How to troubleshoot Active Directory replication error 5 in Windows Server: Access is denied
This article describes the symptoms, cause, and resolution of situations in which Active Directory replication fails with error 5: Access is denied.
Original KB number: 3073945
Symptoms
You might encounter one or more of the following symptoms when Active Directory replications fail with error 5.
Symptom 1
The Dcdiag.exe command-line tool reports that the Active Directory replication test fails with error status code (5). The report resembles the following example:
Testing server: <Site_Name>\<Destination_DC_Name>
Starting test: Replications
Replications Check
[Replications Check,<Destination_DC_Name>] A recent replication attempt failed:
From <Source_DC> to <Destination_DC>
Naming Context: <Directory_Partition_DN_Path>
The replication generated an error (5):
Access is denied.
The failure occurred at <Date> <Time>.
The last success occurred at <Date> <Time>.
<Number> failures have occurred since the last success.
Symptom 2
The Dcdiag.exe command-line tool reports that the DsBindWithSpnEx
function fails with error 5 by running the DCDIAG /test:CHECKSECURITYERROR
command.
Symptom 3
The REPADMIN.exe command-line tool reports that the last replication attempt failed with status 5.
The REPADMIN
commands that frequently cite the 5 status include but aren't limited to the following ones:
REPADMIN /KCC
REPADMIN /REPLICATE
REPADMIN /REPLSUM
REPADMIN /SHOWREPL
REPADMIN /SHOWREPS
REPADMIN /SYNCALL
Sample output from the REPADMIN /SHOWREPL
command is as follows. This output shows incoming replication from DC_2_Name
to DC_1_Name
failing with the "Access is denied" error.
<Site_Name>\<DC_1_Name>
DSA Options: IS_GC
Site Options: (none)
DSA object GUID: <GUID>
DSA invocationID: <invocationID>
==== INBOUND NEIGHBORS======================================
DC= <DomainName>,DC=com
<Site_Name>\<DC_2_Name> via RPC
DSA object GUID: <GUID>
Last attempt @ <Date> <Time> failed, result 5(0x5):
Access is denied.
<#> consecutive failure(s).
Last success @ <Date> <Time>.
Symptom 4
NTDS KCC, NTDS General, or Microsoft-Windows-ActiveDirectory_DomainService events with the 5 status are logged in the Directory Services log in Event Viewer.
The following table summarizes Active Directory events that frequently cite the 8524 status, including but not limited to:
Event ID | Source | Event string |
---|---|---|
1655 | NTDS General | Active Directory tried to communicate with the following global catalog and the attempts were unsuccessful. |
1925 | NTDS KCC | The attempt to establish a replication link for the following writable directory partition failed. |
1926 | NTDS KCC | The attempt to establish a replication link to a read-only directory partition with the following parameters failed. |
Symptom 5
When you right-click the connection object from a source domain controller (DC) in Active Directory Sites and Services and then select Replicate Now, the process fails, and you receive the following error:
Dialog title text: Replicate Now
Dialog message text:
The following error occurred during the attempt to synchronize naming context %<directory partition name>% from Domain Controller <Source DC> to Domain Controller <Destination DC>: Access is denied.
The operation will not continue.
The following screenshot represents a sample of the error:
Workaround
Use the generic DCDIAG and NETDIAG
command-line tools to run multiple tests. Use the DCDIAG /TEST:CheckSecurityError
command-line tool to perform specific tests. (These tests include an SPN registration check.)
To work around this issue, follow these steps:
- At a command prompt, run
DCDIAG
on the destination domain controller. - Run
DCDIAG /TEST:CheckSecurityError
andNETDIAG
. - Resolve any faults that were identified by
DCDIAG
. - Retry the previously failed replication operation. If replications continue to fail, see the following causes and solutions.
The following causes may result in error 5. Some of them have solutions.
Cause 1: There's an invalid security channel or password mismatch on the source or destination DC
Validate the security channel by running one of the following commands:
nltest /sc_query:<Domain Name>
netdom verify <DC Name>
On condition, reset the destination domain controller's password by using NETDOM /RESETPWD
.
Solution
Disable the Kerberos Key Distribution Center (KDC) service on the destination domain controller.
From an elevated command prompt on the destination domain controller, obtain the system's Kerberos ticket by running
Klist -li 0x3e7 purge
.Run
NETDOM RESETPWD
to reset the password of the remote domain controller:c:\>netdom resetpwd /server:<remote_dc_name> /userd: domain_name\administrator /passwordd: administrator_password
Make sure that the potential KDCs and the source domain controller (if they're in the same domain) inbound replicate the destination domain controller's new password.
