Development discussion of WireGuard
 help / color / mirror / Atom feed
From: Nicolas Prochazka <nicolas.prochazka@gmail.com>
To: "Dan Lüdtke" <mail@danrl.com>
Cc: WireGuard mailing list <wireguard@lists.zx2c4.com>
Subject: Re: [ wireguard-dev ] About configuring allowedip
Date: Fri, 24 Feb 2017 16:06:06 +0100	[thread overview]
Message-ID: <CADdae-hddrPL+z3Oy2R=yJXKUoxA2tGScJ9Z2sYyx6KgmV=cTw@mail.gmail.com> (raw)
In-Reply-To: <71F759EB-03D2-4769-9B8D-F92C4A19D2C6@danrl.com>

[-- Attachment #1: Type: text/plain, Size: 4467 bytes --]

ok thanks,
what is confusing me it that allowed ip is for :
- authorized source packet
- routing outgoing packet
and we can set allowedips with a lot of ip / netmask
Regards,
Nicolas

2017-02-24 14:10 GMT+01:00 Dan Lüdtke <mail@danrl.com>:

> Nicolas,
>
> I draw your network including the allowed_ips restrictions.
>
> > ping peer3 --peer1--->peer2 : not ok .
>
> This can not work! Peer 2 does not accept the source address from Peer 3.
> Please review your allowed_ips settings. Draw the things on paper, make
> PostIt notes representing the packets including their destination address
> and source address. Draw a little "firewall" on the tunnels (whitelist is
> allowed_ips, all the rest gets dropped!) and see if the PostIt can make it
> through with it's source address. Yes, this sounds like child play, but it
> works. I have taught complex firewalling and VPN setups to lawyers and law
> makers this way. It helps understanding, if a simple diagram does not cut
> it.
>
> Allowed IPs is probably the most complex thing WireGuard has to offer from
> a user perspective. Rename it to Allowed Source Addrresses in your head it
> becomes clearer.
>
> HTH
>
> Dan
>
> > On 24 Feb 2017, at 11:41, Nicolas Prochazka <nicolas.prochazka@gmail.com>
> wrote:
> >
> > hello again,
> > my configuration ,
> > ping peer 1-->peer 2  : ok   ( on ipv6 wg0 )
> > ping peer 3 --> peer 1 : ok
> > ping peer3 --peer1--->peer2 : not ok .
> >
> >
> > On peer 1 , forwarding is setting
> > net.ipv6.conf.all.forwarding = 1
> > net.ipv4.conf.all.forwarding = 1
> >
> >
> > Peer 1 : wg configuration
> >
> > interface: wg0
> >   public key: q5ypTBI7bN0vPGzvlGYyF6pCqYgrDsEjO827duAwjX4=
> >   private key: (hidden)
> >   listening port: 6081
> >
> > peer: dOXT9AvlEt9KSl3ricE12GuVa+U4XB0s1c92s8W+9VA=
> >   endpoint: 52.49.x.x:6081
> >   allowed ips: ::/0
> >   latest handshake: 8 seconds ago
> >   transfer: 71.29 KiB received, 60.28 KiB sent
> >   persistent keepalive: every 25 seconds
> >
> > peer: bqwiLTe/hr0JJMz3IvnDXqS5nOT6u/WL75dasmTE/ko=
> >   endpoint: 10.10.0.69:6081
> >   allowed ips: fd00::baae:edff:fe72:5094/128
> >   latest handshake: 45 seconds ago
> >   transfer: 5.49 KiB received, 6.36 KiB sent
> >
> >
> > Peer 3 :
> >
> >
> > interface: wg0
> >   public key: bqwiLTe/hr0JJMz3IvnDXqS5nOT6u/WL75dasmTE/ko=
> >   private key: (hidden)
> >   listening port: 6081
> >
> > peer: q5ypTBI7bN0vPGzvlGYyF6pCqYgrDsEjO827duAwjX4=
> >   endpoint: 10.10.99.230:6081
> >   allowed ips: ::/0
> >   latest handshake: 33 seconds ago
> >   transfer: 4.92 KiB received, 7.55 KiB sent
> >   persistent keepalive: every 25 seconds
> >
> >
> > Peer 2 :
> >
> > interface: wg0
> >   public key: dOXT9AvlEt9KSl3ricE12GuVa+U4XB0s1c92s8W+9VA=
> >   private key: (hidden)
> >   listening port: 6081
> >
> > peer: q5ypTBI7bN0vPGzvlGYyF6pCqYgrDsEjO827duAwjX4=
> >   endpoint: 77.156.x.x:58943
> >   allowed ips: fd00::eea8:6bff:fef9:23bc/128
> >   latest handshake: 1 minute, 43 seconds ago
> >   transfer: 52.59 KiB received, 79.01 KiB sent
> >
> >
> > 2017-02-23 14:41 GMT+01:00 Dan Lüdtke <mail@danrl.com>:
> > Nicolas: Could you provide the configuration files? Because from your
> little graphic or schema I can not even derive what you are configuring. I
> guess there is something overlapping prefixes maybe?
> >
> > Jason: I think we are approaching the point in time when there will be a
> -dev and a -users ML :)
> >
> >
> > > On 23 Feb 2017, at 14:03, Nicolas Prochazka <
> nicolas.prochazka@gmail.com> wrote:
> > >
> > > Hello, i'm trying to do this with wireguard, withtout success :
> > >
> > > peer1 ---> peer2   : config ok , works
> > > peer3 ---> peer1  : config ok , works
> > > peer3 --->peer1 ---> peer2  : not ok .
> > >
> > > I suspect allowed-ip configuration, but all my tests does not works.
> > > perhaps I must create two wireguard interface on peer 1 and do
> forwarding/routing ?
> > > i'm using ipv6 as internal ip.
> > >
> > > so my question is :
> > > - two interface ?
> > > - specifiq magic allowedip ?
> > > ( allowed ip is confusing for, it is using for routing and for
> evicting paquet ? )
> > >
> > > Regards,
> > > Nicolas
> > > _______________________________________________
> > > WireGuard mailing list
> > > WireGuard@lists.zx2c4.com
> > > https://lists.zx2c4.com/mailman/listinfo/wireguard
> >
> >
>
>

[-- Attachment #2: Type: text/html, Size: 6065 bytes --]

  reply	other threads:[~2017-02-24 15:04 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-02-23 13:03 Nicolas Prochazka
2017-02-23 13:41 ` Dan Lüdtke
2017-02-24 10:41   ` Nicolas Prochazka
2017-02-24 13:10     ` Dan Lüdtke
2017-02-24 15:06       ` Nicolas Prochazka [this message]
2017-02-23 21:16 ` Baptiste Jonglez

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='CADdae-hddrPL+z3Oy2R=yJXKUoxA2tGScJ9Z2sYyx6KgmV=cTw@mail.gmail.com' \
    --to=nicolas.prochazka@gmail.com \
    --cc=mail@danrl.com \
    --cc=wireguard@lists.zx2c4.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).