Australian cloud for business software — built and run in Sydney since 2002

NetLeverage
NetLeverage

w32tm /resync fails with "The computer did not resync because no time data was available"

Technical article
Troubleshooting

About this article: This article is reviewed against the cited sources. This is source research, not lab validation. Check the stated versions and your own requirements before making changes.

Running w32tm /resync on a Windows Server and getting "The computer did not resync because no time data was available" means the Windows Time client sent an NTP request and did not get a proper NTP reply. Microsoft's page names three scenarios: the client cannot find the time server, UDP port 123 is blocked, or the server it asks is not advertising as an NTP server.

Find out which of the three applies before you change anything. The checks below only read settings. The change that follows alters where the server takes its time from, so it comes with a saved copy of the key and a way back.

What the error means

Microsoft's page "w32tm /resync" fails with "The computer didn't resync because no time data was available." applies to Windows Server (all supported versions) and is dated 2026-02-12. It shows the failure like this:

C:\Windows\system32>w32tm /resync /rediscover
Sending resync command to local computer
The computer did not resync because no time data was available

The same page says the error "generally occurs when the client sends an NTP request but doesn't get a proper NTP response in return". It lists two events that can appear in the System log with source Microsoft-Windows-Time-Service: event 129 ("NtpClient was unable to set a domain peer to use as a time source because of discovery error", error "The entry is not found. (0x800706E1)") and event 134 ("NtpClient was unable to set a manual peer to use as a time source because of DNS resolution error", error "No such host is known. (0x80072AF9)"). Both say NtpClient will try again in 15 minutes and double the reattempt interval thereafter.

Microsoft's Windows Time tools page adds the network facts: W32Time follows the Network Time Protocol specification, which requires UDP port 123 for all time synchronisation, and "The built-in W32Time NTP client can only use UDP 123 as the source port."

Likely causes

Cause What you see How to confirm
Scenario 1: the NTP client cannot find the NTP server Event 134 with "No such host is known", or event 129 for a domain peer; a misspelt name or address in the NTPServer value reg query HKLM\SYSTEM\CurrentControlSet\Services\W32Time\Parameters and read NtpServer; test that the name resolves from this server
Scenario 2: UDP port 123 is blocked No reply even though the name resolves Start a network capture, run w32tm /resync /rediscover, and see whether UDP 123 requests leave and whether replies return
Scenario 3: the target is not advertising as an NTP server The client asks a domain controller or other server that does not answer as a time source Check the Windows Time hierarchy from the PDC down; run dcdiag /test:Advertising on the domain controller

Prove the cause before changing anything

All of these only read. Run them in an elevated Command Prompt on the server that shows the error.

w32tm /query /source
w32tm /query /status
w32tm /query /configuration
w32tm /query /peers
reg query HKLM\SYSTEM\CurrentControlSet\Services\W32Time\Parameters

Microsoft's tools page describes /source (displays the time source), /status (displays the W32Time status), /configuration (runtime configuration and the source of the settings) and /peers (peers and their status). The reg query command reads the NtpServer value. Microsoft adds a caveat: if Group Policy sets the NtpServer value through the Configure Windows NTP Client policy on a domain member, "W32Time doesn't use the NtpServer registry value", and w32tm /query /configuration shows the real configuration.

To test the time server itself, Microsoft documents a strip chart of the offset between this computer and another:

w32tm /stripchart /computer:<target> /samples:5 /dataonly

Replace <target> with the time server name. The tools page says /stripchart uses an ephemeral UDP source port, while the built-in client can only use UDP 123 as its source port. A strip chart that works therefore does not prove that the client's own path through a firewall is open.

For scenario 3, on a domain controller:

dcdiag /test:Advertising

The dcdiag page says the Advertising test fails if the Netlogon service has stopped or failed to start. Microsoft's resync page also lists registry values to confirm on the domain controllers under HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\W32Time: TimeProviders\NtpServer with Enabled=1 on all domain controllers, Config with AnnounceFlags = 5 on the PDC, and AnnounceFlags = 10 on member domain controllers. It also names w32tm /monitor as a quick status check on all domain controllers.

Example (illustrative): at contoso.com.au the member server app01 reports "The computer did not resync because no time data was available". w32tm /query /configuration shows NtpServer as tmie.contoso.com.au,0x9, a misspelling of time.contoso.com.au, and the System log holds event 134 with "No such host is known." Scenario 1 fits. Nothing yet points to the firewall or to the domain controllers.

Fix

Match the fix to the scenario you proved.

