Fix Active Directory replication error 1722 (0x6ba, RPC_S_SERVER_UNAVAILABLE): check DNS and the DC GUID CNAME, test port 135 and the dynamic RPC range, then the RPC ClientProtocols key, NIC drivers, UDP fragmentation and SMB signing.
Home › Guides › AD replication error 1722
Last updated 19 November 2026
Error 1722 (0x6ba, RPC_S_SERVER_UNAVAILABLE, "The RPC server is unavailable.") means the RPC client on the destination DC couldn't connect to the RPC server on the source DC. RPC has no special insight into the failure; it maps a lower-layer problem, most often a failed TCP connect, to this one error. So work up from the network:
dcdiag (Replications test):
[Replications Check,<DC Name>] A recent replication attempt failed: From <source DC> to <destination DC> Naming Context: <DN path of directory partition> The replication generated an error (1722): The RPC server is unavailable. ... [<DC name>] DsBindWithSpnEx() failed with error 1722, The RPC server is unavailable..
repadmin /showreps (also /replsum, /showrepl and /syncall):
Last attempt @ <date> <time> failed, result 1722 (0x6ba): The RPC server is unavailable.
repadmin /syncall names the DNS record it tried to reach:
CALLBACK MESSAGE: Error contacting server <object guid of NTDS Settings object>._msdcs.<forest root domain>.<top level domain> (network error): 1722 (0x6ba): The RPC server is unavailable.
Replicate now in Active Directory Sites and Services fails with "The RPC server is unavailable. This operation will not continue. This condition may be caused by a DNS lookup problem." Promoting a replica DC can also fail because the wizard "could not create the NTDS Settings object" on the helper DC, ending in "The RPC server is unavailable." Directory Service events that commonly carry the 1722 status:
| Source | Event | Text |
|---|---|---|
| Microsoft-Windows-ActiveDirectory_DomainService | 1125 | The Active Directory Domain Services Installation Wizard (Dcpromo) was unable to establish connection with the following domain controller. |
| NTDS KCC | 1311 | The Knowledge Consistency Checker (KCC) has detected problems with the following directory partition. |
| NTDS KCC | 1865 | The Knowledge Consistency Checker (KCC) was unable to form a complete spanning tree network topology. As a result, the following list of sites cannot be reached from the local site. |
| NTDS KCC | 1925 | The attempt to establish a replication link for the following writable directory partition failed. |
| NTDS Replication | 1960 | Internal event: The following domain controller received an exception from a remote procedure call (RPC) connection. The operation may have failed. |
1722 is logged when a lower-layer protocol reports a connectivity failure. Microsoft's list of common causes:
The same 1722 status also shows up outside replication, for example from nltest /sc_query when the Netlogon service isn't running; that case is covered in Netlogon event 5719.
A destination DC finds its replication partner by the objectGUID of the source DC's NTDS Settings object: it looks up the CNAME <objectGUID>._msdcs.<forest root DNS name>, which points to the source DC's host A/AAAA record (the /syncall CALLBACK MESSAGE above shows that exact name). From the destination DC, query each DNS server it is configured to use:
nslookup -type=cname <objectGUID>._msdcs.<forest root domain> <DNS Server IP>
Or in PowerShell (DnsClient module):
Resolve-DnsName -Name <objectGUID>._msdcs.<forest root domain> -Type CNAME -Server <DNS Server IP>
The answer must lead to the source DC's current IP address. Microsoft's 1722 article also lists:
DCDIAG /TEST:DNS /V /E /F:<filename.log> checks DNS health on every DC. Its test groups include Authentication (Auth), Basic (Basc), Records registration (RReg), Dynamic update (Dyn), Delegations (Del) and Forwarders/Root hints (Forw), and the summary table shows PASS, FAIL or n/a per DC.NLTEST /DSGETDC:<netbios or DNS domain name> exercises the DC locator. Add /force to skip the cache, or /gc or /pdc to locate a global catalog or the PDC emulator ("To find the Global Catalog, you must specify a tree name, which is the DNS domain name of the root domain.").ping -a <IP_of_problem_server> is a quick check that the IP address maps back to the right DC.Stale host records, a wrong HOSTS file entry or a stale NTDS Settings object send the destination to the wrong machine. If the name resolves to a computer that isn't a DC (or a DC that hasn't registered the replication service with the endpoint mapper) you get error 1753 instead: "There are no more endpoints available from the endpoint mapper." The full name-to-IP checklist is in replication error -2146893022.
