domain and Default Domain Controller GPO guids} in place.

      a. In Windows Explorer Navigate to your Sysvol\Sysvol\Domain\Policies folder

      b. Verify the following GUIDs are inplace

      {31B2F340-016D-11D2-945F-00C04FB984F9}
      {6AC1786C-016F-11D2-945F-00C04FB984F9}

    Note Upgrading from Windows 2000 is not supported. For more information see Guide for Upgrading to Windows 2008.

    Running ADPREP /domainprep /gpprep

    1. Insert the Windows Server 2008 DVD.
    2. Open a command prompt.
    3. Change your drive letter to the DVD drive.
    4. Change your directory to Sources\Adprep.
    5. Run ADPREP /domainprep /gpprep.

    clip_image022

    ADPREP /domainprep /gpprep without running adprep /domainprep first.

    clip_image024

    ADPREP /domainprep /gpprep after running adprep /domainprep

    Preparing to Run ADPREP /rodcprep

    RODC’s (Read-Only Domain Controllers) are a cool new feature added in Windows Server 2008. The benefits of a RODC in certain domain configurations are well worth the effort of learning and implementing them. For more information on the benefits, see RODC Features on TechNet. If you intend to introduce them into your environment you will have to run ADPREP /rodcprep. This command prepares partitions in Active Directory so RODC’s can be used by adding security to the ForestDNS, DomainDNS, and Domain partitions.

    Prerequisites

    In order to run ADPREP /domainprep /rodcprep you should:

    1. Be a Domain Admin and Enterprise Admin.
    2. Be able to contact all Infrastructure Master role holders in the forest.

    Note ADPREP /rodcprep will let you run without first running ADPREP /forestprep and ADPREP /domainprep, however it is not recommended.

    Running ADPREP /rodcprep

    1. Insert the Windows Server 2008 DVD. 
    2. Open a command prompt.
    3. Change your drive letter to the DVD drive.
    4. Change your directory to Sources\Adprep.
    5. Run ADPREP /domainprep /rodcprep.

    clip_image026

    That concludes this post on running ADPREP. Running through the steps in order should eliminate many of the problems you might otherwise encounter.

    - Rob Newhouse

  • Understanding How to Read a Userenv Log – Part 2

    Mark here. As promised here is the second part of Understanding How to Read a Userenv Log (see Part 1 if you missed it). Most of the time when we need to enable Userenv logging is due to a delay of getting to the desktop after entering our credentials. Normally the user will be stuck at “Applying Personal Settings” is most often the scenario we see. The problem can stem from several different problems. One can be that another process, usually the anti-virus, has a lock on a file within the profile and cannot open the file. When this happens the user will get a new default profile. A reboot of the client will temporarily resolve this forcing all handles to be released. This does not always work but at times it will allow the user to logon and receive his old profile. Eventually the problem will come back. There also could be permissions issue or network issues when attempting to load a roaming profile. There are many different scenarios that plague users and loading their profiles.

    I am going to continue using a Userenv log that I have where the customer was seeing the desktop hanging at “Applying Personal Settings.” To find the portion where the user starts his logon, search the log file for the user alias that they use to logon with:

    USERENV(750.754) 22:01:02:796 LoadUserProfile: lpProfileInfo->lpUserName = <User1>

    You will also see this:

    USERENV(750.754) 22:01:02:796 LoadUserProfile: lpProfileInfo->lpDefaultPath = \\DC1\netlogon\Default User

    By default we always look to the Netlogon share on a domain controller for the Default User profile. This goes way back to the old NT 4.0 days and it still is hanging around. Next we resolve the users name to a SID (Security Identifier) as follows:

    USERENV(750.754) 22:01:02:812 LoadUserProfile: User sid: S-1-5-<domain sid-rid>

    We do a check and see if the user is a local admin on the client they are logging onto:

    USERENV(750.754) 22:01:02:812 RestoreUserProfile: User is a Admin

    Now we actually go look and see if an existing profile exists or not to determine if we need to create a new one or whether the user has an existing profile:

