gsoc/mails/20210622-Re_GSoC Port Forwarding-13697.html

395 lines
29 KiB
HTML
Raw Permalink Normal View History

2021-10-19 18:22:56 +02:00
<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>Marek Marczykowski-Górecki &lt;marmarek@invisiblethingslab.com&gt;</td></tr><tr><td><div class="headerdisplayname" style="display:inline;">Data: </div>22/06/2021, 16:04</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>Giulio <giulio@gmx.com></td></tr><tr><td><div class="headerdisplayname" style="display:inline;">CC: </div>Frédéric Pierret &lt;frederic.pierret@qubes-os.org&gt;</td></tr></table><br>
<div class="moz-text-plain" wrap=true graphical-quote=true style="font-family: -moz-fixed; font-size: 14px;" lang="x-unicode"><pre wrap class="moz-quote-pre">
On Tue, Jun 22, 2021 at 02:28:26PM +0200, Giulio wrote:
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>Hello,
<span class="moz-txt-citetags">&gt; </span>thank you for the detailed response.
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>Il 22/06/2021 04:43, Marek Marczykowski-Górecki ha scritto:
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; </span>Hi, I'm replying to both emails at once:
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>On Sun, Jun 20, 2021 at 10:50:04PM +0200, Giulio wrote:
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>Questions:
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>1) Should we both support internal port forwarding and external port
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>forwarding? Such as exposing a port for another domain or exposing a
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>port through the public network interface? I would say yes.
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>Yes, I think so. Technically, those two cases should be quite similar.
<span class="moz-txt-citetags">&gt; &gt; </span>See also the case of sys-vpn much lower in the email.
<span class="moz-txt-citetags">&gt; &gt; </span>
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>I think that I'm actually failing to picture all the possible internal
<span class="moz-txt-citetags">&gt; </span>scenarios.
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>1) In the case of external port forwarding &lt;sys-net&gt; should forward to
<span class="moz-txt-citetags">&gt; </span>&lt;sys-firewall&gt; and &lt;sys-firewall&gt; then to the &lt;appvm&gt;.
<span class="moz-txt-citetags">&gt; </span>In this case the port gets forwarded on the external interface ie: a LAN
<span class="moz-txt-citetags">&gt; </span>or a public ip address depending on the network environment.
</pre></blockquote><pre wrap class="moz-quote-pre">
Yes.
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>2) In the case of internal port forwarding, the port is forwarded only
<span class="moz-txt-citetags">&gt; </span>from &lt;sys-firewall&gt; to &lt;appvm&gt;. In that case, another &lt;appvm2&gt; can visit
<span class="moz-txt-citetags">&gt; </span>the &lt;appvm&gt; service using &lt;sys-firewall&gt; ip address and the chosen port.
</pre></blockquote><pre wrap class="moz-quote-pre">
Yes.
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>In this case, the ports get exposed on &lt;sys-firewall&gt; and thus depending
<span class="moz-txt-citetags">&gt; </span>on how the rules are implemented, may be available to all the AppVMs
<span class="moz-txt-citetags">&gt; </span>that share the same &lt;sys-firewall&gt;.
</pre></blockquote><pre wrap class="moz-quote-pre">
Yes.
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>In both cases may be important to allow to specify access rules for the
<span class="moz-txt-citetags">&gt; </span>forwarded port, such as the lan/public ip addresses ranges allowed for
<span class="moz-txt-citetags">&gt; </span>case 1 and the appvm name for case 2.
</pre></blockquote><pre wrap class="moz-quote-pre">
Yes, indeed.
</pre><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>3) Since the expire= feature seems to be already implemented (and
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>limited for the expiring full outgoing access) would it be useful to be
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>implemented in gui and cli for every rule? I would say yes since the
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>admin and agent code seems to be already there. The same goes for the
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>"comment=" field.
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>Per-rule expire may be tricky to handle at the GUI level, I have no idea
<span class="moz-txt-citetags">&gt; &gt; </span>how to make the UI for this not very confusing...
<span class="moz-txt-citetags">&gt; &gt; </span>But the comment field is definitely useful to use.
<span class="moz-txt-citetags">&gt; &gt; </span>
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>How do you see the same checkbox that actually allows full internet
<span class="moz-txt-citetags">&gt; </span>access with the 5 minutes expiration time, displayed also on the window
<span class="moz-txt-citetags">&gt; </span>for adding a rule?
