Routed IPsec using if_ipsec VTI interfaces

jimp · May 30, 2018, 11:19 PM

The next round of 2.4.4 snapshots will include support for routed IPsec using FreeBSD's if_ipsec(4) virtual tunnel interfaces (VTI).

With a VTI setup, you define the local/remote network in a P2 entry set to VTI, and these addresses become the local and remote addresses of an ipsecX interface. This interface can then be assigned and enabled (Set IP type to None, just like OpenVPN), and then you can setup static routes, it gets a gateway to use with policy routing, and you can use routing protocols with it as well. Though that last part is not well-tested since FRR has a zebra crash issue going on at the moment on 2.4.4 snaps. You also get interface-specific IPsec firewall rules and NAT, you can packet capture on the interface, and get traffic data.

As things get tested/fixed we'll have a better write-up on how to set it all up, but it's fairly straightforward.

obrienmd · Jun 3, 2018, 4:59 PM

Hi Jim, I tested in 2.4.4 snapshot (6/2), no connectivity yet - posted my results / some info in the redmine feature:

https://redmine.pfsense.org/issues/8544

jimp · Jun 4, 2018, 2:49 PM

Let's keep the discussion here and not on Redmine.

Can you show me:

The output of ifconfig -a or at least the ipsecX interface (where X = that tunnels reqid, which looks to be 1 from your redmine comment)
The contents of the state table for those IP addresses (e.g. pfctl -ss | grep 192.168.240)
The config.xml contents for the IPsec tunnel P1 and P2, and the assigned interface

On mine, the interface does not show as a /24 and it shouldn't since it's a peer-to-peer interface:

ipsec2: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> metric 0 mtu 1400
        tunnel inet 203.0.113.14 --> 203.0.113.2
        inet6 fe80::20d:b9ff:fe33:f70%ipsec2 prefixlen 64 scopeid 0xb 
        inet 10.2.222.2 --> 10.2.222.1 netmask 0xff000000 
        nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL>
        reqid: 2
        groups: ipsec

I do have a static route on the interface, however:

: netstat -rn | egrep '(ipsec2|10.2.222)'
10.2.0.0/24        10.2.222.1         UGS      ipsec2
10.2.222.1         link#11            UH       ipsec2
10.2.222.2         link#11            UHS         lo0
fe80::%ipsec2/64                  link#11                       U        ipsec2
fe80::20d:b9ff:fe33:f70%ipsec2    link#11                       UHS         lo0

obrienmd · Jun 4, 2018, 4:02 PM

Thanks Jim,

Yours shows the same netmask mine does when I use a 10. for the tunnel IPs... I think it's setting some default based on which RFC1918 type space it's in - like Windows auto-populates a /24 netmask for 192.168 addresses and a /8 for 10. addresses (strange, but true).

Here's the info - note, I changed the addresses to 192.168.241.1(Texas) and 192.168.242.1 (Washington) for troubleshooting, but here goes:

There was no tunnel inet when i checked the Texas side's ifconfig. I disabled and re-enabled the interface, and it showed up... Still not passing traffic but pings just timed out, rather than giving the sendto error. Firewall logs don't show any packets blocked on WAN or ipsec (I have both basically open between those IPs for testing).

ifconfig:

#Washington
ipsec2: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> metric 0 mtu 1500
        tunnel inet {washington_wan_ip) --> {texas_wan_ip)
        inet6 fe80::208:a2ff:fe0a:7291%ipsec2 prefixlen 64 scopeid 0x9
        inet 192.168.241.1 --> 192.168.242.1 netmask 0xffffff00
        nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL>
        reqid: 2
        groups: ipsec

#Texas - strange, no tunnel inet
ipsec1: flags=8011<UP,POINTOPOINT,MULTICAST> metric 0 mtu 1400
        inet 192.168.242.1 --> 192.168.241.1 netmask 0xffffff00
        inet6 fe80::215:5dff:fe1f:d816%ipsec1 prefixlen 64 tentative scopeid 0x8
        nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL>
        reqid: 1
        groups: ipsec

states:

#Washington
ipsec2 icmp 192.168.241.1:56811 -> 192.168.242.1:56811       0:0

#Texas
ipsec1 icmp 192.168.242.1:25691 -> 192.168.241.1:25691       0:0

config.xml:

#Washington
        <interfaces>
                <opt1>
                        <descr><![CDATA[Texas]]></descr>
                        <if>ipsec2</if>
                        <spoofmac></spoofmac>
                        <enable></enable>
                </opt1>
        </interfaces>
        <ipsec>
                <phase1>
                        <ikeid>2</ikeid>
                        <iketype>ikev2</iketype>
                        <interface>wan</interface>
                        <remote-gateway>{texas_wan_ip}</remote-gateway>
                        <protocol>inet</protocol>
                        <myid_type>myaddress</myid_type>
                        <myid_data></myid_data>
                        <peerid_type>peeraddress</peerid_type>
                        <peerid_data></peerid_data>
                        <encryption>
                                <item>
                                        <encryption-algorithm>
                                                <name>aes</name>
                                                <keylen>256</keylen>
                                        </encryption-algorithm>
                                        <hash-algorithm>sha256</hash-algorithm>
                                        <dhgroup>2</dhgroup>
                                </item>
                        </encryption>
                        <lifetime>28800</lifetime>
                        <pre-shared-key>{my_pre_shared_key}</pre-shared-key>
                        <private-key></private-key>
                        <certref></certref>
                        <caref></caref>
                        <authentication_method>pre_shared_key</authentication_method>
                        <descr><![CDATA[TXDC]]></descr>
                        <nat_traversal>on</nat_traversal>
                        <mobike>off</mobike>
                        <margintime></margintime>
                        <dpd_delay>10</dpd_delay>
                        <dpd_maxfail>5</dpd_maxfail>
                </phase1>
                <client></client>
                <logging>
                        <dmn>1</dmn>
                        <mgr>1</mgr>
                        <ike>2</ike>
                        <chd>2</chd>
                        <job>1</job>
                        <cfg>2</cfg>
                        <knl>1</knl>
                        <net>1</net>
                        <asn>1</asn>
                        <enc>1</enc>
                        <imc>1</imc>
                        <imv>1</imv>
                        <pts>1</pts>
                        <tls>1</tls>
                        <esp>1</esp>
                        <lib>1</lib>
                </logging>
                <makebeforebreak></makebeforebreak>
                <uniqueids>yes</uniqueids>
                <phase2>
                        <ikeid>2</ikeid>
                        <uniqid>5b141a502eaff</uniqid>
                        <mode>vti</mode>
                        <reqid>1</reqid>
                        <localid>
                                <type>address</type>
                                <address>192.168.241.1</address>
                        </localid>
                        <remoteid>
                                <type>address</type>
                                <address>192.168.242.1</address>
                        </remoteid>
                        <protocol>esp</protocol>
                        <encryption-algorithm-option>
                                <name>aes256gcm</name>
                                <keylen>auto</keylen>
                        </encryption-algorithm-option>
                        <hash-algorithm-option>hmac_sha256</hash-algorithm-option>
                        <pfsgroup>14</pfsgroup>
                        <lifetime>3600</lifetime>
                        <pinghost></pinghost>
                        <descr><![CDATA[TXDC]]></descr>
                </phase2>
        </ipsec>