    USERENV(750.754) 22:01:02:812 ExtractProfileFromBackup: A profile already exists
    USERENV(750.754) 22:01:02:812 PatchNewProfileIfRequred: A profile already exists with the current sid, exitting
    USERENV(750.754) 22:01:02:812 GetExistingLocalProfileImage: Found entry in profile list for existing local profile
    USERENV(750.754) 22:01:02:812 GetExistingLocalProfileImage: Local profile image filename = <C:\Documents and Settings\user1>
    USERENV(750.754) 22:01:02:812 GetExistingLocalProfileImage: Expanded local profile image filename = < C:\Documents and Settings \user1>
    USERENV(750.754) 22:01:02:812 GetExistingLocalProfileImage: No local mandatory profile. Error = 2
    USERENV(750.754) 22:01:02:812 GetExistingLocalProfileImage: Found local profile image file ok < C:\Documents and Settings \user1\ntuser.dat>

    So where do you look to find out if the user has an existing profile?

    HKLM\Software\Microsoft\Windows NT\CurrentVersion\ProfileList

    Under this key will be a list of SID’s corresponding to each user that has logged onto this client machine. If you highlight the ProfileList key you will notice some values there such as AllUsersProfile, DefaultUserProfile and the ProfilesDirectory. These values tell the system where the default directory is for the profiles then under that we know which directory the All Users and the Default Profile are. Drilling down underneath you can find the SID for the user you are looking for. Highlight it and note the ProfileImagePath value. This will point you to the directory where the system thinks your profile should be. There are other values here as well but we are not going to get into those. Note in the above output we check and see if the profile is a mandatory profile or not. How does it know whether the profile is mandatory or not? See the next line in the output at the end of the string we see Ntuser.dat in the profile path. If the Ntuser ended with a .man that would signify that the profile is mandatory. As you can see it ends with the .dat so the error = 2 means, “The system cannot find the file specified.” You can simply run “net helpmsg 2” from a command prompt to resolve some of the error messages to find out what they actually mean. In this case it did not find the Ntuser.man as the file it was looking for. So another question, what is the ntuser.dat file? This is the actual HKEY_CURRENT_USER hive you see in the registry when opening Regedit. It is simply a registry hive for the user’s profile to contain certain settings like desktop wallpaper, screen saver, application settings, etc. Once the profile is loaded we should see a line in the debug like this:

    USERENV(750.754) 22:01:02:874 LoadUserProfile: Leaving with a value of 1.

    The value of 1 means it was successful. This is not the time where the desktop starts showing up so don’t get confused here with that - the profile was loaded successfully. This simply means we loaded the Ntuser.dat file into the registry as the HKEY_CURRENT_USER hive. The next thing we do is start the Group Policy Processing for the user account:

    USERENV(750.dec) 22:01:03:796 ProcessGPOs: Starting user Group Policy (Background) processing...

    We go through the same as we did for the computer account; we need to determine the LDAP path of the user so we know where to look for any policies to apply.

    USERENV(750.dec) 22:01:03:828 ProcessGPOs: User name is: CN=User1,OU=Users,OU=TX,OU=USA,DC=Domain,DC=com, Domain name is: Domain
    USERENV(750.dec) 22:01:03:828 ProcessGPOs: Domain controller is: \\DC1.Domain.COM Domain DN is Domain.COM

    I won’t go through all the policy processing as we did in part one as it is all the same. The GPC and the GPT version numbers are checked, the display name, the common name is all the same. To know when we are done applying the user group policies look for the following in the log file:

    USERENV(750.dec) 22:08:54:183 ProcessGPOs: User Group Policy has been applied.
    USERENV(750.dec) 22:08:54:183 ApplyGroupPolicy: Leaving successfully.

    So if you notice now between the time we logged in (22:01:02) loaded the profile and applied group policies successfully (22:08:54) took approximately 8 seconds. Also note that the desktop is still not loaded by default, we are not done yet. Now comes the time to run any logon scripts:

    USERENV(46c.4c4) 22:08:54:777 LibMain: Process Name: C:\WINNT\System32\WScript.exe

    Now we are almost done, the GINA calls userinit.exe which in turn calls explorer.exe. Explorer is the process that brings up the shell which is the desktop with all its icons, etc. So in troubleshooting why it is taking so long for either the Userinit being called or Explorer being called you have to look to the log and see what is occurring.