</pre></blockquote><pre wrap class="moz-quote-pre">
This may be more relevant to longer times. With times like 5min, just
setting the rules up (if you want more than one of them) may already eat
up significant portion of the expiration time...
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>However I think there is time to think more through this as the UI will
<span class="moz-txt-citetags">&gt; </span>be the last component.
<span class="moz-txt-citetags">&gt; </span>
</pre><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>4) How would you implement the management of forwarding rules in the
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>network providing domain (sys-net)? Shall the user add a rule both in
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>the target domain (ie the one with webserver and another one in sys-net)
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>or should it be fully automatic from the first?
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>From the user point of view, I think it should be automated as much as
<span class="moz-txt-citetags">&gt; &gt; </span>possible. Like, let the user choose which port in which VM redirect to
<span class="moz-txt-citetags">&gt; &gt; </span>where. There may be cases when such redirection won't be possible - if
<span class="moz-txt-citetags">&gt; &gt; </span>there is no network path between the two points.
<span class="moz-txt-citetags">&gt; &gt; </span>
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>I agree with you. We might just check when the user adds an internal
<span class="moz-txt-citetags">&gt; </span>forwarding rule if both the source and the destination shares the same
<span class="moz-txt-citetags">&gt; </span>&lt;firewallvm&gt;, don't we?
</pre></blockquote><pre wrap class="moz-quote-pre">
I think we can simplify it even further: allow forwarding ports only
from (any of) upstream VMs. For example in this case:
appvm1 -&gt; sys-firewall -&gt; sys-net
| |
appvm2 ------+ |
|
appvm3 -&gt; some-other-firewall -+
Allow forwarding to appvm1 only from sys-net (external case) or
sys-firwall, but not appvm3 or some-other-firewall.
Then, within the forward rule configuration you can restrict access
rules (like you propose below, with default 0.0.0.0/0). This restriction
will work for VMs directly connected to sys-firewall only, because there
is NAT (sys-net does not know whether its appvm1 or appvm2 - it only
sees sys-firewall IP in those cases). But I think it's ok to make this
limitation and require VMs to be connected to the same &lt;firewallvm&gt; if
you want to forward traffic between them.
I think you did it right with the internal/external type.
Allowing forwarding from others (like some-other-firewall in the picture
above) may be tricky (and unreliable), as it will be hard to restrict
who can really connect (sys-net have no idea which VM behind
sys-firewall/some-other-firewall really connects).
</pre><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>5) Users should be able to set forward rules using domain names and not
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>static ip addresses. In this case, the actual ip addresses of the dst
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>domains should be collected in a similr way as currently DNS are
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>resolved in `/core-agent-linux/qubesagent/firewall.py`, would this be good?
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>But here we are mostly talking about IP addresses of different VMs,
<span class="moz-txt-citetags">&gt; &gt; </span>right? Those can (and should) be resolved at core-admin side, so the VM
<span class="moz-txt-citetags">&gt; &gt; </span>applying the rules will have all the IP given. In fact VM may not be
<span class="moz-txt-citetags">&gt; &gt; </span>able to resolve IP of another VM at all.
<span class="moz-txt-citetags">&gt; &gt; </span>
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>Thanks for the insight, it totally makes sense.
<span class="moz-txt-citetags">&gt; </span>
</pre><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>Proposed XML Syntax:
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>&lt;rule&gt;
<span class="moz-txt-citetags">&gt; &gt; &gt; </span> &lt;properties&gt;
<span class="moz-txt-citetags">&gt; &gt; &gt; </span> &lt;property name="action"&gt;forward&lt;/property&gt;
<span class="moz-txt-citetags">&gt; &gt; &gt; </span> &lt;property name="proto"&gt;udp&lt;/property&gt;
<span class="moz-txt-citetags">&gt; &gt; &gt; </span> &lt;property name="dstports"&gt;443-8080-5555&lt;/property&gt;
<span class="moz-txt-citetags">&gt; &gt; &gt; </span> &lt;/properties&gt;
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>&lt;rule&gt;
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>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>.
