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.
Error 64 is the same code as 0x40. Microsoft's page lists it as hexadecimal error 0x40, decimal 64, symbolic string ERROR_NETNAME_DELETED, with the friendly text "The specified network name is no longer available."
Two Microsoft pages cover specific cases: a domain join that fails because Kerberos traffic to the domain controller is dropped by a WAN accelerator or firewall, and a Windows Server 2012 R2 or Windows 8.1 share reached over SMB version 1 that fails at random. Neither Microsoft page covers a file copy from a share dropping with error 64, so these pages give no cause for that case.
What the error means
The domain join page, "Domain join error 0x40 The specified network name is no longer available", shows a table of the error: hexadecimal 0x40, decimal 64, ERROR_NETNAME_DELETED, friendly error "The specified network name is no longer available." In the domain join case, the netsetup.log file contains entries like these:
NetUseAdd to \\dc1.adatum.com\IPC$ returned 64
NetpJoinDomain: status of connecting to dc '\\dc1.adatum.com': 0x40
NetpJoinDomainOnDs: Function exits with status of: 0x40
Microsoft's explanation: the issue "is related to getting Kerberos Tickets for a Server Message Block (SMB) session." The client sends a TGS request for the service name cifs/DC1.ADATUM.COM, the domain controller sends back a TCP acknowledgment, which Microsoft says suggests it received the request, but it "doesn't reply with a valid TGS response", and the client ends the TCP connection.
The second page, for Windows Server 2012 R2 and Windows 8.1, describes a different path to the same text. Shared folders are created on a computer running Windows Server 2012 R2, Windows 8.1 or Windows RT 8.1, SMB version 1 is enabled, and an older computer (Windows Server 2003 R2 or Windows XP) tries to reach them over SMB1. The message appears "randomly". Microsoft's cause: "there is a race condition between SMB version 1 Service driver loading and the Server service initialization." That page shows no date.
Likely causes
| Cause | What you see | How to confirm |
|---|---|---|
| A WAN accelerator acknowledges the Kerberos service-ticket request but the response never reaches the domain controller (domain join) | returned 64 in netsetup.log during a domain join |
Network trace: IP time-to-live values of 64 or lower on the frames, which Microsoft says is the TTL used by Unix-like devices |
| A firewall or other network device between client and domain controller drops the KDC response | Same netsetup.log lines | Concurrent network trace taken on the domain controller side |
| SMB1 race at server start on Windows Server 2012 R2 or Windows 8.1 | Random failures reaching a share from Windows XP or Server 2003 R2 | SMB1 is enabled on the server, the clients are old, and the update rollup below is not installed |
| Anything else | The same message from a copy or a mapped drive | Microsoft's pages give no cause, so none is claimed here |
Prove the cause before changing anything
These steps change nothing on the server.
For the domain join case, Microsoft's troubleshooting steps are: use the net use command to access the same UNC path and reproduce the issue, and collect a network trace of the net use run. Then read the trace for the sequence Microsoft describes: the client sends a TGS request for the cifs/ service name of the domain controller, retransmits the segment and reduces the frame size, and the domain controller acknowledges but never returns a valid ticket. The page says a full understanding "needs a trace from the DC network". In Microsoft's worked trace, the sequence numbers show that the server did send data the client never received, and the page says "This discrepancy directly indicates issues with the correct flow of network traffic." A gap between what the domain controller sent and what the client received is what to look for in your own trace.
Microsoft's walk-through of that trace gives the detail to look for. The client retransmits the segment carrying the request (frames 1541 and 1542) and then sends a smaller frame (frame 1545), which the domain controller acknowledges in frame 1546. The client keeps sending until the last segment in frame 1550, which carries the Push bit and ends at byte 1299631114, and the server acknowledges that in frame 1555. Meanwhile the client sends a further request in frames 1552 and 1556, ending at byte 1299632506. Until the client ends the session, only a single message from the server (frame 1553) reaches it. The server's sequence numbers show it did send other data, which the client never saw, so the client keeps acknowledging the same number until it gives up. Microsoft's page notes that a full explanation needs the trace taken on the domain controller's network as well, and says that trace might also show why the client does not reset the session earlier. Microsoft's page lists two causes: a WAN accelerator that responds to acknowledge the TGS request but whose response does not arrive at the Key Distribution Center (KDC), and a network device such as a firewall that dropped the KDC response.
For the SMB1 case, check whether SMB1 is enabled on the server. Microsoft's SMB page gives this read-only command:
Get-SmbServerConfiguration | Select EnableSMB1Protocol
Also check the list of installed updates for the August 2014 update rollup, Microsoft's KB 2975719, which Microsoft's page names as the resolution.