    USERENV(8c4.8c8) 22:08:55:808 LibMain: Process Name: C:\WINNT\system32\userinit.exe
    USERENV(8d8.8dc) 22:28:57:824 LibMain: Process Name: C:\WINNT\Explorer.EXE

    Note the approximately 20 minutes difference in the time stamps of the two events above. So what is going on during this time? This is what was happening:

    USERENV(6e4.6d4) 22:10:16:856 GetUserDNSDomainName: Domain name is NT Authority. No DNS domain name available.
    USERENV(6e4.6d4) 22:10:17:090 ImpersonateUser: Failed to impersonate user with 5.

    This was being logged into the Userenv log constantly. If you remember from Part 1, the 6e4 was the PID of the process that was running. In this log we could see that this PID was delaying Explorer from being called to start the shell. The PID 6e4 (hex) resolves to 1764 (decimal). If we look back in Task Manager we can see that the PID resolves to a 3rd party security service that was running. In this case we disabled that service and rebooted. We tested and no other issues came up with a delay in getting to the desktop. Here is another tidbit to help you out in case you are troubleshooting a remote machine. Let’s say someone sends you a Userenv.log file to look over, you see something similar as above but the user has either rebooted or shut down the machine so you have no way to look up that PID. Every time a service restarts or the machine restarts the process will have a new PID. Have the user generate a System Information file by going to Start – Run – Msinfo32 and hit enter. Once the System Information file comes up save the file as a System Information File with a .NFO extension. Now have the user send that file as well then in the left hand column expand Software Environment and highlight Running Tasks. There you will see all the processes running and their PID’s.

    I will touch on some aspects of a roaming profile, all the information above has been for local profiles. When using a roaming profile you will see some slight differences in the Userenv log. For example if the roaming profile path is set on the user properties on the Account tab you will see the following:

    USERENV(750.754) 22:01:02:796 LoadUserProfile: lpProfileInfo->lpUserName = <User1>
    USERENV(750.754) 22:01:02:796 LoadUserProfile: lpProfileInfo->lpProfilePath =
    \\fileserver\profiles\User1

    This tells us that the roaming profile is located on the server called fileserver at the share called profiles then the user name. Once the SID has been resolved to a name we check to see if it is a cross forest logon.

    USERENV(750.754) 22:01:02:796 CheckXForestLogon: checking x-forest logon, user handle = 560
    USERENV(750.754) 22:01:02:796 CheckXForestLogon: not XForest logon.

    There is a policy that allows cross forest roaming profiles to be loaded. By default if user account is located in one forest and they attempt to logon to a machine in another forest they will get a new default profile unless the following policy is set:

    Computer Configuration \ Administrative Templates \ System \ Group Policy \ Allow Cross-Forest User Policy and Roaming User Profiles

    This policy will need to be set in the domain where the machine account is located. Next we actually map a drive to the share \\fileserver\profiles and most of the time it will be E drive. The drive will look like E:\user1 now. Next we go through the same procedure as before, pinging the DC to determine if we are on a fast link or not. Now we check to see if the user account has ownership of the files in the profile:

    USERENV(750.754) 22:01:02:796 CheckRoamingShareOwnership: checking ownership for E:\User11
    USERENV(750.754) 22:01:02:796 CheckRoamingShareOwnership: owner is the right user

    At times the user may not be the owner of the files located in the share. Let’s say the file server was changed and now points to a new file server then the files copied over manually to the new server share. The Administrators group will most likely be the new owner of the files. So when the user logs on they will get an error attempting to load the profile unless the following policy is set:

    Computer Configuration \ Administrative Templates \ System \ User Profiles \ Do not check for user ownership of Roaming Profile Folders

    Next we perform the same checks to see if it is a mandatory profile or not then start reconciling the roaming profile with the locally cached profile if one exists. One thing to keep in mind is that all changes made to the profile are done locally and are not propagated to the server share until the user logs off.

    I hope these two parts help you in understanding how to read a Userenv log. By no means will you be an expert in reading one the first time. It takes practice and looking at many of them to determine what is happening or not happening as the case may be. Hopefully it will give you a better understanding of how the OS is designed and what is actually occurring during the profile load.