Start the KDC service on the destination domain controller and retry the replication operation.
For more information, see How to use Netdom.exe to reset machine account passwords of a domain controller.
Cause 2: The "CrashOnAuditFail" setting in the registry of the destination DC has a value of "2"
A CrashOnAuditFail
value of 2
is triggered if the Audit: Shut down system immediately if unable to log security audits policy setting in Group Policy is enabled and the local security event log is full.
Active Directory domain controllers are especially prone to maximum-capacity security logs when auditing is enabled and the size of the security event log is constrained by the Do not overwrite events (clear log manually) and Overwrite as needed options in Event Viewer or their Group Policy equivalents.
Solution
Important
This section, method, or task contains steps that tell you how to modify the registry. However, serious problems might occur if you modify the registry incorrectly. Therefore, make sure that you follow these steps carefully. For protection, back up the registry before you modify it so that you can restore it if a problem occurs. For more information about how to back up and restore the registry, see How to back up and restore the registry in Windows.
Clear the security event log and save it to another location as required.
Reevaluate any size constraints on the security event log. This includes policy-based settings.
Delete and then re-create a
CrashOnAuditFail
registry entry as follows:Registry subkey:
HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\LSA
Value Name:CrashOnAuditFail
Value Type:REG_DWORD
Value Data:1
Restart the destination domain controller.
On seeing a CrashOnAuditFail
value of 0
or 1
, some CSS engineers have resolved "Access is denied" errors by clearing the security event log again, deleting the CrashOnAuditFail
registry value, and rebooting the destination domain controller.
For more information, see Manage auditing and security log.
Cause 3: Invalid trust (the secure channel on the source or destination DC is invalid, or trust relationships in the trust chain are broken or invalid)
If Active Directory replication fails between domain controllers in different domains, you should verify the health of trust relationships along the trust path.
You can try the NetDiag Trust Relationship test to check for broken trusts. The Netdiag.exe utility identifies broken trusts by displaying the following text:
Trust relationship test. . . . . . : Failed
Test to ensure DomainSid of domain '<domainname>' is correct.
[FATAL] Secure channel to domain '<domainname>' is broken.
[% <variable status code> %]
For example, if you have a multi-domain forest that contains a root domain (Contoso.COM
), a child domain (B.Contoso.COM
), a grandchild domain (C.B.Contoso.COM
), and a tree domain in same forest (Fabrikam.COM
) and if replication is failing between domain controllers in the grandchild domain (C.B.Contoso.COM
) and the tree domain (Fabrikam.COM)
, you should verify trust health between C.B.Contoso.COM
and B.Contoso.COM
, between B.Contoso.COM
and Contoso.COM
, and then finally between Contoso.COM
and Fabrikam.COM
.
If a shortcut trust exists between the destination domains, you don't have to validate the trust path chain. Instead, you should validate the shortcut trust between the destination and source domain.
Check for recent password changes to the trust by running the following command:
Repadmin /showobjmeta * <DN path for TDO in question>
Verify that the destination domain controller is transitively inbound replicating the writable domain directory partition where trust password changes may take effect.
Commands to reset trusts from the root domain PDC are as follows:
netdom trust <Root Domain> /Domain:<Child Domain> /UserD:CHILD /PasswordD:* /UserO:ROOT /PasswordO:* /Reset /TwoWay
Commands to reset trusts from the child domain PDC are as follows:
netdom trust <Child Domain> /Domain:<Root Domain> /UserD:Root /PasswordD:* /UserO:Child /PasswordO:* /Reset /TwoWay
Cause 4: Excessive time skew
There's a time difference between the KDC used by the destination DC and the source DC. The time difference exceeds the maximum time skew that's allowed by Kerberos defined in the Default Domain policy.
Kerberos policy settings in the default domain policy allow for a five-minute difference in system time (this is the default value) between KDC domain controllers and Kerberos target servers to prevent replay attacks. Some documentation states that the system time of the client and that of the Kerberos target must be within five minutes of one another. Other documentation states that, in the context of Kerberos authentication, the time that is important is the delta between the KDC that is used by the caller and the time on the Kerberos target. Also, Kerberos doesn't care whether the system time on the relevant domain controllers matches current time. It cares only that the relative time difference between the KDC and target domain controller is within the maximum time skew that Kerberos policy allows. (The default time is five minutes or less.)
