Linux Netfilter discussions
 help / color / mirror / Atom feed
From: Matt Domsch <matt@domsch.com>
To: Harald Welte <laforge@netfilter.org>, netfilter@lists.netfilter.org
Subject: Re: ip_nat_pptp ICMP rejected failures
Date: Wed, 5 Oct 2005 22:54:22 -0500	[thread overview]
Message-ID: <20051006035422.GA8836@domsch.com> (raw)
In-Reply-To: <20051005154453.GC4184@rama>

[-- Attachment #1: Type: text/plain, Size: 5466 bytes --]

On Wed, Oct 05, 2005 at 05:44:53PM +0200, Harald Welte wrote:
> On Wed, Oct 05, 2005 at 10:13:09AM -0500, Matt Domsch wrote:
> there have been patches for lots of 2.4 and 2.6 releases, though.

Indeed, I've been lazy in this respect, no disrespect intended.
 
> thanks for the detailed bugreport, I'll try to analyze the problem once
> I'm back from http://workshop.netfilter.org/

Enjoy the workshop!
 
> Please try to explicitly add a drop rule for the ICMP packets and see
> whether it works then.  Sounds strange, but I have my reasons for asking
> ;)


Progress.  To the raw table I added:
-A OUTPUT -p icmp --icmp-type protocol-unreachable --destination PPTP_SERVER -j DROP

Now what happens is that the LCP Configuration Requests and LCP
Configuration Acks generated by the PPTP_SERVER aren't getting NAT'd
by the firewall coming back to my client, so LCP is never
established.  Removing ip_nat_pptp resolves this of course.


No.     Time        Source                Destination           Protocol Info
     47 1.029527    CLIENT_IP           PPTP_SERVER         TCP      4668 > 1723 [SYN] Seq=0 Ack=0 Win=64512 Len=0 MSS=1460
     48 1.029729    FW_PUBLIC_IP        PPTP_SERVER         TCP      4668 > 1723 [SYN] Seq=0 Ack=0 Win=64512 Len=0 MSS=1460
     49 1.056471    PPTP_SERVER         FW_PUBLIC_IP        TCP      1723 > 4668 [SYN, ACK] Seq=0 Ack=1 Win=16384 Len=0 MSS=1460
     50 1.056586    PPTP_SERVER         CLIENT_IP           TCP      1723 > 4668 [SYN, ACK] Seq=0 Ack=1 Win=16384 Len=0 MSS=1460
     51 1.071807    CLIENT_IP           PPTP_SERVER         PPTP     Start-Control-Connection-Request
     52 1.071899    FW_PUBLIC_IP        PPTP_SERVER         PPTP     Start-Control-Connection-Request
     53 1.096896    PPTP_SERVER         FW_PUBLIC_IP        PPTP     Start-Control-Connection-Reply
     54 1.096993    PPTP_SERVER         CLIENT_IP           PPTP     Start-Control-Connection-Reply
     55 1.108357    CLIENT_IP           PPTP_SERVER         PPTP     Outgoing-Call-Request
     56 1.108461    FW_PUBLIC_IP        PPTP_SERVER         PPTP     Outgoing-Call-Request
     57 1.133729    PPTP_SERVER         FW_PUBLIC_IP        PPTP     Outgoing-Call-Reply
     58 1.133875    PPTP_SERVER         CLIENT_IP           PPTP     Outgoing-Call-Reply
     59 1.137014    CLIENT_IP           PPTP_SERVER         PPTP     Set-Link-Info
     60 1.137101    FW_PUBLIC_IP        PPTP_SERVER         PPTP     Set-Link-Info
     61 1.149890    CLIENT_IP           PPTP_SERVER         PPP LCP  Configuration Request
     62 1.150004    FW_PUBLIC_IP        PPTP_SERVER         PPP LCP  Configuration Request
     63 1.174587    PPTP_SERVER         FW_PUBLIC_IP        PPP LCP  Configuration Request
====== The previous packet should be NAT'd here, but it's not.
     64 1.174765    PPTP_SERVER         FW_PUBLIC_IP        PPP LCP  Configuration Ack
====== The previous packet should be NAT'd here, but it's not.
     71 1.335423    PPTP_SERVER         FW_PUBLIC_IP        TCP      1723 > 4668 [ACK] Seq=189 Ack=349 Win=17172 Len=0
     72 1.335582    PPTP_SERVER         CLIENT_IP           TCP      1723 > 4668 [ACK] Seq=189 Ack=349 Win=17172 Len=0
     96 2.319976    PPTP_SERVER         FW_PUBLIC_IP        PPP LCP  Configuration Request
====== The previous packet should be NAT'd here, but it's not.
    125 3.136456    CLIENT_IP           PPTP_SERVER         PPP LCP  Configuration Request
    126 3.136592    FW_PUBLIC_IP        PPTP_SERVER         PPP LCP  Configuration Request
    127 3.160169    PPTP_SERVER         FW_PUBLIC_IP        PPP LCP  Configuration Ack
====== The previous packet should be NAT'd here, but it's not.
    181 5.319708    PPTP_SERVER         FW_PUBLIC_IP        PPP LCP  Configuration Request
====== The previous packet should be NAT'd here, but it's not.
    213 6.133984    CLIENT_IP           PPTP_SERVER         PPP LCP  Configuration Request
    214 6.134116    FW_PUBLIC_IP        PPTP_SERVER         PPP LCP  Configuration Request
    215 6.159191    PPTP_SERVER         FW_PUBLIC_IP        PPP LCP  Configuration Ack
====== The previous packet should be NAT'd here, but it's not.
    308 9.319669    PPTP_SERVER         FW_PUBLIC_IP        PPP LCP  Configuration Request
====== The previous packet should be NAT'd here, but it's not.
    340 10.144477   CLIENT_IP           PPTP_SERVER         PPP LCP  Configuration Request
    341 10.144617   FW_PUBLIC_IP        PPTP_SERVER         PPP LCP  Configuration Request
    342 10.167478   PPTP_SERVER         FW_PUBLIC_IP        PPP LCP  Configuration Ack
    427 13.061012   CLIENT_IP           PPTP_SERVER         PPTP     Set-Link-Info
    428 13.061137   FW_PUBLIC_IP        PPTP_SERVER         PPTP     Set-Link-Info
    429 13.071593   CLIENT_IP           PPTP_SERVER         PPP LCP  Termination Request
    430 13.071680   FW_PUBLIC_IP        PPTP_SERVER         PPP LCP  Termination Request
    431 13.095415   PPTP_SERVER         FW_PUBLIC_IP        PPP LCP  Termination Ack
    435 13.258806   PPTP_SERVER         FW_PUBLIC_IP        TCP      1723 > 4668 [ACK] Seq=189 Ack=373 Win=17148 Len=0
    436 13.258922   PPTP_SERVER         CLIENT_IP           TCP      1723 > 4668 [ACK] Seq=189 Ack=373 Win=17148 Len=0


There's no apparent difference between the packets that are NATd
without ip_nat_pptp and the packets that aren't when using ip_nat_pptp.

Thanks,
Matt

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

  reply	other threads:[~2005-10-06  3:54 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-10-05 15:13 ip_nat_pptp ICMP rejected failures Matt Domsch
2005-10-05 15:44 ` Harald Welte
2005-10-06  3:54   ` Matt Domsch [this message]
2005-10-08  5:05     ` Matt Domsch

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=20051006035422.GA8836@domsch.com \
    --to=matt@domsch.com \
    --cc=laforge@netfilter.org \
    --cc=netfilter@lists.netfilter.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox