pptp stops receiving packets when bandwidth spike

I have produced the patch for version 1.7.2 of the pptp-client that does the following:
1) sets default MISSING_WINDOW value to 6000 instead of old 300
2) allows to specify the MISSING_WINDOW value in the command-line with `--missing-window` option to pptp
Looks like this patch either solves the problem or at least greatly reduces it's chances of happening.

The attachment "Patch allows flexible tuning of MISSING_WINDOW" of this bug report has been identified as being a patch. The ubuntu-reviewers team has been subscribed to the bug report so that they can review the patch. In the event that this is in fact not a patch you can resolve this situation by removing the tag 'patch' from the bug report and editing the attachment so that it is not flagged as a patch. Additionally, if you are member of the ubuntu-reviewers team please also unsubscribe the team from this bug report.

[This is an automated message performed by a Launchpad user owned by Brian Murray. Please contact him regarding any issues with the action taken in this bug report.]

Hello,
I have patched pptpclient-1.7.2, too. The main difference to the excellent patch of Ruslan is that here 'missing window' validation is disabled by default. The command line switch is implemented as laid out by Ruslan. After some discussion, colleagues of mine and me came to the conclusion that this validation check offers no benefit whatsoever. There are far more pressing security issues surrounding pptp in general and spoofing can't be detected reliably in this way. The 'missing window' approach does not detect bad packages either - in a nutshell: it is a bit arbitrary to say the least. Deactivating _might_ have the advantage of scalability when going 30mbit/s and beyond. A very rough estimate suggests, that a range of 6000 packets in a .3 second window could be surpassed when 27mbit/s are reached (and something disturbs the force so to speak).

I've build a version of pptp-linux inculding ~mybrain's patch in my ppa: https://code.launchpad.net/~mwhudson/+archive/ppa, which may help people affected by this bug.

(for precise only so far)

I've installed mwhudson's pptp-linux deb, and so far I haven't had had any disconnection issues in the last two days. (previously it would disconnect on average once every 8 minutes, if there was any data going through.

I've sent the patch on to debian: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=680455 (although I don't know if I did this in a useful way!)

Thank you Michael for filing with Debian. Seems though that not much has happened. I updated the patch in a very small way. I adjusted the log level for the "lost / reordered" case. This spammed my log files heavily (several GB in a matter of hours.) Now it is grouped together with most of the of the other cases and is reported for loglevel >= 2, as 1 is the default.

Experienced exactly same behavior (using commercial PureVPN service). When traffic spikes (torrents or direct downloads via browser) my VPN connection died. syslog showed: 'No response to 5 echo-requests'.

The patch of mwhudson works for me perfect. Running Ununtu 12.04.

Kudo's for Michael Hudson-Doyle!

Confirmed this patch also works on 13.10.

Confirmed this patch also work perfectly on 14.04.
pptp-linux-1.7.2-7

It's a SHAME that such a good fix is not yet in the "official" package.
It seems that debian mantainer doesn't care too.
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=680455

"Thanks a lot for this report. I'll look into this one" since Mon, 30 Jul 2012 23:09:42 +0200

@asavah, you wrote "It's a SHAME that such a good fix is not yet in the "official" package."

I'm laughing like anything. I'm the upstream and I make the official package, yet this fix has NEVER come to me. I just happen to notice it now, here. Next time, make sure you talk to upstream before complaining about upstream. ;-)

Merged as df45680 after fixing indentation. Thanks for the patch.
https://sourceforge.net/p/pptpclient/git/ci/df4568002b67de068bbcf16bbdbca800986f325b/

Ruslan, mail me at <email address hidden> if you want your e-mail address added to AUTHORS; I'd normally do it, but it wasn't in the diff.