<span class="moz-txt-citetags">&gt; &gt; </span>
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>Proposed Admin API Syntax:
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>action=forward proto=udp dstports=443-8080-5555 [expire=&lt;unix
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>timestamp&gt;] [comment=random text]
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>Similar here, there needs to be a forward target (IP, and possibly a
<span class="moz-txt-citetags">&gt; &gt; </span>port)
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>On Tue, Jun 22, 2021 at 01:49:15AM +0200, Giulio wrote:
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>Since in the case of port forwarding the target ip address would always be the &lt;vmname&gt; IP address,
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>This is very true. But there needs to be an information where to forward
<span class="moz-txt-citetags">&gt; &gt; </span>the traffic to (as noted earlier). Plus, possibly a second set of ports
<span class="moz-txt-citetags">&gt; &gt; </span>(if you want to redirect to a different port).
<span class="moz-txt-citetags">&gt; &gt; </span>
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>I am still failing to understand something here, could you give me a an
<span class="moz-txt-citetags">&gt; </span>example on when the dsthosts would different rather than the &lt;appvm&gt; or
<span class="moz-txt-citetags">&gt; </span>&lt;firewallvm&gt; ip?
</pre></blockquote><pre wrap class="moz-quote-pre">
I'm talking about the other end of the connection. Let me summarize
this for different rules:
1. For allow/deny rules, it is about where the VM can connect to -
outgoing connection. The source IP is always the VM's IP (implicitly),
and the rule can specify destination IP, protocol, port.
2. For forward rules, it is about incoming connection - the destination
IP is VM's IP (implicitly), but then the connection is redirected to
somewhere else (like some appvm) - and the rule needs to point out to
where it should redirect.
If you have simplified case like this:
sys-net -&gt; appvm
Then, the rule in sys-net not only needs to know what to redirect (like,
TCP port 80), but also needs to know to redirect it to appvm, not
anywhere else.
Similarly, if the rule is stored with appvm, it needs to know to
install the redirect in sys-net and not some intermediate other VM (as
talked about internal, or VPN or Tor cases).
Ok, reading your response further, I think you covered it with
external/internal type, so it should be good.
</pre><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>I think my main concern now is the question 4 from the aforementioned
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>email. Shall the rules be automatically implemented in all 3 involved
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>vms? (&lt;netvm,firewallvm,appvm&gt;). I think yes, because otherwise it would
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>be counterintuitive to be a partially manual and partially automatic
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>operation. But since it actually 'automatically' exposes more attack
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>surface, by loosening up all 3 vms network rules, I guess that maybe
<span class="moz-txt-citetags">&gt; &gt; &gt; </span>more reasoning on it would be helpful.
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>Yes, but you need to pass the traffic somehow. The direct connection can
<span class="moz-txt-citetags">&gt; &gt; </span>be achieved with qvm-connnect-tcp (connecting to the target directly
<span class="moz-txt-citetags">&gt; &gt; </span>using qrexec, bypassing intermediate VMs), but it has its limits (low
<span class="moz-txt-citetags">&gt; &gt; </span>performance, TCP only). To keep it as actual IP traffic, you need to
<span class="moz-txt-citetags">&gt; &gt; </span>change firewall rules at all intermediate VMs too.
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>Lets have a specific example: in default setup, redirect TCP port 80
<span class="moz-txt-citetags">&gt; &gt; </span>from the outside, to 'work' VM port 8080.
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>The setup looks like this:
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span> sys-net -&gt; sys-firewall -&gt; work
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>For this, you will need those rules:
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>1. In sys-net: forward TCP port 80 to sys-firewall
<span class="moz-txt-citetags">&gt; &gt; </span>2. In sys-firewall: forward TCP port 80 to work, port 8080
<span class="moz-txt-citetags">&gt; &gt; </span>3. In work: allow TCP port 8080
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>Now is the important design question: how to store those rules? If you
<span class="moz-txt-citetags">&gt; &gt; </span>store them at all three places separately, it
<span class="moz-txt-citetags">&gt; &gt; </span>will be easier to apply them at runtime, but it will be harder to
<span class="moz-txt-citetags">&gt; &gt; </span>correlate them in UI. Plus, if any of them get modified/removed, it may
<span class="moz-txt-citetags">&gt; &gt; </span>be non-trivial to troubleshoot the issue.
<span class="moz-txt-citetags">&gt; &gt; </span>The other approach is to store the forward rules only in one place (the
<span class="moz-txt-citetags">&gt; &gt; </span>target, 'work' in this example? or the source, 'sys-net' here?). This
<span class="moz-txt-citetags">&gt; &gt; </span>way, it's harder to mess thing up. But when applying the rules (building
<span class="moz-txt-citetags">&gt; &gt; </span>rule sets for qubes-firewall service in all the involved VMs), you need
<span class="moz-txt-citetags">&gt; &gt; </span>to check several places.
