SoftEther VPN User Forum

I recently found that the vpnserver service had crashed on my server and it had deleted _all_ of its configuration files, including the dozen backup files.

how even...?

Did you install the VPN server on temporary file system?
If so, newly created files will be deleted when OS re-start.

thisjun wrote:
> Did you install the VPN server on temporary file system?
> If so, newly created files will be deleted when OS re-start.

no... this is a permanent server that's been running for a good few years now, the vpn server simply deleted all of its config files after a reboot for no apparent reason. Never did it before...

What OS did you use?
Where did you install SoftEther VPN in?

thisjun wrote:
> What OS did you use?
> Where did you install SoftEther VPN in?

CentOS 7.3 and vpnserver is under /usr/local, pretty sure this is the default when you install it.

Thats scary, but I don't think that server would do that by itself it sounds more like something that would be caused by the user. Have you checked server logs for unauthorized access? See the logs of what could of caused server to reboot? Anything suspicious in the VPN server logs?

andr1717 wrote:
> Thats scary, but I don't think that server would do that by itself it
> sounds more like something that would be caused by the user. Have you
> checked server logs for unauthorized access? See the logs of what could of
> caused server to reboot? Anything suspicious in the VPN server logs?

From what I remember the only things logged were 'no config found' etc, it probably deleted its recent log files too. It was likely triggered due to a sudden lack of disk space due to USB bugs with an external drive and for some reason the kernel was writing the backups to the main disk instead of the external one...

Regardless of a lack of space or not, it definitely shouldn't flat-out erase _existing_ files from all existence...

The server's kept pretty secure, so no known breaches of any kind, I'm the only one using it . I had also done multiple disk checks upon finding files were missing in the off-chance a corruption had occurred when it had shut down, but no file errors had been found, disks are perfectly healthy.

I don't have a lot of experience with CentOS but lack of disk space should just produce error message something like "No enough space can't save whatever" but not erase unless there is a cron task that rotates files and deletes the old ones, I could see it then going unexpectedly and erasing stuff.

Also, only SothEther files were affected or some other files too?

If disk space run out, these situation will happen.