    For more information check out some of the following articles:

    Interpreting Userenv Log Files
    http://technet.microsoft.com/en-us/library/cc786775.aspx

    Group Policy Wiki
    http://grouppolicy.editme.com

    250842 Troubleshooting Group Policy application problems
    http://support.microsoft.com/kb/250842

    221833 How to enable user environment debug logging in retail builds of Windows
    http://support.microsoft.com/kb/221833

    - Mark Ramey

  • Understanding How to Read a Userenv Log – Part 1

    Hey everyone, this is Mark from the Directory Services team. We get calls all the time where enabling Userenv logging is necessary to see exactly what is happening with group policy and profile loading. If you have ever looked at one it can be confusing to say the least. One thing to remember is that if the logging is not enabled then do not try and interpret the log since very minimal logging is enabled by default! I am going to break this down into 2 parts with one being the Computer side and its processing then the second part of the User side and its processing. So let’s jump in and get our feet wet with some of the basics.

    First, what is Userenv logging? This is short for “User Environment.” How do I enable the logging? You can reference the following KB article:

    221833 How to enable user environment debug logging in retail builds of Windows
    http://support.microsoft.com/kb/221833

    Note that Userenv logging per this article does not work on any version of Windows Vista or Windows Server 2008. It will work only on Windows 2000, 2003 or XP.

    Open Regedit on the problem computer and drill down to:

    HKLM\Software\Microsoft\Windows NT\CurrentVersion\Winlogon.

    Create a REG_DWORD with the value called UserEnvDebugLevel then set the value to 0x10002 in hexadecimal. The value is not case sensitive.

    Logging will start immediately to the Userenv.log file located in the %SystemRoot%\Debug\UserMode folder (no reboot or restart of services is required). If the Userenv.log file is larger than 300 KB, the file is renamed Userenv.bak, and a new Userenv.log file is created. This action occurs when a user logs on locally or by using Terminal Services, and the Winlogon process starts. However, because the size check only occurs when a user logs on, the Userenv.log file may grow beyond the 300 KB limit. If you need to read the log or .bak files then you can simply open them with Notepad. Since you want to see what the computer is doing when it starts, reboot the client computer.

    One problem with Userenv logging, especially on a busy terminal server with lots of logon activity, is that the log is overwritten before you get a chance to find the useful information in it. While there is no way to increase the 300 KB limit on the log file, if you make Userenv.bak read-only, Winlogon can’t rename Userenv.log to Userenv.bak, so it just keeps logging to the Userenv.log indefinitely. If you decide to use this method, it is critical that you monitor the size of the Userenv.log to make sure it does not fill up the drive. Then remove the read-only attribute as soon as you are done troubleshooting.

    After a reboot and once you are logged onto the client computer, open the Userenv log; you should notice information such as this:

    USERENV(78c.790) 22:00:04:218 LoadUserProfile: lpProfileInfo->lpUserName = <NetworkService>

    Or

    USERENV(78c.790) 22:00:04:546 LoadUserProfile: lpProfileInfo->lpUserName = <LocalService>

    This is normal as there are profiles for the Network Service and Local Service. These accounts have to logon just as a normal user account since they are used to start services that are running on the local computer. You may see a line in the log such as this:

    USERENV(78c.790) 22:00:04:515 GetUserDNSDomainName: Domain name is NT Authority. No DNS domain name available.

    No need to worry if it is associated with one of the services logging on. This is due to if you look at the SID (Security Identifier) for either the Network Service (S-1-5-20) or the Local Service (S-1-5-19) these are what we call Well Known Security Identifiers (SID’s). You can see a list of them in:

    243330 Well-known security identifiers in Windows operating systems
    http://support.microsoft.com/kb/243330

    These accounts are not associated to a domain such as a user or computer account would be. So these can be ignored. Once the services have started up or still in the process of starting you will see the following:

    USERENV(750.280) 22:00:43:203 ProcessGPOs: Starting computer Group Policy (Background) processing...

    This signifies that the group policy for the computer is about to start processing. First we ping the server to determine if we are on a fast link or not. We do this to determine if we are on a slow link where processing group policies could be done in a different fashion. So if ICMP is not allowed through any routers and in order to ping a DC, if we go through a router, then do not expect policies to get applied.

    USERENV(750.280) 22:00:43:203 PingComputer: PingBufferSize set as 2048
    USERENV(750.280) 22:00:43:203 PingComputer: Adapter speed 100000000 bps
    USERENV(750.280) 22:00:43:203 PingComputer: First time: 0
    USERENV(750.280) 22:00:43:203 PingComputer: Fast link. Exiting.

    Next we have to determine where this computer account resides in Active Directory.