In the context of Active Directory operations, the target server is the source domain controller that is contacted by the destination domain controller. Every domain controller in an Active Directory forest that is currently running the KDC service is a potential KDC. Therefore, you have to consider time accuracy on all other domain controllers against the source domain controller. This includes time on the destination domain controller itself.
You can use the following two commands to check time accuracy:
DCDIAG /TEST:CheckSecurityError
W32TM /MONITOR
You can find sample output from DCDIAG /TEST:CheckSecurityError
in the "More information" section. This sample shows excessive time skew on domain controllers.
Look for LSASRV 40960 events on the destination domain controller when the replication request fails. Look for events that cite a GUID in the CNAME record of the source domain controller with the extended error 0xc000133. Look for events that resemble the following one:
0xc000133: The time at the Primary Domain Controller is different than the time at the Backup Domain Controller or member server by too large an amount
Network traces that capture the destination computer that connects to a shared folder on the source domain controller (and also other operations) may show the "An extended error has occurred" on-screen error, but a network trace displays the following information:
-> KerberosV5 KerberosV5:TGS Request Realm: <- TGS request from source DC
<- Kerberosvs Kerberosvs:KRB_ERROR - KRB_AP_ERR_TKE_NVV (33) <- TGS response where "KRB_AP_ERR_TKE_NYV
<- maps to "Ticket not yet valid" <- maps to "Ticket not yet valid"
The TKE_NYV
response indicates that the date range on the TGS ticket is newer than the time on the target. This indicates excessive time skew.
Note
W32TM /MONITOR
checks time only on domain controllers in the test computers domain, so you have to run this in each domain and compare time between the domains.- If the system time was found to be inaccurate, try to figure out why and what can be done to prevent inaccurate time in the future. Was the forest root PDC configured with an external time source? Are reference time sources online and available on the network? Was the time service running? Are DC role computers configured to use NT5DS hierarchy to source time?
- When the time difference is too large on destination domain controllers, the Replicate now command in DSSITE.MSC fails with the "There is a time and/or date difference between the client and the server" on-screen error. This error string maps to error 1398 (decimal) or 0x576 (hexadecimal) with the ERROR_TIME_SKEW symbolic error name.
For more information, see Setting Clock Synchronization Tolerance to Prevent Replay Attacks.
Cause 5: The "RestrictRemoteClients" setting in the registry has a value of "2"
If the Restrictions for Unauthenticated RPC clients policy setting is enabled and set to Authenticated without exceptions, the RestrictRemoteClients
registry value is set to a value of 0x2
in the HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\Windows NT\RPC
registry subkey.
This policy setting enables only authenticated remote procedure call (RPC) clients to connect to RPC servers that are running on the computer to which the policy setting is applied. It doesn't allow for exceptions. If you select this option, the system can't receive remote anonymous calls by using RPC. This setting should never be applied to a domain controller.
Solution
Disable the Restrictions for Unauthenticated RPC clients policy setting that restricts the
RestrictRemoteClients
registry value to2
.Note
The policy setting is located in the following path:
Computer Configuration\Administrative Templates\System\Remote Procedure Call\Restrictions for Unauthenticated RPC clients.
Delete the
RestrictRemoteClients
registry setting, and then restart.
For more information, see Restrictions for Unauthenticated RPC Clients: The group policy that punches your domain in the face and RestrictRemoteClients registry key is enabled.
Cause 6: The "Access this computer from network" user right isn't granted to the "Enterprise Domain Controllers" group or a user who triggered the replication
In a default installation of Windows, the default domain controller policy is linked to the domain controller's organization unit (OU). The OU grants the Access this computer from network user right to the following security groups:
Local policy | Default domain controller policy |
---|---|
Administrators | Administrators |
Authenticated Users | Authenticated Users |
Everyone | Everyone |
Enterprise Domain Controllers | Enterprise Domain Controllers |
[Pre-Windows 2000 compatible Access] | Pre-Windows 2000 compatible Access |
If Active Directory operations fail with error 5, you should verify the following points:
Security groups in the table are granted the Access this computer from network user right in the default domain controller's policy.
Domain controller computer accounts are located in the domain controller's OU.
The default domain controller's policy is linked to the domain controller's OU or to alternative OUs that are hosting the domain controller computer accounts.
Group Policy is applied on the destination domain controller that currently logs error 5.
The Deny access this computer from network user right is enabled or doesn't reference direct or transitive groups that the security context being used by the domain controller or user account triggering the replication.
