All of lore.kernel.org
 help / color / mirror / Atom feed
From: Steven O'Connor <steven@suse.com.au>
To: Robert White <rwhite@pobox.com>, netfilter@vger.kernel.org
Subject: Re: PPTP passthrough
Date: Thu, 4 May 2017 13:46:13 +1000	[thread overview]
Message-ID: <901f9141-693e-2ec8-772a-5961fd6ddfff@suse.com.au> (raw)
In-Reply-To: <019de373-baa3-0e9e-d7f0-4fa63c106091@pobox.com>



On 03/05/17 20:38, Robert White wrote:
> On 05/03/17 02:13, Steven O'Connor wrote:
>> PPTP pass-through seems to be broken. When the client tries to connect,
>> a gre packet is sent but the reply gre packet is dropped at my firewall.
>>
>> The relevant conntrack dump shows a mismatch between the expected reply
>> and the packet received, srckey/dstkey do not match. Is that significant?
>>
>>
>> gre      47 27 src=aaa.bbb.cc.ddd dst=www.xxx.yy.zz srckey=0x0
>> dstkey=0xb053 [UNREPLIED] src=www.xxx.yy.zz dst=aaa.bbb.cc.ddd
>> srckey=0xb053 dstkey=0x0 mark=0 use=1
>> gre      47 27 src=192.168.0.212 dst=aaa.bbb.cc.ddd srckey=0x0
>> dstkey=0x1380 [UNREPLIED] src=aaa.bbb.cc.ddd dst=www.xxx.yy.zz
>> srckey=0x1380 dstkey=0x0 mark=0 use=1
>>
>>
> Doesn't help without your firewall rules.
>
> Do you have a expected,related,established rule in your outgoing chain?
>
> The private address (the only one you didn't blank) suggests that some
> SNAT rule is just a little too ambitious for your own good.
>
> One of the things I do is limit SNAT rules to packets that come from
> real internal-network interfaces. (I use interface group numbers or
> wildcards to make that easier.) Basically you probably don't want to
> SNAT any the packets generated by the local machine.
>
>
>
>
I'm not using any SNAT only masquerade.

The default policy LAN->NET is accept. I have also added a rule to 
accept gre.

It has been working previously but after an update to the kernel or 
shorewall it has stopped working.  I only use pptp occasionally so I 
cannot be sure when it stopped.

The firewall can accept pptp connections from the net and it is only the 
passthru that is broken.


-- 
Steven O'Connor
=============================================
Support Services Pty Ltd   ABN 49 006 226 428
2 Commonwealth Terrace, Sandhurst VIC 3977
Fax 03 9888 2677, Mobile 0417 571 113
=============================================


  reply	other threads:[~2017-05-04  3:46 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-03  2:13 PPTP passthrough Steven O'Connor
2017-05-03 10:38 ` Robert White
2017-05-04  3:46   ` Steven O'Connor [this message]
2017-05-04 19:58     ` Robert White
2017-05-03 14:45 ` Rob Sterenborg (lists)
2017-05-04  3:42   ` Steven O'Connor
2017-05-04  6:07     ` Rob Sterenborg (Lists)
2017-05-04  6:11       ` Rob Sterenborg (Lists)
2017-05-04 18:46       ` Pascal Hambourg

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=901f9141-693e-2ec8-772a-5961fd6ddfff@suse.com.au \
    --to=steven@suse.com.au \
    --cc=netfilter@vger.kernel.org \
    --cc=rwhite@pobox.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.