    USERENV(750.280) 22:00:53:953 ProcessGPOs: User name is: CN=Machine,OU=Workstations,OU=TX,OU=USA,DC=Domain,DC=com, Domain name is: Domain
    USERENV(750.280) 22:00:53:953 ProcessGPOs: Domain controller is: \\DC1.DOMAIN.COM Domain DN is DOMAIN.COM

    Notice that we print out the LDAP path to the computer account. This is done so that we know where to check for the policies at the OU levels. It will also report which DC that the computer has contacted in order to pull the policies from.

    Next the client side extensions are checked where the GUID between the {} will be different.

    USERENV(750.280) 22:00:53:968 ReadExtStatus: Reading Previous Status for extension {25537BA6-77A8-11D2-9B6C-0000F8080861}

    Once we have all that we start checking for policies that might be linked at an OU level, domain level or site level . We first start at the OU that the computer is located at or if the computer lies in a container the next level up that a policy can be linked at.

    USERENV(750.280) 22:00:54:000 GetGPOInfo: Server connection established.
    USERENV(750.280) 22:00:54:031 GetGPOInfo: Bound successfully.
    USERENV(750.280) 22:00:54:046 SearchDSObject: Searching <OU=Workstations,OU=TX,OU=USA,DC=Domain,DC=com >
    USERENV(750.280) 22:00:54:046 SearchDSObject: No GPO(s) for this object.
    USERENV(750.280) 22:00:54:046 SearchDSObject: Searching < OU=TX,OU=USA,DC=Domain,DC=com >
    USERENV(750.280) 22:00:54:046 SearchDSObject: Found GPO(s): <[LDAP://CN={Policy GUID},CN=Policies,CN=System, DC=Domain,DC=com;0]>
    USERENV(750.280) 22:00:54:046 SearchDSObject: Searching < OU=USA,DC=Domain,DC=com >
    USERENV(750.280) 22:00:54:046 SearchDSObject: Searching < DC=Domain,DC=com >
    USERENV(750.280) 22:00:54:046 SearchDSObject: Searching <CN=MYSite,CN=Sites,CN=Configuration, DC=Domain,DC=com >

    At any point in this time the log may show something like this:

    USERENV(750.280) 22:00:54:046 SearchDSObject: Found GPO(s):

    Then the log shows a list of GPO’s that are linked to that particular OU/Domain/Site. Note that you could run across this in the output:

    USERENV(750.280) 22:00:54:046 SearchDSObject: Found GPO(s): <[LDAP://CN={Policy GUID},CN=Policies,CN=System, DC=Domain,DC=com;0]>

    Note the 0 at the end, this is the default setting, 1 means the policy is set to disabled which means the policy is linked to that particular OU / domain or site level but is disabled. If the value is set to 2 this means that the policy has been set to “No Override” also known as “Enforced.” A setting of “No Override” means that if 2 separate GPOs have the same setting are defined but hold different values, the one that is set to no override wins out on what is actually applied to the client. If a policy is set to “No Override” or “Enforced” at an OU/domain level then an OU below that is set to block inheritance, the policy set for “No Override” will still apply. You cannot block a policy from applying if the “No Override” or “Enforce” has been set on a policy.

    Once we have gone all the way through we start actually applying the policies:

    USERENV(750.280) 22:00:54:093 ProcessGPO: Searching <CN={GPO GUID},CN=Policies,CN=System, DC=Domain,DC=com >

    The path above is where in AD the particulars of the policy are stored.

    USERENV(750.280) 22:00:54:093 ProcessGPO: Machine has access to this GPO.

    We need to check and see if the computer has access to the policy if we do then the computer can apply the policy; if we do not have access then we cannot apply it.

    USERENV(750.280) 22:00:54:109 FilterCheck: Found WMI Filter id of: <[DOMAIN.COM;{Policy GUID};0]>

    If a policy has a WMI filter being applied we have to check it and see if the policy is going to apply to this computer or not. The WMI filter can be seen if you go to the properties of the policy then to the WMI tab. If you are using GPMC then this can be found in the right hand pane at the very bottom box after highlighting the policy.

    USERENV(750.280) 22:00:54:093 ProcessGPO: GPO passes the filter check.

    Or

    USERENV(750.280) 22:00:55:250 ProcessGPO: The GPO does not pass the filter check and so will not be applied.
    USERENV(750.280) 22:00:54:093 ProcessGPO: Found functionality version of: 2

    Functionality version has to be a 2 for a Windows 2000 or later OS to apply the policy.