Policy precedence, blocked inheritance, Microsoft Windows Management Instrumentation (WMI) filtering, or the like, isn't preventing the policy setting from applying to domain controller role computers.
Note
- Policy settings can be validated with RSOP.MSC, but
GPRESULT
is the preferred tool because it's more accurate, for example,GPRESULT /H c:\temp\GPOResult.html
orGPRESULT /Z
. - Local policy takes precedence over policies defined in sites, domains, and OUs.
- At one time, it was common for administrators to remove the Enterprise Domain Controllers and Everyone groups from the Access this computer from network policy setting in the default domain controller's policy. However, removing both groups is fatal. There's no reason to remove Enterprise domain controllers from this policy setting, because only domain controllers are members of this group.
Cause 7: There's an SMB signing mismatch between the source and destination DCs
The best compatibility matrix for SMB signing is defined by four policy settings and their registry-based equivalents:
Policy setting | Registry path |
---|---|
Microsoft network client: Digitally sign communications (if server agrees) | HKEY_LOCAL_MACHINE\SYSTEM\CCS\Services\Lanmanworkstation\Parameters\Enablesecuritysignature |
Microsoft network client: Digitally sign communications (always) | HKEY_LOCAL_MACHINE\SYSTEM\CCS\Services\Lanmanworkstation\Parameters\Requiresecuritysignature |
Microsoft network server: Digitally sign communications (if server agrees) | HKEY_LOCAL_MACHINE\SYSTEM\CCS\Services\Lanmanserver\Parameters\Enablesecuritysignature |
Microsoft network server: Digitally sign communications (always) | HKEY_LOCAL_MACHINE\SYSTEM\CCS\Services\Lanmanserver\Parameters\Requiresecuritysignature |
You should focus on SMB signing mismatches between the destination and source domain controllers. The classic cases involve a setting that is enabled or required on one side but disabled on the other.
Note
Internal testing showed SMB signing mismatches, causing replication to fail with error 1722: "The RPC Server is unavailable."
Cause 8: UDP-formatted Kerberos packet fragmentation
Network routers and switches may fragment or completely drop large User Datagram Protocol (UDP)-formatted network packets that are used by Kerberos and Extension Mechanisms for DNS (EDNS0).
Solution
From the console of the destination domain controller, ping the source domain controller by its fully qualified computer name to identify the largest packet supported by the network route. To do this, run the following command:
c:\>Ping <Source_DC_hostname>.<Fully_Qualified_Computer_Name> -f -l 1472
If the largest non-fragmented packet is less than 1,472 bytes, try one of the following methods (in order of preference):
- Change the network infrastructure to appropriately support large UDP frames. This may require a firmware upgrade or configuration change on routers, switches, or firewalls.
- Set maxpacketsize (on the destination domain controller) to the largest packet identified by the
PING -f -l
command less 8 bytes to account for the TCP header, and then restart the changed domain controller. - Set maxpacketsize (on the destination domain controller) to a value of
1
. This triggers Kerberos authentication to use a TCP. Restart the changed domain controller to make the change take effect.
Retry the failed Active Directory operation.
Cause 9: Network adapters have the "Large Send Offload" feature enabled
Solution
- On the destination domain controller, open network adapter properties.
- Select the Configure button.
- Select the Advanced tab.
- Disable the IPv4 Large Send Offload property.
- Restart the domain controller.
Cause 10: Invalid Kerberos realm
The Kerberos realm is invalid if one or more of the following conditions are true:
- The
KDCNames
registry entry incorrectly contains the local Active Directory domain name. - The
PolAcDmN
registry key and thePolPrDmN
registry key don't match.
Solutions
Important
This section, method, or task contains steps that tell you how to modify the registry. However, serious problems might occur if you modify the registry incorrectly. Therefore, make sure that you follow these steps carefully. For protection, back up the registry before you modify it so that you can restore it if a problem occurs. For more information about how to back up and restore the registry, see How to back up and restore the registry in Windows.
Solution to the incorrect "KDCNames" registry entry
- On the destination domain controller, run
REGEDIT
. - Locate the following subkey in the registry:
HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\Lsa\Kerberos\Domains
- For each <Fully_Qualified_Domain> under the subkey, verify that the value of the
KdcNames
registry entry refers to a valid external Kerberos realm and not the local domain or another domain in the same Active Directory forest.
Solution to the mismatched "PolAcDmN" and "PolPrDmN" registry keys
Start Registry Editor.
In the navigation pane, expand Security.