<span class="moz-txt-citetags">&gt; &gt; </span>Plus, the UI should clearly show such redirected ports at all involved
<span class="moz-txt-citetags">&gt; &gt; </span>places, because it does affect system security - it must be easy to spot
<span class="moz-txt-citetags">&gt; &gt; </span>if any redirects are enabled.
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>To make things more complex (sorry...), there may be a VPN or other
<span class="moz-txt-citetags">&gt; &gt; </span>proxy service (Tor?) involved. For example:
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>sys-net -&gt; sys-firewall -&gt; sys-vpn -&gt; work
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>In such a case, the "external" VM for 'work' is not really sys-net, but
<span class="moz-txt-citetags">&gt; &gt; </span>rather sys-vpn. And actually you need to be careful to not accidentally
<span class="moz-txt-citetags">&gt; &gt; </span>bypass VPN either by allowing 'work' to communicate outside of the VPN,
<span class="moz-txt-citetags">&gt; &gt; </span>or (maybe even worse) systems on the LAN (via sys-net) reach inside VPN.
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>This case is not easy to solve, because currently core-admin has no idea
<span class="moz-txt-citetags">&gt; &gt; </span>whether sys-vpn (or other such VM) do any of such tunnelling. Maybe we
<span class="moz-txt-citetags">&gt; &gt; </span>need to (finally) introduce some flag to mark such VMs?
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>And another question: what should happen if you change netvm of 'work'.
<span class="moz-txt-citetags">&gt; &gt; </span>For example switch to something like:
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span> sys-net -&gt; sys-firewall -&gt; (other VMs, but not 'work')
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span> sys-wifi -&gt; work
<span class="moz-txt-citetags">&gt; &gt; </span>
<span class="moz-txt-citetags">&gt; &gt; </span>Should the redirection stay active via sys-wifi? I think it should not,
<span class="moz-txt-citetags">&gt; &gt; </span>at least not automatically (maybe have an option for that?).
<span class="moz-txt-citetags">&gt; &gt; </span>
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>I understand all of your points and consequently it is hard to figure
<span class="moz-txt-citetags">&gt; </span>out a catch-all solution.
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>I tried charting the flow of the possible solution.
<span class="moz-txt-citetags">&gt; </span><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>
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>As a sum up:
<span class="moz-txt-citetags">&gt; </span>1) Rules are stored only in &lt;appvm&gt;/firewall.xml
<span class="moz-txt-citetags">&gt; </span>2) Rules can either be internal or exteranl (ie: they are applied only
<span class="moz-txt-citetags">&gt; </span>to &lt;firewallvm&gt; or both to &lt;firewallvm&gt; and its &lt;netvm&gt;)
</pre></blockquote><pre wrap class="moz-quote-pre">
Please note there may be more VMs in between, some examples at:
<a class="moz-txt-link-freetext" href="https://www.qubes-os.org/doc/firewall/">https://www.qubes-os.org/doc/firewall/</a>
So, I think it's better to define it as:
1. internal: redirect from immediate parent only (VM's 'netvm' property)
2. external: redirect on the whole chain up to the top (follow 'netvm'
property until you get a VM with it set to None).
My idea of redirect source was to explicitly point where the redirection
starts, instead of this automatic internal/external. But indeed the
automatic may be easier to use.
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>3) Forwarding rules should be purged if &lt;appvm&gt; changes &lt;firewall&gt;
<span class="moz-txt-citetags">&gt; </span>(maybe also if &lt;firewallvm&gt; changes &lt;netvm&gt;? But that would be harde to
<span class="moz-txt-citetags">&gt; </span>detect I guess)
</pre></blockquote><pre wrap class="moz-quote-pre">
Or maybe it should be configurable? I'd hate to loose the configuration just
because I temporarily switched to another netvm (and then switched it
back)...
Or, allow both automatic internal/external and explicit redirect source.
Then, setting redirect type to 'internal' would set the redirect point to
the VM's direct upstream (and would automatically adjust if you change
netvm), setting to 'external' would follow the whole chain. But setting
to explicit sys-net for example would always try to apply rule there, if
reachable (and add no rules, if not reachable).
Does it make sense?
Maybe let me explain it on a diagram (see attachment). Especially see
how redirections to appvm3 has changed after switching it from sys-vpn
to sys-firewall: the "internal" redirection (in green) remained there,
but the "sys-vpn" one (in blue) did not.
