This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(217.60000000000002, 33%, 31%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EJG%3C/text%3E%3C/svg%3E)
I use the built-in L2TP/Ipsec client to attach to my office. mainly for doing backup. When I disconnect the VPN, it sometimes puts my Windows 10 desktop into a limbo, where the internet is inaccessible.
The failure is intermittent. After disconnecting the VPN, I can't find anything amiss. It might be more common when an XCOPY is in progress. Here are the symptoms I know of:
1) The routing table is exactly like the normal one (when the failure doesn't happen).
2) I can ping 8.8.8.8 but I can't ping google.com
3) Ipconfig/all shows the usual IP (which is static), gateway, and DNS (which is 8.8.8.8)
4) Other wired and wifi clients on our home network work just fine.
5) Thunderbird and all browsers have the problem, including after an app restart.
I've tried the following to fix the issue
1) Disable/enable the ethernet wired interface
2) Uninstall/reinstall the same interface
3) same actions for the disconnected VPN interface
4) Change the IP, including making it dynamic
Sometimes, it will start working after 10/15 minutes. Most of the time, I have to reboot.
Hello Jim,
If you are willing to share trace data, you could try the following the next time the problem occurs:
netsh trace start report=disabled provider=Microsoft-Windows-TCPIP provider=Microsoft-Windows-DNS-Client provider=Microsoft-Windows-WFP tracefile=why.etl netsh trace stop The resulting trace file (why.etl) might contain some useful hints about what is happening - if you are happy to share it then I will take a look...
Gary
This failure can be fairly rare - I'll share it when I can.
Hello Jim,
No problem; if you want to be sure of getting my attention, include @Gary Nebbett in the message - that triggers an e-mail to me.
There is no guarantee that the trace will reveal the cause of the problem, but it is a structured way of tackling the problem (rather than just changing, resetting, reinstalling, etc. things in a scattergun fashion).
This image gives a hint about the type of data gathered (DNS lookup, route (interface, next hop) selection, etc.):
Gary
I had the failure this morning, but the ETL file is unreadable (by Notepad, at least) and this forum won't allow me to attach it.
Hello Jim,
There are several options; here is a non-exhaustive list:
Gary
I don't have a linkedin account (and I don't want one). You could email me at jjgurleyATgmail
Here's a Google Drive copy of my trace @Gary Nebbett
https://drive.google.com/file/d/1-R2bIq-q7xG8IzM6e7wyMhHD4J3P29OL/view?usp=sharing
Hello Jim,
I e-mailed your gmail address as soon as you published it; I clicked on your Google Drive link as soon as you published it and requested access (the link requires access to be explicitly granted).
Somewhere there should be two or more e-mail messages waiting for you (from me directly and from Google); when and if you find and act upon them then we can look at the trace data.
Gary