Example (illustrative): at contoso.com.au, the new server app01 at a branch fails to join the domain. The netsetup.log lines show NetUseAdd to \\dc1.contoso.com.au\IPC$ returned 64. A trace of net use shows the Kerberos service-ticket request sent several times with the frame size shrinking and no valid reply from dc1. Frames from the branch show a time-to-live of 64, the value Microsoft says Unix-like devices use, and the branch has a WAN accelerator in the path. That matches Microsoft's first cause, so the next step is to look at the accelerator, not at the server.
Fix
Domain join case
Microsoft's page names the two causes and gives no reconfiguration steps. The fix belongs to whoever runs the WAN accelerator or the firewall that the trace points to. Microsoft's page does not give product-specific settings. Take the trace and the log lines to that team.
SMB1 case on Windows Server 2012 R2 or Windows 8.1
Microsoft's resolution is to install update rollup 2975719, the August 2014 update rollup for Windows RT 8.1, Windows 8.1 and Windows Server 2012 R2. Microsoft's page does not say which computer to install it on; its workaround section refers to the computer on which you create a shared folder.
Risk: installing an update rollup changes the server, so the owner of the server plans it like any other update; Microsoft's page gives no steps for removing it.
Microsoft's page also describes a registry workaround under a "Workaround" heading, in which the DependOnService entry of the LanManServer service is edited and the computer is restarted. The page prints the new value as a single line, SamSS Srv2 Srv, and does not say whether the three service names go on one line or each on its own line in the Value data box. If the update cannot be installed, take that question to Microsoft support before editing the value.
To see what is set now, without changing it, run this read-only command and keep the output with your notes:
reg query "HKLM\SYSTEM\CurrentControlSet\services\LanManServer" /v DependOnService
Confirm the fix
- For the domain join case, the join completes and netsetup.log no longer shows
returned 64. - For the SMB1 case, reach the shared folder from the old client several times over a day. Microsoft says the failure was random, so one success proves little.
- Check the installed updates for the August 2014 update rollup, Microsoft's 2975719.
What to watch afterwards
- SMB1 itself. Microsoft's SMB page says SMBv1 has significant security vulnerabilities and that Microsoft strongly encourages you not to use it, and that it recommends updating the SMB server that still requires it rather than reinstalling SMBv1. If the old clients can be retired, the race stops mattering.
- Network devices on the Kerberos path, after any change to a WAN accelerator or firewall. Microsoft says the trace from the domain controller's side is what shows where the response was dropped.
- New occurrences that do not match either case, which need their own diagnosis.
Read next
Related articles you may want next.
- TroubleshootingSystem error 53 "The network path was not found": name, port 445 or the Server serviceSystem error 53 means the client could not reach the share. Check the server's name, TCP port 445, the Server service and the firewall at the other end, and the client's own TCP/IP NetBIOS Helper service. Prove each one in turn.Read article
- TroubleshootingSMB Signing Now Required: NAS and Old Client Access FailuresNAS share fails after Windows 11 24H2 or Server 2025? Tell signing (STATUS_INVALID_SIGNATURE) from blocked guest access, audit with SMBClient events, fix the device first.Read article
- Troubleshooting"There are currently no logon servers available to service the logon request" at Windows sign-inSign-in says no logon servers are available? Microsoft's three causes: purged cached credential (event 45058), cloned DC in DSRM, broken secure channel.Read article
Risks and limits
These commands have not been run or lab-tested.
Microsoft's domain join page states supported versions of Windows Server and gives no earlier version limit. The SMB1 page applies only to Windows Server 2012 R2, Windows 8.1 and Windows RT 8.1, and shows no date. Neither page covers a copy or backup job over a normal share failing with error 64. Microsoft's registry workaround for the SMB1 case is not set out, because its page does not say how the three names are entered.
Sources
- Domain join error 0x40 "The specified network name is no longer available" — Microsoft Learn, page dated 2026-02-12, accessed 22 November 2026
- Shared folder in Windows Server 2012 R2 or Windows 8.1 cannot be accessed by using SMB version 1 protocol — Microsoft Support, date not shown, accessed 22 November 2026
- Detect, enable, and disable SMBv1, SMBv2, and SMBv3 in Windows — Microsoft Learn, page dated 2026-11-18, accessed 22 November 2026
- reg query — Microsoft Learn, page dated 2023-02-03, accessed 22 November 2026
This article was originally published on .