On the Security menu, select Permissions to grant the Administrators local group full control of the
SECURITY
hive and its child containers and objects.Locate the following subkey in the registry:
HKEY_LOCAL_MACHINE\SECURITY\Policy\PolPrDmN
In the right-side pane of Registry Editor, select the No Name: REG_NONE registry entry one time.
On the View menu, select Display Binary Data.
In the Format section of the dialog, select Byte.
The domain name is displayed as a string on the right side of the Binary Data dialog. The domain name is the same as the Kerberos realm.
Locate the following subkey in the registry:
HKEY_LOCAL_MACHINE\SECURITY\Policy\PolACDmN
In the right-side pane of Registry Editor, double-click the No Name: REG_NONE entry.
In the Binary Editor dialog, paste the value from the
PolPrDmN
registry subkey. (The value from thePolPrDmN
registry subkey is the NetBIOS domain name).Restart the domain controller. If the domain controller isn't functioning correctly, see other methods.
Cause 11: There's a LAN Manager Compatibility (LM Compatibility) mismatch between the source and destination DCs
Cause 12: Service principal names are either not registered or not present because of simple replication latency or a replication failure
Cause 13: Antivirus software uses a mini-firewall network adapter filter driver on the source or destination DC
More information
Active Directory errors and events such as those described in the "Symptoms" section can also fail with error 8453 together with the error string similar to the following one:
Replication Access was denied.
The following situations can cause Active Directory operations to fail with error 8453. However, these situations don't cause failures with error 5.
- The naming context (NC) head isn't permitted with the Replicating Directory Changes permission.
- The security principal starting replication isn't a member of a group that is granted the Replicating Directory Changes permission.
- Flags are missing in the
UserAccountControl
attribute. These flags includeSERVER_TRUST_ACCOUNT
andTRUSTED_FOR_DELEGATION
. - The read-only domain controller (RODC) joined the domain without first running the
ADPREP /RODCPREP
command.
Sample output from "DCDIAG /TEST:CheckSecurityError"
Sample output from DCDIAG /CHECKSECURITYERROR
for a domain controller is as follows. This is caused by excessive time skew.
Doing primary tests
Testing server: <Site_Name>\<Destination_DC_Name>
Starting test: CheckSecurityError
Source DC <Source DC> has possible security error (5). Diagnosing...
Time skew error between client and 1 DCs! ERROR_ACCESS_DENIED or down machine recieved by:
<Source DC>
Source DC <Source DC>_has possible security error (5). Diagnosing...
Time skew error: 7205 seconds different between:.
<Source DC>
<Destination_DC>
[<Source DC>] DsBindWithSpnEx() failed with error 5,
Access is denied..
Ignoring DC <Source DC> in the convergence test of object CN=<Destination_DC>,OU=Domain Controllers,DC=<DomainName>,DC=com, because we cannot connect!
......................... <Destination_DC> failed test CheckSecurityError
Sample output from DCDIAG /CHECKSECURITYERROR
is as follows. It shows the missing SPN names. (The output might vary from environment to environment.)
Doing primary tests
Testing server: <site name>\<dc name>
Test omitted by user request: Advertising
Starting test: CheckSecurityError
* Dr Auth: Beginning security errors check'
Found KDC <KDC DC> for domain <DNS Name of AD domain> in site <site name>
Checking machine account for DC <DC name> on DC <DC Name>
* Missing SPN :LDAP/<hostname>.<DNS domain name>/<DNS domain name>
* Missing SPN :LDAP/<hostname>.<DNS domain name>
* Missing SPN :LDAP/<hostname>
* Missing SPN :LDAP/<hostname>.<DNS domain name>/<NetBIOS domain name>
* Missing SPN :LDAP/bba727ef-be4e-477d-9796-63b6cee3bSf.<forest root domain DN>
* SPN found :E3514235-4B06-I1D1-ABØ4-00c04fc2dcd2/<NTDS Settings object GUID>/<forest root domain DNS name>
* Missing SPN :HOST/<hostname>.<DNS domain name>/<DNS domain name>
* SPN found :HOST/<hostname>.<DNS domain name>
* SPN found :HOST/<hostname>
* Missing SPN :HOST/<hostname>.<DNS domain name>/<NetBIOS domain name>
* Missing SPN :GC/<hostname>.<DNS domain name>/<DNS domain name>
Unable to verify the machine account (<DN path for Dc machine account>) for <DC Name> on <DC name>.
Data collection
If you need assistance from Microsoft support, we recommend you collect the information by following the steps mentioned in Gather information by using TSS for Active Directory replication issues.