    USERENV(750.280) 22:00:54:093 ProcessGPO: Found file system path of: \\DOMAIN.com\SysVol\DOMAIN.com\Policies\{GPO GUID}

    This above is the path of the policy in the file system on a DC.

    USERENV(750.280) 22:00:54:109 ProcessGPO: Found common name of: <{GPO GUID}>
    USERENV(750.280) 22:00:54:109 ProcessGPO: Found display name of: <MY Domain Policy>
    USERENV(750.280) 22:00:54:109 ProcessGPO: Found machine version of: GPC is 77, GPT is 77

    Here is where we check the GPC (Group Policy Container) and the GPT (Group Policy Template) for the version numbers. We check the version numbers to determine if the policy has changed since the last time it was applied. If the version numbers are different then we either have an AD replication or File replication latency problem.

    USERENV(750.280) 22:00:54:109 ProcessGPO: Found flags of: 0
    USERENV(750.280) 22:00:54:109 ProcessGPO: Found extensions: [{35378EAC-683F-11D2-A89A-00C04FBBCFA2}{53D6AB1D-2488-11D1-A28C-00C04FB94F17}][{827D319E-6EAC-11D2-A4EA-00C04F79F83A}{803E14A0-B4FB-11D0-A0D0-00A0C90F574B}][{B1BE8D72-6EAC-11D2-A4EA-00C04F79F83A}{53D6AB1D-2488-11D1-A28C-00C04FB94F17}]

    The extensions above refer to the CSE (client-side extensions) and will vary from policy to policy. For a list of the CSE’s see:

    216357 Identifying Group Policy Client-Side Extensions
    http://support.microsoft.com/kb/216357

    943729 Information about new Group Policy preferences in Windows Server 2008
    http://support.microsoft.com/kb/943729

    Moving right along now we come to the part where we process CSE’s for particular settings such as Folder Redirection, Disk Quota, etc. If the particular extension is not being used then you can simply ignore this section.

    USERENV(750.280) 22:00:56:359 ProcessGPOs: Processing extension Folder Redirection
    USERENV(750.280) 22:00:56:359 CompareGPOLists: The lists are the same.
    USERENV(750.280) 22:00:56:359 CheckGPOs: No GPO changes but couldn't read extension Folder Redirection's status or policy time.
    USERENV(750.280) 22:00:56:359 ProcessGPOs: Extension Folder Redirection skipped with flags 0x7.

    Now we have pretty much come to the end of the computer processing which we know by the following:

    USERENV(750.280) 22:00:56:390 SetFgRefreshInfo: Previous Machine Fg policy Synchronous, Reason: SyncPolicy.
    USERENV(750.280) 22:00:56:390 SetFgRefreshInfo: Next Machine Fg policy Synchronous, Reason: SyncPolicy.
    USERENV(750.280) 22:00:56:390 ProcessGPOs: No WMI logging done in this policy cycle.
    USERENV(750.280) 22:00:56:390 LeaveCriticalPolicySection: Critical section 0x6d0 has been released.
    USERENV(750.280) 22:00:56:390 ProcessGPOs: Computer Group Policy has been applied.
    USERENV(750.280) 22:00:56:390 ProcessGPOs: Leaving with 1.
    USERENV(750.280) 22:00:56:390 ApplyGroupPolicy: Leaving successfully.
    USERENV(750.860) 22:00:56:640 GPOThread: Next refresh will happen in 115 minutes

    This is telling us that we will be refreshing the policy in the foreground and the next one will occur the same way in 115 minutes and we are leaving successfully.

    Now for a note about the next blog for the user processing, no one runs a pure Microsoft environment and almost every Userenv log I have ever seen had something in the order of this showing up:

    USERENV(6e4.6d4) 22:00:56:890 GetUserNameAndDomain Failed to impersonate user
    USERENV(6e4.6d4) 22:00:56:890 GetUserDNSDomainName: Domain name is NT Authority. No DNS domain name available.