#Texas
        <interfaces>
                <opt2>
                        <descr><![CDATA[Washington]]></descr>
                        <if>ipsec1</if>
                        <spoofmac></spoofmac>
                        <enable></enable>
                </opt2>
        </interfaces>
        <ipsec>
                <phase1>
                        <ikeid>1</ikeid>
                        <iketype>ikev2</iketype>
                        <interface>wan</interface>
                        <remote-gateway>{washington_wan_ip}</remote-gateway>
                        <protocol>inet</protocol>
                        <myid_type>myaddress</myid_type>
                        <myid_data></myid_data>
                        <peerid_type>peeraddress</peerid_type>
                        <peerid_data></peerid_data>
                        <encryption>
                                <item>
                                        <encryption-algorithm>
                                                <name>aes</name>
                                                <keylen>256</keylen>
                                        </encryption-algorithm>
                                        <hash-algorithm>sha256</hash-algorithm>
                                        <dhgroup>2</dhgroup>
                                </item>
                        </encryption>
                        <lifetime>28800</lifetime>
                        <pre-shared-key>{my_pre_shared_key}</pre-shared-key>
                        <private-key></private-key>
                        <certref></certref>
                        <caref></caref>
                        <authentication_method>pre_shared_key</authentication_method>
                        <descr><![CDATA[Washington]]></descr>
                        <nat_traversal>on</nat_traversal>
                        <mobike>off</mobike>
                        <margintime></margintime>
                        <dpd_delay>10</dpd_delay>
                        <dpd_maxfail>5</dpd_maxfail>
                </phase1>
                <client></client>
                <logging>
                        <dmn>1</dmn>
                        <mgr>1</mgr>
                        <ike>2</ike>
                        <chd>2</chd>
                        <job>1</job>
                        <cfg>2</cfg>
                        <knl>1</knl>
                        <net>1</net>
                        <asn>1</asn>
                        <enc>1</enc>
                        <imc>1</imc>
                        <imv>1</imv>
                        <pts>1</pts>
                        <tls>1</tls>
                        <esp>1</esp>
                        <lib>1</lib>
                </logging>
                <makebeforebreak></makebeforebreak>
                <uniqueids>yes</uniqueids>
                <phase2>
                        <ikeid>1</ikeid>
                        <uniqid>5b141a51cd2c8</uniqid>
                        <mode>vti</mode>
                        <reqid>1</reqid>
                        <localid>
                                <type>address</type>
                                <address>192.168.242.1</address>
                        </localid>
                        <remoteid>
                                <type>address</type>
                                <address>192.168.241.1</address>
                        </remoteid>
                        <protocol>esp</protocol>
                        <encryption-algorithm-option>
                                <name>aes256gcm</name>
                                <keylen>auto</keylen>
                        </encryption-algorithm-option>
                        <hash-algorithm-option>hmac_sha256</hash-algorithm-option>
                        <pfsgroup>14</pfsgroup>
                        <lifetime>3600</lifetime>
                        <pinghost></pinghost>
                        <descr><![CDATA[Auburn]]></descr>
                </phase2>
        </ipsec>

obrienmd · Jun 4, 2018, 4:24 PM

Thought this was interesting, here's a tcpdump of the Washington WAN interface for all packets going to the Texas WAN IP.

During this dump, I was pinging the Texas side of the tunnel from the Washington side (both internal tunnel IPs, not WAN to WAN), but it seems that no ESP traffic was being sent.

#Washington
#tcpdump -vvnni {wan_int} "dst host {texas_wan_ip}"
tcpdump: listening on igb0, link-type EN10MB (Ethernet), capture size 262144 bytes
09:11:28.527012 IP (tos 0x0, ttl 64, id 48076, offset 0, flags [none], proto UDP (17), length 108)
    {washington_wan_ip}.500 > {texas_wan_ip}.500: [udp sum ok] isakmp 2.0 msgid 00000543 cookie 976f62e3b6408d81->db4ad4f0b86ca2a2: child_sa  inf2:
    (v2e: len=48)
09:11:38.550511 IP (tos 0x0, ttl 64, id 19398, offset 0, flags [none], proto UDP (17), length 108)
    {washington_wan_ip}.500 > {texas_wan_ip}.500: [udp sum ok] isakmp 2.0 msgid 00000544 cookie 976f62e3b6408d81->db4ad4f0b86ca2a2: child_sa  inf2:
    (v2e: len=48)
09:11:48.626281 IP (tos 0x0, ttl 64, id 22543, offset 0, flags [none], proto UDP (17), length 108)
    {washington_wan_ip}.500 > {texas_wan_ip}.500: [udp sum ok] isakmp 2.0 msgid 00000545 cookie 976f62e3b6408d81->db4ad4f0b86ca2a2: child_sa  inf2:
    (v2e: len=48)
09:11:58.715107 IP (tos 0x0, ttl 64, id 48507, offset 0, flags [none], proto UDP (17), length 108)
    {washington_wan_ip}.500 > {texas_wan_ip}.500: [udp sum ok] isakmp 2.0 msgid 00000546 cookie 976f62e3b6408d81->db4ad4f0b86ca2a2: child_sa  inf2:
    (v2e: len=48)
09:11:58.731185 IP (tos 0x0, ttl 64, id 19883, offset 0, flags [none], proto UDP (17), length 108)
    {washington_wan_ip}.500 > {texas_wan_ip}.500: [udp sum ok] isakmp 2.0 msgid 000005a3 cookie 976f62e3b6408d81->db4ad4f0b86ca2a2: child_sa  inf2[R]:
    (v2e: len=48)

jimp · Jun 4, 2018, 6:08 PM

The missing tunnel inet line is probably part of it, but I'm also wary of the netmask it added. I believe that should be /32 (0xFFFFFFFF) but I need to look into why it's ending up like it is. Possibly because of a default per block like you mentioned.

It can take an actual subnet there and not just a single address, so I may need to reconsider locking down the GUI controls like I have.

I'll work up a fix and push it shortly

obrienmd · Jun 4, 2018, 6:16 PM

Thanks, ready to test as soon as it's in snapshots!

obrienmd · Jun 4, 2018, 6:22 PM

Note, the reason I set them in different /24s is that I was hoping it would cause the traffic to "route" rather than going off into the ether somewhere (thinking the addy was in-net maybe and trying to just broadcast it), given it was destined for a different subnet, and the default it was giving me for 192.168/16 RFC1918 addresses was /24.

However, I still see traffic exiting the interface in tcpdump, but don't see any ESP going out the WAN. It's weird, I have no idea where the packets are going :)

Also, yours worked, and they were adjacent addresses in a /8 per your ifconfig entries, so that's a mystery :)

jimp · Jun 4, 2018, 6:29 PM

The far side of mine wasn't pfSense, but TNSR, and it was set to /24, pfSense was set to /8 (incorrectly) but yeah it was working, oddly enough.

I do see ESP on mine, yours is only showing IKE traffic so maybe your tunnel isn't even connecting?

Here are the changes I just pushed so that it will respect the subnet mask set for the local side, and so it applies the interface changes when IPsec is applied: https://github.com/pfsense/pfsense/commit/65767828cb2c6d14648bd86aced1b325b172ce43

They should show up in snapshots in a few hours or by the morning, but you can apply that with the system patches package for now.

obrienmd · Jun 4, 2018, 7:14 PM

OK, patches applied and both are now /32s in ifconfig :)

Still no traffic, will start digging on ipsec logs - P1s are "ESTABLISHED" and P2s show in IPSec/Overview. I'm not an ipsec guru, but I'll let you know what I find.

ipsec --statusall on each box:

#Washington
Status of IKE charon daemon (strongSwan 5.6.2, FreeBSD 11.2-RC1, amd64):
  uptime: 26 hours, since Jun 03 09:49:40 2018
  worker threads: 11 of 16 idle, 5/0/0/0 working, job queue: 0/0/0/0, scheduled: 3
  loaded plugins: charon unbound aes des blowfish rc2 sha2 sha1 md4 md5 random nonce x509 revocation constraints pubkey pkcs1 pkcs7 pkcs8 pkcs12 pgp dnskey sshkey ipseckey pem openssl fips-prf curve25519 xcbc cmac hmac curl attr kernel-pfkey kernel-pfroute resolve socket-default stroke vici updown eap-identity eap-sim eap-md5 eap-mschapv2 eap-dynamic eap-radius eap-tls eap-ttls eap-peap xauth-generic xauth-eap whitelist addrblock counters
Listening IP addresses:
  {all system IPs}
Connections:
        con2:  {washington_wan_ip}...{texas_wan_ip}  IKEv2, dpddelay=10s
        con2:   local:  [{washington_wan_ip}] uses pre-shared key authentication
        con2:   remote: [{texas_wan_ip}] uses pre-shared key authentication
        con2:   child:  192.168.241.1/32|/0 === 192.168.242.1/32|/0 TUNNEL, dpdaction=restart
Security Associations (1 up, 0 connecting):
        con2[8]: ESTABLISHED 114 minutes ago, {washington_wan_ip}[{washington_wan_ip}]...{texas_wan_ip}[{texas_wan_ip}]
        con2[8]: IKEv2 SPIs: 7a626268c7c2243f_i 94e70afd7a796de2_r*, pre-shared key reauthentication in 5 hours
        con2[8]: IKE proposal: AES_CBC_256/HMAC_SHA2_256_128/PRF_HMAC_SHA2_256/MODP_1024
        con2{48}:  INSTALLED, TUNNEL, reqid 3, ESP SPIs: c659a939_i c37c85a6_o
        con2{48}:  AES_GCM_16_256/MODP_2048, 77812 bytes_i, 0 bytes_o, rekeying in 22 minutes
        con2{48}:   192.168.241.1/32|/0 === 192.168.242.1/32|/0