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>4) Users should be able to specify both the forwarded port and
<span class="moz-txt-citetags">&gt; </span>destination port as you were saying
<span class="moz-txt-citetags">&gt; </span>5) Users should be able to eventually restrict forwarding to designated
<span class="moz-txt-citetags">&gt; </span>networks (with 0.0.0.0/0 being the wildcard instead of being a wildcard
<span class="moz-txt-citetags">&gt; </span>by default)
</pre></blockquote><pre wrap class="moz-quote-pre">
I wonder if this should be combined in the same rule, or maybe separate
rule for limiting? But maybe indeed putting it into the same rule may be
easier (avoids duplicating port numbers etc).
</pre><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>However, in this case it will surely be harder to display the rules in
<span class="moz-txt-citetags">&gt; </span>all the affected vms.
<span class="moz-txt-citetags">&gt; </span>The other approach, as you were suggesting, of adding each specific rule
<span class="moz-txt-citetags">&gt; </span>in each vm conf does make sense, but I think then it would necessary
<span class="moz-txt-citetags">&gt; </span>something to keep track of the rule dependencies (such as a unique
<span class="moz-txt-citetags">&gt; </span>identifier). Furthermore there is a higher risk of having orphaned rules
<span class="moz-txt-citetags">&gt; </span>or a inconsistent state.
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>Furthermore, in the "internal" vpn case that I have in mind, the idea is
<span class="moz-txt-citetags">&gt; </span>to forward the local port via the VPN interface or Tor (but in the Tor
<span class="moz-txt-citetags">&gt; </span>case users should just stick to Whonix). Some providers, such as
<span class="moz-txt-citetags">&gt; </span>Mullvad, AirVPN, PIA etc allows port forwarding this way and I think
<span class="moz-txt-citetags">&gt; </span>that's the most relevant case since it allows exposing a service on the
<span class="moz-txt-citetags">&gt; </span>internet while maintaining a bit of privacy/anonimity and whithout
<span class="moz-txt-citetags">&gt; </span>needing to bypass the local network NAT. Is this the same case you are
<span class="moz-txt-citetags">&gt; </span>referring to?
</pre></blockquote><pre wrap class="moz-quote-pre">
Yes, exactly.
</pre><blockquote type=cite style="color: #007cff;"><blockquote type=cite style="color: #007cff;"><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; &gt; </span>And finally, don't forget IPv6 exists. Which means you can have the same
<span class="moz-txt-citetags">&gt; &gt; </span>port on IPv4 and IPv6. And theoretically they could be redirected to
<span class="moz-txt-citetags">&gt; &gt; </span>different places (but I'm not sure if that's a good idea...).
<span class="moz-txt-citetags">&gt; &gt; </span>
</pre></blockquote><pre wrap class="moz-quote-pre">
<span class="moz-txt-citetags">&gt; </span>
<span class="moz-txt-citetags">&gt; </span>I think that once we have figured out the overall logic to implement, it
<span class="moz-txt-citetags">&gt; </span>should not be hard to duplicate it for ipv4/ipv6. I think the main
<span class="moz-txt-citetags">&gt; </span>problem to think about is to insert proper checks to prevent users from
<span class="moz-txt-citetags">&gt; </span>adding mixed rules.
</pre></blockquote><pre wrap class="moz-quote-pre">
Yes, that sounds about right.
<div class="moz-txt-sig">--
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab
</div>
</fieldset>
</pre></div><BR><DIV CLASS="moz-attached-image-container"><IMG CLASS="moz-attached-image" shrinktofit="yes" SRC="EmbeddedImages/0.jpg"></DIV><br><hr><br><div style="font-size:12px;color:black;"><img src="data:image/gif;base64,R0lGODdhDwAPAOMAAP///zEwYmJlzQAAAPr6+vv7+/7+/vb29pyZ//39/YOBg////////////////////ywAAAAADwAPAAAESRDISUG4lQYr+s5bIEwDUWictA2GdBjhaAGDrKZzjYq3PgUw2co24+VGLYAAAesRLQklxoeiUDUI0qSj6EoH4Iuoq6B0PQJyJQIAOw==">
<ul><li><a href="Attachments/network-graph.jpg">Attachments/network-graph.jpg</li></a></ul></div><div class='' ></div></body>
</html>