    So what is this? Note that the very first of all the sample lines that I have added so far has started with:

    USERENV(750.280), the first number, 750, is the process identifier (PID) and the second number is the thread identifier (TID). You can open Calc and change the View to Scientific. Change the type to Hex then enter 750 and hit the Decimal radio button. Now you have a number of 1872, this is the PID for the process. Open Task Manager, go to View – Select Columns and put a check mark in the box for PID and hit OK. Now in the column next to the Image Name will be the PID. Sort the PID’s and look for the process associated with the number 1872 and that is the process that the Userenv logging is referring to. In this case the PID of 1872 was Winlogon.exe. If we do the same for the problem PID (6e4) we get 1764. That PID resolved to a third-party service that was causing problems.

    I hope this helps in understanding how to read part of the Userenv logging that we can enable to help in determining issues when booting up, applying group policies, etc. I will complete the User processing in Part 2. Also, you can find additional information on this topic at Interpreting Userenv Log Files on TechNet.

    - Mark Ramey

  • DFSR Does Not Replicate Temporary Files

    Note that this post has been added to the TechNet Wiki to allow for community editing.

    If you notice that DFS Replication (DFSR) is not replicating certain files, one simple reason is that the temporary attribute is set on them. By design, DFSR does not replicate files if they have the temporary attribute set on them, and it cannot be configured to do so.

    This may not be obvious because nearly all the normal methods you would use in Windows to check file attributes do not show the temporary attribute. Specifically, all of the following do not show the temporary attribute - Attrib.exe, Explorer's file properties, FileSystemObject in Windows Scripting Host, and CIM_Datafile in WMI. Also, DFSR does not log any errors to the event log or to the debug logs to show temporary files are not being replicated. There is a relevant entry in the debug logs, but it is not an error because this behavior is by design.

    The reason DFSR does not replicate files with the temporary attribute set is that they are considered short-lived files that you would never actually want to replicate. Using the temporary attribute on a file keeps that file in memory and saves on disk I/O. Therefore applications can use it on short-lived files to improve performance.

    An application can use FILE_ATTRIBUTE_TEMPORARY when calling the CreateFile function if they want a temporary file. But an even better way is to also specify FILE_FLAG_DELETE_ON_CLOSE so the temporary file is deleted when all handles are closed. That way you get the performance benefit of a temporary file (it’s kept in memory) and it is removed when handles are closed so administrators don’t come along and wonder why DFSR isn’t replicating it.

    If you have temporary files that you want DFSR to replicate, you may think it is enough to just remove the temporary attribute on those files and be on your way. And you can do that. But since you got in this situation once, it is likely you still have an application that will come right back and create more temporary files. So you need to get at the crux of the issue – why do you want to replicate files that an application is specifically creating to be temporary? Either the application must change its behavior, or you must except that temporary files won’t be replicated, because there is no way to make DFSR replicate files as long as the temporary attribute is set on them.

    Checking the Temporary Attribute on a File using Fsutil

    But wait, you say, maybe I don’t even know yet if these files that aren’t replicating are temporary! So let’s find out. As mentioned before, almost none of the ways to check attributes in Windows will actually show the temporary attribute. But there is one that does – the handy Fsutil tool that is included in Windows.

    fsutil usn readdata c:\data\test.txt

    Major Version : 0x2
    Minor Version : 0x0
    FileRef# : 0x0021000000002350
    Parent FileRef# : 0x0003000000005f5e
    Usn : 0x000000004d431000
    Time Stamp : 0x0000000000000000 12:00:00 AM 1/1/1601
    Reason : 0x0
    Source Info : 0x0
    Security Id : 0x5fb
    File Attributes : 0x120

    File Name Length : 0x10
    File Name Offset : 0x3c
    FileName : test.txt

    File Attributes is a bitmask that indicates which attributes are set. In the above example, 0x120 indicates the temporary attribute is set because that is 0x100 and 0x20 (Archive) = 0x120.

    Here are the possible values:

    READONLY 0x1
    HIDDEN 0x2
    SYSTEM 0x4
    DIRECTORY 0x10
    ARCHIVE 0x20
    DEVICE 0x40
    NORMAL 0x80
    TEMPORARY 0x100
    SPARSE_FILE 0x200
    REPARSE_POINT 0x400
    COMPRESSED 0x800
    OFFLINE 0x1000
    NOT_CONTENT_INDEXED 0x2000
    ENCRYPTED 0x4000

    You combine the values to come up with the File Attributes bitmask value.