#Texas
Status of IKE charon daemon (strongSwan 5.6.2, FreeBSD 11.2-RC1, amd64):
  uptime: 24 hours, since Jun 03 11:20:24 2018
  worker threads: 11 of 16 idle, 5/0/0/0 working, job queue: 0/0/0/0, scheduled: 4
  loaded plugins: charon unbound aes des blowfish rc2 sha2 sha1 md4 md5 random nonce x509 revocation constraints pubkey pkcs1 pkcs7 pkcs8 pkcs12 pgp dnskey sshkey ipseckey pem openssl fips-prf curve25519 xcbc cmac hmac curl attr kernel-pfkey kernel-pfroute resolve socket-default stroke vici updown eap-identity eap-sim eap-md5 eap-mschapv2 eap-dynamic eap-radius eap-tls eap-ttls eap-peap xauth-generic xauth-eap whitelist addrblock counters
Listening IP addresses:
  {all system IPs}
Connections:
   bypasslan:  %any...%any  IKEv1/2
   bypasslan:   local:  uses public key authentication
   bypasslan:   remote: uses public key authentication
   bypasslan:   child: {local_lan}/24|/0 === {local_lan}/24|/0 PASS
        con1:  {texas_wan_ip}...{washington_wan_ip}  IKEv2, dpddelay=10s
        con1:   local:  [{texas_wan_ip}] uses pre-shared key authentication
        con1:   remote: [{washington_wan_ip}] uses pre-shared key authentication
        con1:   child:  192.168.242.1/32|/0 === 192.168.241.1/32|/0 TUNNEL, dpdaction=restart
Shunted Connections:
   bypasslan:  {local_lan}/24|/0 === {local_lan}/24|/0 PASS
Security Associations (1 up, 0 connecting):
        con1[4]: ESTABLISHED 114 minutes ago,{texas_wan_ip[{texas_wan_ip}]...{washington_wan_ip}[{washington_wan_ip}]
        con1[4]: IKEv2 SPIs: 7a626268c7c2243f_i* 94e70afd7a796de2_r, pre-shared key reauthentication in 5 hours
        con1[4]: IKE proposal: AES_CBC_256/HMAC_SHA2_256_128/PRF_HMAC_SHA2_256/MODP_1024
        con1{37}:  INSTALLED, TUNNEL, reqid 1, ESP SPIs: c37c85a6_i c659a939_o
        con1{37}:  AES_GCM_16_256/MODP_2048, 0 bytes_i, 235872 bytes_o, rekeying in 23 minutes
        con1{37}:   192.168.242.1/32|/0 === 192.168.241.1/32|/0

jimp · Jun 4, 2018, 7:15 PM

Try putting them into a unique common /30, like you would an OpenVPN tunnel network.

If you need to route LAN-to-LAN then setup static routes or try a dynamic routing protocol (OSPF, BGP), though admittedly the routing protocol part has not been tested yet, only static routes.

obrienmd · Jun 4, 2018, 7:42 PM

Sure, I'm pretty comfortable with routing, just not the ipsec side. I'll try a shared /30.

Right now, status is very strange. Gateways are showing up (i.e. they can ping each other, and I see that in tcpdump), but when I try to ping from CLI, even using -S (to set source in the same subnet, just in case), I get nothing in reply. Very odd.

States are weird, some are going through enc0, some on the actual int.

Washington:
enc0 icmp 10.91.110.2:8558 <- 10.91.110.1:8558       0:0
ipsec1 icmp 10.91.110.2:10026 -> 10.91.110.1:10026       0:0
ipsec1 icmp 10.91.110.2:33717 -> 10.91.110.1:33717       0:0

Texas:
ipsec1 icmp 10.91.110.1:8558 -> 10.91.110.2:8558       0:0
enc0 icmp 10.91.110.1:33717 <- 10.91.110.2:33717       0:0

obrienmd · Jun 4, 2018, 7:48 PM

Eureka - putting them in a shared /30 seems to work. Now to see if all the GRE weirdness in previous uses get me with VTI :)

State matching: https://redmine.pfsense.org/issues/4479
Another: GREs sometimes open states before ipsec, then can't "get going" until states are cleared

jimp · Jun 4, 2018, 7:57 PM

@obrienmd said in Routed IPsec using if_ipsec VTI interfaces:

Eureka - putting them in a shared /30 seems to work. Now to see if all the GRE weirdness in previous uses get me with VTI :)

State matching: https://redmine.pfsense.org/issues/4479

I haven't tested TCP much but it shouldn't have the same issues.

Another: GREs sometimes open states before ipsec, then can't "get going" until states are cleared

That you can solve by putting floating rules outbound on WAN to stop your traffic from leaking and making incorrect states.

obrienmd · Jun 4, 2018, 10:46 PM

Thanks for all the help thus far Jim!

This is super exciting - it seems to work great thus far. The floating rules thing was always a little iffy for us (one in 5-6 reboots would get bad states even though non-IKE/ESP traffic was forbidden), though I'm with you in principle.

One last (I hope) weird issue: Firewall-originated traffic targeting anything outside the ipsec tunnel ip of the far firewall goes out the ipsec interface (i.e. route works as expected), but a dump of the far side interface doesn't show the traffic incoming. So:

From Texas LAN host to Washington firewall - pings, services work
From Texas LAN host to Washington LAN host - pings, services work
From Texas firewall to Washington firewall ipsec tunnel IP - pings, services work
From Texas firewall to Washington firewall LAN IP - pings, services fail (see outbound in tcpdump, not inbound on far side)
From Texas firewall to Washington LAN host - pings, services fail (see outbound in tcpdump, not inbound on far side)

jimp · Jun 5, 2018, 2:22 PM

I'll have to setup a better test to try that out, but I found an issue with the interface numbering/reqids that I need to fix before getting back to that.

obrienmd · Jun 5, 2018, 3:09 PM

@jimp yep, I think I'm seeing the reqid issue myself. Every few reboots, I get this complaint and no traffic flows:

Jun 5 08:08:59	charon		12[KNL] received an SADB_ACQUIRE with policy id 2 but no matching policy found
Jun 5 08:08:59	charon		12[KNL] creating acquire job for policy {near_wan_ip}/32|/0 === {far_wan_ip}/32|/0 with reqid {0}
Jun 5 08:08:59	charon		14[CFG] trap not found, unable to acquire reqid 0

jimp · Jun 5, 2018, 3:11 PM

If it works at all, it probably isn't the same issue. In my case I'm seeing the interface end up with one number but the reqid in the ipsec config has a different number, so no traffic ever reaches the interface due to the mismatch. That should be an all-or-nothing situation.

If you only have one P1/P2 or even only one P2 per P1 then it should be OK as-is, just by coincidence.

obrienmd · Jun 5, 2018, 3:13 PM

When I see those errors, it really doesn't work at all. I have connected P1s and P2s, but traffic isn't flowing at all (not the previous situation two posts up).

jimp · Jun 5, 2018, 9:04 PM

I just pushed some changes to how the IPsec interfaces are formed. The numbering of the interfaces has changed, so to be safe you should unassign the interface before upgrading. I'll work on some upgrade code in the morning, but it should hopefully now align better in terms of how strongswan forms the reqid vs how the if_ipsec interfaces want it so everything will line up better. I need to do some more testing, but it should be better.