Scenario 1: a wrong or unreachable NtpServer value

Microsoft's page says to check the spelling and accuracy of the NTPServer registry value, confirm the server is valid with the correct name or address, and make sure the DNS servers defined can resolve it. If the name is wrong, the tools page gives a command to set the manual peer list.

Before you change it, export the Windows Time key to a new file:

reg export HKLM\SYSTEM\CurrentControlSet\Services\W32Time C:\Temp\w32time-before.reg

The reg export page shows the form reg export <keyname> <filename> and says the file must have a .reg extension. Create C:\Temp first if it does not exist.

Disruptive step: the next command changes the list of time servers this server uses, and the resync you run after it asks the server to resynchronise as soon as possible. The tools page says that when the offset is larger than the MaxAllowedPhaseOffset value, w32tm sets the computer clock immediately instead of adjusting it gradually.

Run, with your own server names in place of the placeholders:

w32tm /config /manualpeerlist:"<time-server-1> <time-server-2>" /syncfromflags:manual /update

The tools page says /manualpeerlist is a space-delimited list that must be enclosed in quotation marks when it holds several peers, and that /update notifies W32Time that the configuration is changing. The tools page also says: "if you want to set the local time client to point to multiple peers, we recommend that you prepare three or more time servers." Then follow "Confirm the fix" below, which carries the resync command and its warning.

To undo it, stop the service, import the saved file and start the service again:

net stop w32time
reg import C:\Temp\w32time-before.reg
net start w32time

The import writes the whole saved key back, so it also overwrites any other Windows Time setting changed under that key since the export. Microsoft's reg import page does not say that a value absent from the file is removed, so after an undo run w32tm /query /configuration and compare it with what you recorded.

Scenario 1 on a domain member that should follow the domain

The tools page says: "By default, a computer that's joined to a domain synchronizes time through a domain hierarchy of time sources." It also notes that a computer set to a manual source can be switched back. If the server should follow the domain, export the key again to a different new file name, so the first saved copy is not overwritten:

reg export HKLM\SYSTEM\CurrentControlSet\Services\W32Time C:\Temp\w32time-before-domhier.reg

Disruptive step: the next commands stop the server using its manually listed time servers, make it take time from a domain controller in the hierarchy, and restart the Windows Time service, as the tools page's example does. If the domain controllers have their own time problem, the server inherits it.

w32tm /config /syncfromflags:domhier /update
net stop w32time
net start w32time

To undo it, stop the service, import the second saved file and start the service again:

net stop w32time
reg import C:\Temp\w32time-before-domhier.reg
net start w32time

Scenario 2: UDP port 123 blocked

Microsoft's page says that if the capture shows no UDP 123 request leaving, the local Windows Firewall is the likely block, and if requests leave with no reply, a non-Microsoft firewall is the likely block. The page gives no firewall rule to add. The rule belongs to whoever manages that firewall, so hand them the capture.

Scenario 3: the target does not advertise as a time server

Work from the PDC downward: confirm each domain controller receives time from its own source, confirm advertising with dcdiag /test:Advertising, and compare the registry values listed above. Microsoft's page gives the expected values but no command to set them, so changes there need the domain controller's owner and Microsoft's own tools page for the registry settings.

Confirm the fix

Run the resync again and read the status.

Disruptive step: the resync command asks this server to resynchronise its clock as soon as possible. The tools page says that when the offset is larger than MaxAllowedPhaseOffset the computer clock is set immediately, which can step the time that running applications see. There is no undo other than the next synchronisation.

w32tm /resync /rediscover
w32tm /query /status
w32tm /query /source

The resync should return without the "no time data" message, and /query /source should show the time server you expect. Event 129 and event 134 should stop appearing in the System log.

What to watch afterwards

  • The System log for further Microsoft-Windows-Time-Service warnings. Microsoft says the client doubles the reattempt interval after each failure, so a fixed server can take time to show recovery on its own.
  • The offset on the server against its time source with w32tm /stripchart.
  • Any change to firewall rules or to the time server's address, which can bring scenario 2 or 1 back.
  • Group Policy: if a policy sets the NTP client, a registry edit does not change the effective source.

Risks and limits

These commands have not been run or lab-tested.

Microsoft's resync page covers three scenarios and gives no figures for how many cases each one covers. It gives no firewall rule and no command for the AnnounceFlags or NtpServer registry values on domain controllers. The tools page lists its supported systems as Windows Server 2016 to 2025, Windows 10 and Windows 11, while the resync page says all supported versions of Windows Server. Microsoft's pages name no other causes of a failed resync. Make the change at a time when a clock step on the server is acceptable.

Sources