    If you need a sanity check:

    1. Start, Run, Calc.
    2. Change to Hex and paste in the File Attributes value from the Fsutil command. Say for example, 4925.
    3. Hit the And button, then type 100.
    4. Hit equals and if it returns 100, then the temporary attribute is set. If it returns 0, the temporary attribute is not set.

    Checking for Temporary Files in the Debug Logs with Findstr

    Another way to check if files are not replicating because they have the temporary attribute set is to use Findstr (included in Windows) to look for the FILE_ATTRIBUTE_TEMPORARY text string in the DFSR debug logs.

    First you need to extract out all of the debug logs, because all except the active log will be compressed, as indicated by a .GZ extension. The DFSR debug logs (Dfsr*.log and Dfsr*.log.gz) reside by default under %windir%\debug. All the popular compression tools such as Winzip and Winrar can handle .GZ compression.

    Let’s say you extracted the debug logs to C:\Logs. You can then run the following Findstr command to look for temporary files.

    Findstr FILE_ATTRIBUTE_TEMPORARY c:\logs\dfsr*.log

    That will output the entire line for every line in the debug log that contains a match to that string. If it doesn't find any matches, it will return to a prompt and not show anything.

    Sample output from a matching entry:

    C:\WINDOWS\debug\Dfsr00018.log:20080903 16:14:29.390 1808 USNC 1204 UsnConsumer::ProcessUsnRecord Skipping USN_RECORD with FILE_ATTRIBUTE_TEMPORARY flag:

    If it does find any matches, you can then open the specified log file, search on the string FILE_ATTRIBUTE_TEMPORARY (Ctrl+F or Edit | Find in Notepad) and then you will see the actual file name for the file that was skipped because the temporary attribute is set on it.

    Removing the Temporary Attribute from Multiple Files with Powershell

    So you figured out that DFSR is not replicating some files because they have the temporary attribute set. There is no way to change this behavior in DFSR, so the only option is to live with it, or remove the temporary attribute from the files you want to replicate. An application in your environment has created these temporary files, so just treating the symptom isn’t enough, you need to find the application that creates them and either change its behavior, or accept that those files will not be replicated.

    Since Attrib is not aware of the temporary attribute, we need to go to greater lengths to remove it. First you need to have Powershell installed on the machine - www.microsoft.com/powershell

    Then bring up a Powershell prompt (Start, Run, Powershell or from the Programs menu) and run this command to remove the temporary attribute from all files in the specified directory, including subdirectories (in this example, D:\Data):

    Get-childitem D:\Data -recurse | ForEach-Object -process {if (($_.attributes -band 0x100) -eq 0x100) {$_.attributes = ($_.attributes -band 0xFEFF)}}

    If you don’t want it to work against subdirectories just remove the -recurse parameter.

    Additional Information

    Win32 CreateFile Function
    http://msdn.microsoft.com/en-us/library/aa363858(VS.85).aspx

    It’s Only Temporary
    http://blogs.msdn.com/larryosterman/archive/2004/04/19/116084.aspx

    103237 INFO: Using Temporary Files Can Improve Application Performance
    http://support.microsoft.com/kb/103237

    131324 Use New Flags to Speed Up C Run-time Low-Level I/O Functions
    http://support.microsoft.com/kb/131324

    Designing Distributed File Systems (see Files Not Replicated under DFSR Replication Compatibility)
    http://technet.microsoft.com/en-us/library/cc772778.aspx

    - Craig Landis

  • New KB Articles 11/2-11/9

    New KB articles related to Directory Services for the week of 11/2-11/9.

    950234

     

    Error message when you try to open a shared file in Windows Explorer  on a Windows XP-based computer: "<SharePath> is not accessible. Access is denied"

    933071

    The "Effective Permissions" tab may report incorrect permissions in Windows Server 2003

    955457

     

    It takes longer to log on to a Windows Server 2003-based terminal server than expected when you enable Group Policy preferences in the terminal server

    894270

     

    The error message of the Reg command may be truncated when you redirect a standard error message to a file on a Windows XP Service Pack 2-based computer

    958285

     

    Error message when you access the shared resources in a cooperation network over a VPN connection from a Windows Vista or Windows Server 2008-based client computer: "0x80004005" or "0x80070035"

    943790

    A new version of the File system API that provides discovery and restoration for deleted files in Windows Vista is available

    951710

     

    You receive error messages or warnings when you change an Active Directory schema so that the Company property supports more than 64 characters