The RPC Endpoint Mapper listens on TCP 135 and tells the client which dynamically assigned port the service (AD replication, FRS and so on) is listening on, so both must be reachable. Ports for DCs on Windows Server 2008 and later, from Microsoft's firewall article (not every port is needed in every scenario):
| Server port | Service |
|---|---|
| 135/TCP | RPC Endpoint Mapper |
| 49152-65535/TCP | RPC for LSA, SAM, NetLogon (also FRS RPC and DFSR RPC) |
| 389/TCP/UDP | LDAP |
| 636/TCP | LDAP SSL |
| 3268/TCP, 3269/TCP | LDAP GC, LDAP GC SSL |
| 53/TCP/UDP | DNS |
| 88/TCP/UDP | Kerberos |
| 464/TCP/UDP | Kerberos password change |
| 445/TCP | SMB |
| 123/UDP | W32Time |
| 9389/TCP | Active Directory Web Services (ADWS) |
Windows Server 2008 and later use a dynamic port range of 49152 through 65535; earlier versions used 1025 through 5000, so old firewall rules are a classic cause. Check the range on a DC:
netsh int ipv4 show dynamicport tcp netsh int ipv4 show dynamicport udp netsh int ipv6 show dynamicport tcp netsh int ipv6 show dynamicport udp
Test from the destination DC. Microsoft's article uses PortQry:
portqry -n <problem_server> -e 135 portqry -n <problem_server> -r 1024-5000
(The -r example is the old 1025-5000 era; on Windows Server 2008 and later the dynamic range is 49152-65535.) Without PortQry, Test-NetConnection -ComputerName <problem_server> -Port 135 reports TcpTestSucceeded : True when the TCP connection works.
If you can't open the whole range, you can pin AD to fixed ports on each DC (both values are needed, because clients also use SamRPC, LSARPC and the DRS interface, not just Netlogon):
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\NTDS\Parameters Registry value: TCP/IP Port Value type: REG_DWORD Value data: (available port) HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Netlogon\Parameters Registry value: DCTcpipPort Value type: REG_DWORD Value data: (available port)
Restart the computer for the NTDS value and the Netlogon service for DCTcpipPort. Setting both to the same port logs Netlogon event 5809 (port in use); 5809 after a Netlogon restart with a unique port is by design and can be ignored. The alternative is shrinking the dynamic range, for example netsh int ipv4 set dynamicport tcp start=10000 num=1000 (ports 10000-10999; minimum range 255 ports, minimum start port 1025, maximum end port 65535).
Under HKEY_LOCAL_MACHINE\Software\Microsoft\Rpc\ClientProtocols these four default values must exist:
| Protocol name | Type | Data value |
|---|---|---|
| ncacn_http | REG_SZ | rpcrt4.dll |
| ncacn_ip_tcp | REG_SZ | rpcrt4.dll |
| ncacn_np | REG_SZ | rpcrt4.dll |
| ncacn_ip_udp | REG_SZ | rpcrt4.dll |
If the key or any of the values are missing, import the key from a known good server.
RequireSecuritySignature (always) and EnableSecuritySignature (if the other side agrees), 0 = disable and 1 = enable, under HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\LanManWorkstation\Parameters and ...\LanManServer\Parameters.Microsoft's next step is NTDS diagnostic logging: level 1 is basic, 2 and 3 add more verbose information and 5 logs extended information. NTDS Replication events 1960, 1961, 1962 and 1963 carry extended RPC information (Microsoft's article notes this for Windows Server 2003 SP2). Microsoft's -2146893022 article raises the DS RPC Client logging entry to 3, triggers replication and reads the fully qualified CNAME in events 1962 and 1963; the destination DC should be able to ping that name and get the source DC's current IP address.
Look up 1925, 1311 and 1960 in the Windows Event ID Lookup, keep the repadmin, dcdiag, Test-NetConnection and Resolve-DnsName lines from the PowerShell Cheat Sheet handy, and filter repadmin /showrepl * /csv output for 1722 rows with the Regex Tester.
Look up 1125, 1311, 1865, 1925 and 1960 in the Windows Event ID Lookup, copy the repadmin, dcdiag and Test-NetConnection lines from the PowerShell Cheat Sheet, filter repadmin CSV output with the Regex Tester, check firewall rule ranges and DC subnets with the IPv4 Subnet Calculator or IPv6 Subnet Calculator, and identify the adapter vendor in dcdiag output with the MAC Vendor Lookup. Related guides: replication error 8453, replication error -2146893022, replication error 8614 and Netlogon event 5719.
1722 (0x6ba, RPC_S_SERVER_UNAVAILABLE) means the RPC client on the destination DC could not connect to the RPC server on the source DC. RPC maps a lower-layer failure, most often a failed TCP connection, to this error, so the cause is usually DNS, a firewall or the network rather than AD itself.
DNS. Microsoft says DNS lookup failures cause a large number of 1722 replication errors. Check that the source DC's objectGUID CNAME under _msdcs of the forest root and its host record resolve to its current IP address on every DNS server the destination DC uses.
TCP 135 for the RPC Endpoint Mapper plus the dynamic RPC range, which is 49152-65535 on Windows Server 2008 and later (1025-5000 on older versions). Microsoft's DC port list also includes LDAP 389, 636, 3268 and 3269, DNS 53, Kerberos 88 and 464, SMB 445, W32Time 123 and ADWS 9389.
Yes. Set the REG_DWORD TCP/IP Port under HKLM\SYSTEM\CurrentControlSet\Services\NTDS\Parameters and DCTcpipPort under the Netlogon Parameters key on each DC, use two different ports, open both on the firewall, then restart the computer and the Netlogon service.
1722 means the RPC connection to the source failed. 1753 (There are no more endpoints available from the endpoint mapper) means the connection reached a machine that does not have the AD replication service registered with its endpoint mapper, for example a non-DC that a bad name-to-IP mapping pointed to.
Runbooks, checklists and helpdesk templates: IT Admin Packs on Gumroad.