Guest User!

You are not Sophos Staff.

This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

SSL VPN issue after UTM upgrade to 9.404-5

Hello,

after the UTM upgrade from 9.403-4 to 9.404-5 the SSL VPN connection is no longer working. I changed nothing on the configuration.

Now I get following error message:

...

2016:06:28-12:24:27 firewall openvpn[9229]: SENT CONTROL [firewall]: 'PUSH_REQUEST' (status=1)

2016:06:28-12:24:27 firewall openvpn[9229]: TCPv4_CLIENT WRITE [56] to [AF_INET]213.136.68.103:44344 (via [AF_INET]10.10.10.254:35371): P_CONTROL_V1 kid=0 [ ] pid=5 DATA len=42
2016:06:28-12:24:27 firewall openvpn[9229]: TCPv4_CLIENT READ [22] from [AF_INET]213.136.68.103:44344 (via [AF_INET]10.10.10.254:35371): P_ACK_V1 kid=0 [ 5 ]
2016:06:28-12:24:27 firewall openvpn[9229]: TCPv4_CLIENT READ [466] from [AF_INET]213.136.68.103:44344 (via [AF_INET]10.10.10.254:35371): P_CONTROL_V1 kid=0 [ ] pid=6 DATA len=452
2016:06:28-12:24:27 firewall openvpn[9229]: PUSH: Received control message: 'PUSH_REPLY,topology subnet,route-gateway 192.168.55.1,route 192.168.54.0 255.255.255.0,route 192.168.55.0 255.255.255.0,setenv-safe remote_network_1 192.168.54.0/24,setenv-safe remote_network_2 192.168.55.0/24,setenv-safe local_network_1 192.168.5.0/24,setenv-safe local_network_2 192.168.111.0/24,setenv-safe local_network_3 192.168.250.0/24,setenv-safe local_network_4 192.168.110.0/24,ifconfig 192.168.54.6 192.168.54.5'
2016:06:28-12:24:27 firewall openvpn[9229]: OPTIONS IMPORT: --ifconfig/up options modified
2016:06:28-12:24:27 firewall openvpn[9229]: OPTIONS IMPORT: route options modified
2016:06:28-12:24:27 firewall openvpn[9229]: OPTIONS IMPORT: route-related options modified
2016:06:28-12:24:27 firewall openvpn[9229]: OPTIONS IMPORT: environment modified
2016:06:28-12:24:27 firewall openvpn[9229]: ROUTE_GATEWAY 10.10.10.1/255.255.255.0 IFACE=eth0.10 HWADDR=00:15:5d:6f:14:09
2016:06:28-12:24:27 firewall openvpn[9229]: TUN/TAP device tun1 opened
2016:06:28-12:24:27 firewall openvpn[9229]: TUN/TAP TX queue length set to 100
2016:06:28-12:24:27 firewall openvpn[9229]: do_ifconfig, tt->ipv6=0, tt->did_ifconfig_ipv6_setup=0
2016:06:28-12:24:27 firewall openvpn[9229]: /bin/ip link set dev tun1 up mtu 1500
2016:06:28-12:24:27 firewall openvpn[9229]: /bin/ip addr add dev tun1 192.168.54.6/11 broadcast 255.255.255.254
2016:06:28-12:24:27 firewall openvpn[9229]: /bin/ip route change dev tun1 192.168.54.4/11 proto 41 src 192.168.54.6
2016:06:28-12:24:27 firewall openvpn[9229]: MANAGEMENT: Client disconnected
2016:06:28-12:24:27 firewall openvpn[9229]: Linux ip route change failed: external program exited with error status: 2
2016:06:28-12:24:27 firewall openvpn[9229]: Exiting due to fatal error
2016:06:28-12:24:35 firewall openvpn[6482]: MANAGEMENT: Client disconnected

Because tun1 is not available I tryed to execute this command for a test on another interface and then I got following error message:

firewall:/var/sec/chroot-openvpn/etc/openvpn/conf.d # /bin/ip route change dev tun0 192.168.54.4/11 proto 41 src 192.168.54.6
RTNETLINK answers: Invalid argument

I hope you can help me!

Many Thanks!

Regards

Simon



This thread was automatically locked due to age.
Parents
  • Hi All,

    the engineering team is working on a fix for this issue. Unfortunately the fix was not ready in time to make it into the 9.405 maintenance release, the issue will be resolved with the next maintenance update.

    Jan

  • Groundhog Day:

    9.406003  released,...engineering team seems to be on holiday, problem still exists since 24 Mai.

    Great Job Sophos!

  • Above answer is not working.

    Same thing here. Problem has existed since upgrading from 9.355-1 to 9.405-5

    We fixed it by changing the openvpn binary back to 9.355-1 - this seems to work! But is a bad solution, because it breaks the idea of having an appliance. This might also void warranty - but when dealing with critical VPN connections being broken what do you do.... right?

    Premium 24/7 support - no solution by Sophos - They saw "wait for the next update" but no date is given. So the answer from Sophos seems to be "Wait for the next update"

    For reference, this is where it breaks on ours:

    The problem seems to be when the ip route script is called. The route it adds is a /10 that seems wrong to us. I've debugged what happens and this is the moment openvpn crashes with an error but Sophos seems to be unable to find the bug. A change in arguments in openvpn maybe?

    Version 9.405-5 (broken)

    2016:09:01-14:16:09 redacted openvpn[26247]: /bin/ip link set dev tun0 up mtu 1500 
    2016:09:01-14:16:09 redacted openvpn[26247]: /bin/ip addr add dev tun0 local 10.242.2.6 peer 10.242.2.5
    2016:09:01-14:16:09 redacted openvpn[26247]: /bin/ip route change dev tun0 10.242.2.4/10 proto 41 src 10.242.2.6   <<< error here!!!

     Version 9.355-1 Before upgrade (working) (2 weeks ago).

    2016:08:11-06:02:43 redacted openvpn[28719]: /bin/ip link set dev tun0 up mtu 1500
    2016:08:11-06:02:43 redacted openvpn[28719]: /bin/ip addr add dev tun0 local 10.242.2.6 peer 10.242.2.5
    2016:08:11-06:02:43 redacted openvpn[28719]: PLUGIN_CALL: POST /usr/lib/openvpn/plugins/openvpn-plugin-utm.so/PLUGIN_UP status=0
    2016:08:11-06:02:43 redacted openvpn[28719]: /bin/ip route add xxx.xx.0.0/16 proto 41 dev tun0 via 10.242.2.5

    Dear Sophos: Get your stuff together.

Reply
  • Above answer is not working.

    Same thing here. