239 lines
13 KiB
HTML
239 lines
13 KiB
HTML
|
<html>
|
||
|
|
<head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
|
||
|
|
<title>Re: GSoC Port Forwarding</title>
|
||
|
|
<link rel="important stylesheet" href="">
|
||
|
|
<style>div.headerdisplayname {font-weight:bold;}
|
||
|
|
</style></head>
|
||
|
|
<body>
|
||
|
|
<table border=0 cellspacing=0 cellpadding=0 width="100%" class="header-part1"><tr><td><div class="headerdisplayname" style="display:inline;">Oggetto: </div>Re: GSoC Port Forwarding</td></tr><tr><td><div class="headerdisplayname" style="display:inline;">Mittente: </div>Giulio <giulio@gmx.com></td></tr><tr><td><div class="headerdisplayname" style="display:inline;">Data: </div>22/06/2021, 14:28</td></tr></table><table border=0 cellspacing=0 cellpadding=0 width="100%" class="header-part2"><tr><td><div class="headerdisplayname" style="display:inline;">A: </div>Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com></td></tr><tr><td><div class="headerdisplayname" style="display:inline;">CC: </div>Frédéric Pierret <frederic.pierret@qubes-os.org></td></tr></table><br>
|
||
|
|
<div class="moz-text-flowed" style="font-family: -moz-fixed; font-size: 14px;" lang="x-unicode">Hello,
|
||
|
|
<br>thank you for the detailed response.
|
||
|
|
<br>
|
||
|
|
<br>Il 22/06/2021 04:43, Marek Marczykowski-Górecki ha scritto:
|
||
|
|
<br><blockquote type=cite style="color: #007cff;">Hi, I'm replying to both emails at once:
|
||
|
|
<br>
|
||
|
|
<br>On Sun, Jun 20, 2021 at 10:50:04PM +0200, Giulio wrote:
|
||
|
|
<br><blockquote type=cite style="color: #007cff;">Questions:
|
||
|
|
<br>
|
||
|
|
<br>1) Should we both support internal port forwarding and external port
|
||
|
|
<br>forwarding? Such as exposing a port for another domain or exposing a
|
||
|
|
<br>port through the public network interface? I would say yes.
|
||
|
|
<br></blockquote>
|
||
|
|
<br>Yes, I think so. Technically, those two cases should be quite similar.
|
||
|
|
<br>See also the case of sys-vpn much lower in the email.
|
||
|
|
<br>
|
||
|
|
<br></blockquote>
|
||
|
|
<br>I think that I'm actually failing to picture all the possible internal
|
||
|
|
scenarios.
|
||
|
|
<br>
|
||
|
|
<br>1) In the case of external port forwarding <sys-net> should forward to
|
||
|
|
<sys-firewall> and <sys-firewall> then to the <appvm>.
|
||
|
|
<br>In this case the port gets forwarded on the external interface ie: a LAN
|
||
|
|
or a public ip address depending on the network environment.
|
||
|
|
<br>2) In the case of internal port forwarding, the port is forwarded only
|
||
|
|
from <sys-firewall> to <appvm>. In that case, another <appvm2> can visit
|
||
|
|
the <appvm> service using <sys-firewall> ip address and the chosen port.
|
||
|
|
<br>In this case, the ports get exposed on <sys-firewall> and thus depending
|
||
|
|
on how the rules are implemented, may be available to all the AppVMs
|
||
|
|
that share the same <sys-firewall>.
|
||
|
|
<br>
|
||
|
|
<br>In both cases may be important to allow to specify access rules for the
|
||
|
|
forwarded port, such as the lan/public ip addresses ranges allowed for
|
||
|
|
case 1 and the appvm name for case 2.
|
||
|
|
<br>
|
||
|
|
<br><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;">3) Since the expire= feature seems to be already implemented (and
|
||
|
|
<br>limited for the expiring full outgoing access) would it be useful to be
|
||
|
|
<br>implemented in gui and cli for every rule? I would say yes since the
|
||
|
|
<br>admin and agent code seems to be already there. The same goes for the
|
||
|
|
<br>"comment=" field.
|
||
|
|
<br></blockquote>
|
||
|
|
<br>Per-rule expire may be tricky to handle at the GUI level, I have no idea
|
||
|
|
<br>how to make the UI for this not very confusing...
|
||
|
|
<br>But the comment field is definitely useful to use.
|
||
|
|
<br>
|
||
|
|
<br></blockquote>
|
||
|
|
<br>How do you see the same checkbox that actually allows full internet
|
||
|
|
access with the 5 minutes expiration time, displayed also on the window
|
||
|
|
for adding a rule?
|
||
|
|
<br>However I think there is time to think more through this as the UI will
|
||
|
|
be the last component.
|
||
|
|
<br>
|
||
|
|
<br><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;">4) How would you implement the management of forwarding rules in the
|
||
|
|
<br>network providing domain (sys-net)? Shall the user add a rule both in
|
||
|
|
<br>the target domain (ie the one with webserver and another one in sys-net)
|
||
|
|
<br>or should it be fully automatic from the first?
|
||
|
|
<br></blockquote>
|
||
|
|
<br>From the user point of view, I think it should be automated as much as
|
||
|
|
<br>possible. Like, let the user choose which port in which VM redirect to
|
||
|
|
<br>where. There may be cases when such redirection won't be possible - if
|
||
|
|
<br>there is no network path between the two points.
|
||
|
|
<br>
|
||
|
|
<br></blockquote>
|
||
|
|
<br>I agree with you. We might just check when the user adds an internal
|
||
|
|
forwarding rule if both the source and the destination shares the same
|
||
|
|
<firewallvm>, don't we?
|
||
|
|
<br>
|
||
|
|
<br>
|
||
|
|
<br><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;">5) Users should be able to set forward rules using domain names and not
|
||
|
|
<br>static ip addresses. In this case, the actual ip addresses of the dst
|
||
|
|
<br>domains should be collected in a similr way as currently DNS are
|
||
|
|
<br>resolved in `/core-agent-linux/qubesagent/firewall.py`, would this be good?
|
||
|
|
<br></blockquote>
|
||
|
|
<br>But here we are mostly talking about IP addresses of different VMs,
|
||
|
|
<br>right? Those can (and should) be resolved at core-admin side, so the VM
|
||
|
|
<br>applying the rules will have all the IP given. In fact VM may not be
|
||
|
|
<br>able to resolve IP of another VM at all.
|
||
|
|
<br>
|
||
|
|
<br></blockquote>
|
||
|
|
<br>Thanks for the insight, it totally makes sense.
|
||
|
|
<br>
|
||
|
|
<br><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;">Proposed XML Syntax:
|
||
|
|
<br><rule>
|
||
|
|
<br> <properties>
|
||
|
|
<br> <property name="action">forward</property>
|
||
|
|
<br> <property name="proto">udp</property>
|
||
|
|
<br> <property name="dstports">443-8080-5555</property>
|
||
|
|
<br> </properties>
|
||
|
|
<br><rule>
|
||
|
|
<br></blockquote>
|
||
|
|
<br>I don't see an important information here: forward to <span class="moz-txt-underscore"><span class="moz-txt-tag">_</span>where<span class="moz-txt-tag">_</span></span>.
|
||
|
|
<br>
|
||
|
|
<br><blockquote type=cite style="color: #007cff;">Proposed Admin API Syntax:
|
||
|
|
<br>action=forward proto=udp dstports=443-8080-5555 [expire=<unix
|
||
|
|
<br>timestamp>] [comment=random text]
|
||
|
|
<br></blockquote>
|
||
|
|
<br>Similar here, there needs to be a forward target (IP, and possibly a
|
||
|
|
<br>port)
|
||
|
|
<br>
|
||
|
|
<br>On Tue, Jun 22, 2021 at 01:49:15AM +0200, Giulio wrote:
|
||
|
|
<br><blockquote type=cite style="color: #007cff;">Since in the case of port forwarding the target ip address would always be the <vmname> IP address,
|
||
|
|
<br></blockquote>
|
||
|
|
<br>This is very true. But there needs to be an information where to forward
|
||
|
|
<br>the traffic to (as noted earlier). Plus, possibly a second set of ports
|
||
|
|
<br>(if you want to redirect to a different port).
|
||
|
|
<br>
|
||
|
|
<br></blockquote>
|
||
|
|
<br>I am still failing to understand something here, could you give me a an
|
||
|
|
example on when the dsthosts would different rather than the <appvm> or
|
||
|
|
<firewallvm> ip?
|
||
|
|
<br>
|
||
|
|
<br>
|
||
|
|
<br><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;">I think my main concern now is the question 4 from the aforementioned
|
||
|
|
<br>email. Shall the rules be automatically implemented in all 3 involved
|
||
|
|
<br>vms? (<netvm,firewallvm,appvm>). I think yes, because otherwise it would
|
||
|
|
<br>be counterintuitive to be a partially manual and partially automatic
|
||
|
|
<br>operation. But since it actually 'automatically' exposes more attack
|
||
|
|
<br>surface, by loosening up all 3 vms network rules, I guess that maybe
|
||
|
|
<br>more reasoning on it would be helpful.
|
||
|
|
<br></blockquote>
|
||
|
|
<br>Yes, but you need to pass the traffic somehow. The direct connection can
|
||
|
|
<br>be achieved with qvm-connnect-tcp (connecting to the target directly
|
||
|
|
<br>using qrexec, bypassing intermediate VMs), but it has its limits (low
|
||
|
|
<br>performance, TCP only). To keep it as actual IP traffic, you need to
|
||
|
|
<br>change firewall rules at all intermediate VMs too.
|
||
|
|
<br>
|
||
|
|
<br>Lets have a specific example: in default setup, redirect TCP port 80
|
||
|
|
<br>from the outside, to 'work' VM port 8080.
|
||
|
|
<br>
|
||
|
|
<br>The setup looks like this:
|
||
|
|
<br>
|
||
|
|
sys-net -> sys-firewall -> work
|
||
|
|
<br>
|
||
|
|
<br>For this, you will need those rules:
|
||
|
|
<br>
|
||
|
|
<br>1. In sys-net: forward TCP port 80 to sys-firewall
|
||
|
|
<br>2. In sys-firewall: forward TCP port 80 to work, port 8080
|
||
|
|
<br>3. In work: allow TCP port 8080
|
||
|
|
<br>
|
||
|
|
<br>Now is the important design question: how to store those rules? If you
|
||
|
|
<br>store them at all three places separately, it
|
||
|
|
<br>will be easier to apply them at runtime, but it will be harder to
|
||
|
|
<br>correlate them in UI. Plus, if any of them get modified/removed, it may
|
||
|
|
<br>be non-trivial to troubleshoot the issue.
|
||
|
|
<br>The other approach is to store the forward rules only in one place (the
|
||
|
|
<br>target, 'work' in this example? or the source, 'sys-net' here?). This
|
||
|
|
<br>way, it's harder to mess thing up. But when applying the rules (building
|
||
|
|
<br>rule sets for qubes-firewall service in all the involved VMs), you need
|
||
|
|
<br>to check several places.
|
||
|
|
<br>Plus, the UI should clearly show such redirected ports at all involved
|
||
|
|
<br>places, because it does affect system security - it must be easy to spot
|
||
|
|
<br>if any redirects are enabled.
|
||
|
|
<br>
|
||
|
|
<br>
|
||
|
|
<br>To make things more complex (sorry...), there may be a VPN or other
|
||
|
|
<br>proxy service (Tor?) involved. For example:
|
||
|
|
<br>
|
||
|
|
<br>sys-net -> sys-firewall -> sys-vpn -> work
|
||
|
|
<br>
|
||
|
|
<br>In such a case, the "external" VM for 'work' is not really sys-net, but
|
||
|
|
<br>rather sys-vpn. And actually you need to be careful to not accidentally
|
||
|
|
<br>bypass VPN either by allowing 'work' to communicate outside of the VPN,
|
||
|
|
<br>or (maybe even worse) systems on the LAN (via sys-net) reach inside VPN.
|
||
|
|
<br>
|
||
|
|
<br>This case is not easy to solve, because currently core-admin has no idea
|
||
|
|
<br>whether sys-vpn (or other such VM) do any of such tunnelling. Maybe we
|
||
|
|
<br>need to (finally) introduce some flag to mark such VMs?
|
||
|
|
<br>
|
||
|
|
<br>
|
||
|
|
<br>And another question: what should happen if you change netvm of 'work'.
|
||
|
|
<br>For example switch to something like:
|
||
|
|
<br>
|
||
|
|
<br> sys-net -> sys-firewall -> (other VMs, but not 'work')
|
||
|
|
<br>
|
||
|
|
<br> sys-wifi -> work
|
||
|
|
<br>
|
||
|
|
<br>Should the redirection stay active via sys-wifi? I think it should not,
|
||
|
|
<br>at least not automatically (maybe have an option for that?).
|
||
|
|
<br>
|
||
|
|
<br></blockquote>
|
||
|
|
<br>I understand all of your points and consequently it is hard to figure
|
||
|
|
out a catch-all solution.
|
||
|
|
<br>
|
||
|
|
<br>I tried charting the flow of the possible solution.
|
||
|
|
<br><a class="moz-txt-link-freetext" href="https://git.lsd.cat/Qubes/gsoc/src/master/assets/implementation.png">https://git.lsd.cat/Qubes/gsoc/src/master/assets/implementation.png</a>
|
||
|
|
<br>
|
||
|
|
<br>As a sum up:
|
||
|
|
<br>1) Rules are stored only in <appvm>/firewall.xml
|
||
|
|
<br>2) Rules can either be internal or exteranl (ie: they are applied only
|
||
|
|
to <firewallvm> or both to <firewallvm> and its <netvm>)
|
||
|
|
<br>3) Forwarding rules should be purged if <appvm> changes <firewall>
|
||
|
|
(maybe also if <firewallvm> changes <netvm>? But that would be harde to
|
||
|
|
detect I guess)
|
||
|
|
<br>4) Users should be able to specify both the forwarded port and
|
||
|
|
destination port as you were saying
|
||
|
|
<br>5) Users should be able to eventually restrict forwarding to designated
|
||
|
|
networks (with 0.0.0.0/0 being the wildcard instead of being a wildcard
|
||
|
|
by default)
|
||
|
|
<br>
|
||
|
|
<br>However, in this case it will surely be harder to display the rules in
|
||
|
|
all the affected vms.
|
||
|
|
<br>The other approach, as you were suggesting, of adding each specific rule
|
||
|
|
in each vm conf does make sense, but I think then it would necessary
|
||
|
|
something to keep track of the rule dependencies (such as a unique
|
||
|
|
identifier). Furthermore there is a higher risk of having orphaned rules
|
||
|
|
or a inconsistent state.
|
||
|
|
<br>
|
||
|
|
<br>Furthermore, in the "internal" vpn case that I have in mind, the idea is
|
||
|
|
to forward the local port via the VPN interface or Tor (but in the Tor
|
||
|
|
case users should just stick to Whonix). Some providers, such as
|
||
|
|
Mullvad, AirVPN, PIA etc allows port forwarding this way and I think
|
||
|
|
that's the most relevant case since it allows exposing a service on the
|
||
|
|
internet while maintaining a bit of privacy/anonimity and whithout
|
||
|
|
needing to bypass the local network NAT. Is this the same case you are
|
||
|
|
referring to?
|
||
|
|
<br>
|
||
|
|
<br><blockquote type=cite style="color: #007cff;">And finally, don't forget IPv6 exists. Which means you can have the same
|
||
|
|
<br>port on IPv4 and IPv6. And theoretically they could be redirected to
|
||
|
|
<br>different places (but I'm not sure if that's a good idea...).
|
||
|
|
<br>
|
||
|
|
<br></blockquote>
|
||
|
|
<br>I think that once we have figured out the overall logic to implement, it
|
||
|
|
should not be hard to duplicate it for ipv4/ipv6. I think the main
|
||
|
|
problem to think about is to insert proper checks to prevent users from
|
||
|
|
adding mixed rules.
|
||
|
|
<br>
|
||
|
|
<br>Cheers
|
||
|
|
<br>Giulio
|
||
|
|
<br></div></body>
|
||
|
|
</html>
|
||
|
|
</table></div>
|