Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Windows OS: Error occurred after upgrade from 1.6 to 1.8.3 on windows server environment. #1920

Closed
darshinimashar opened this issue May 13, 2021 · 9 comments
Closed

Windows OS: Error occurred after upgrade from 1.6 to 1.8.3 on windows server environment. #1920

darshinimashar opened this issue May 13, 2021 · 9 comments
Assignees
@crawshaw
Labels
L3 Some users Likelihood OS-windows Issues involving Tailscale for Windows P5 Halts deployment Priority level T6 Major usability Issue type

Comments

@darshinimashar
Copy link

darshinimashar commented May 13, 2021 •
edited
Loading

Customer upgraded environment on SCCM or other windows server and getting following error on startup failure,

A fatal error occurred while creating a TLS client credential. The internal error state is 10013. (source Schannel)
Initialization failed because the driver device could not be created. Use the string "0000000001003200000002000000000000000000000000000000**********" to identify the interface for which initialization failed. It represents the MAC address of the failed interface or the Globally Unique Interface Identifier (GUID) if NetBT was unable to map from GUID to MAC address. If neither the MAC address nor the GUID was available, the string represents a cluster device name. (source NetBT)
The server could not bind to the transport \Device\NetBT_Tcpip_{************************} because another computer on the network has the same name. The server could not start. (source Server)

untitled (1)

When error appears it unable to get bug report too,
image

Front logo Front conversations

Platform: Windows server 2016
@darshinimashar
Copy link
Author

Bugreport marker created:
BUG-bbe2e1538db7dd00aad95b72d54ff18689e623f334ef4af8881d7afa45a81950-20210513142502Z-226f488cc0c259e4

From another customer from the front conversation: https://forum.tailscale.com/t/tailscale-stops-working-after-1-8-0-3-update-windows-server-rds-issue-mutliple-users/635

@darshinimashar
Copy link
Author

One more customer reported the issue,https://forum.tailscale.com/t/windows-server-2016-and-2019-rds/318/11.

@evilhamsterman
Copy link

evilhamsterman commented May 14, 2021 •
edited
Loading

I get the "Failed to connect to Tailscale service." error on Windows 10 whenever we do a fast user switch from my account to my wife's account. Tailscale will be running fine in my account but when she switches to her account she will get repeated alerts from Tailscale and I have to end up killing the service and all Tailscale processes to clear it. It has been happening ever since I updated from 1.6.0 to 1.8.0

@a-sorochak
Copy link

I have the same issue on Windows 10 after Tailscale update from 1.6.0 to 1.8.3.
Tailscale was installed with "Run unattended" option and when any other Windows user logs in, he sees multiple error messages "Failed to connect to Tailscale service". Tailscaled is running meanwhile.

Fragment of Tailscale's log
2021-05-16T10:07:02.223+03:00: panic: runtime error: invalid memory address or nil pointer dereference 2021-05-16T10:07:02.223+03:00: [signal 0xc0000005 code=0x0 addr=0x40 pc=0x129ee2b] 2021-05-16T10:07:02.223+03:00: 2021-05-16T10:07:02.223+03:00: goroutine 502 [running]: 2021-05-16T10:07:02.223+03:00: tailscale.com/ipn.NewBackendServer(0xc0000888a0, 0x0, 0x0, 0xc0000aa078, 0x173cf30) 2021-05-16T10:07:02.223+03:00: tailscale.com@v0.0.0-00010101000000-000000000000/ipn/message.go:108 +0xcb 2021-05-16T10:07:02.223+03:00: tailscale.com/ipn/ipnserver.(*server).serveConn(0xc0001867e0, 0x1737b30, 0xc000036040, 0x173cf30, 0xc0001ce018, 0xc0000888a0) 2021-05-16T10:07:02.223+03:00: tailscale.com@v0.0.0-00010101000000-000000000000/ipn/ipnserver/server.go:255 +0x330 2021-05-16T10:07:02.223+03:00: created by tailscale.com/ipn/ipnserver.Run 2021-05-16T10:07:02.223+03:00: tailscale.com@v0.0.0-00010101000000-000000000000/ipn/ipnserver/server.go:729 +0x917 2021-05-16T10:07:02.304+03:00: BabysitProc: backoff: 7 msec 2021-05-16T10:07:02.476+03:00: 2021/05/16 10:07:02 Program starting: v1.8.3-td0e86b08c-g03346d1fb: []string{"C:\\Program Files (x86)\\Tailscale IPN\\tailscaled.exe", "/subproc", "32167d78dac5e623d6ddab1526511cda953abd4a92514cdaa714c2a817ac7db0"} 2021-05

I have found workaround by removing tailscale-ipn.lnk from common Windows startup directory (C:\ProgramData\Microsoft\Windows\Start Menu\Programs\StartUp).

@crawshaw crawshaw self-assigned this May 18, 2021
@crawshaw
Copy link
Contributor

There are several code paths in ipn/ipnserver/server.go that are wrong. We are calling NewBackendServer with a nil b Backend and a set sendNotifyMsg parameter, which immediately trips over this code:

        if sendNotifyMsg != nil {
                b.SetNotifyCallback(bs.send)
        }

That causes the panic above.

@crawshaw
Copy link
Contributor

OK the panic is in error cases, and is easily fixed. Unfortunately, fixing it doesn't necessarily fix the underlying problem.

crawshaw added a commit that referenced this issue May 18, 2021
@crawshaw
ipn: allow b to be nil in NewBackendServer
0c55ab0 
A couple of code paths in ipnserver use a NewBackendServer with a nil
backend just to call the callback with an encapsulated error message.
This covers a panic case seen in logs.

For #1920

Signed-off-by: David Crawshaw <crawshaw@tailscale.com>
@crawshaw crawshaw mentioned this issue May 18, 2021
Merged
crawshaw added a commit that referenced this issue May 19, 2021
@crawshaw
ipn: allow b to be nil in NewBackendServer
293a2b1 
A couple of code paths in ipnserver use a NewBackendServer with a nil
backend just to call the callback with an encapsulated error message.
This covers a panic case seen in logs.

For #1920

Signed-off-by: David Crawshaw <crawshaw@tailscale.com>
crawshaw added a commit that referenced this issue May 19, 2021
@crawshaw
ipn: allow b to be nil in NewBackendServer
256db54 
A couple of code paths in ipnserver use a NewBackendServer with a nil
backend just to call the callback with an encapsulated error message.
This covers a panic case seen in logs.

For #1920

Signed-off-by: David Crawshaw <crawshaw@tailscale.com>
@DentonGentry DentonGentry changed the title Windows OS: Error occurred after upgrade form 1.6 to 1.8.3 on windows server environment. Windows OS: Error occurred after upgrade from 1.6 to 1.8.3 on windows server environment. May 20, 2021
@DentonGentry DentonGentry added L3 Some users Likelihood OS-windows Issues involving Tailscale for Windows P5 Halts deployment Priority level T6 Major usability Issue type and removed needs-triage labels May 20, 2021
@darshinimashar
Copy link
Author

This should resolve: #1953 & #1775

@bradfitz
Copy link
Member

@darshinimashar, did you mean #1775 or https://github.com/tailscale/corp/issues/1775?

@bradfitz
Copy link
Member

Emailed customer, but I think this is done.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@crawshaw crawshaw

Labels
L3 Some users Likelihood OS-windows Issues involving Tailscale for Windows P5 Halts deployment Priority level T6 Major usability Issue type
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
6 participants
@bradfitz @evilhamsterman @crawshaw @DentonGentry @a-sorochak @darshinimashar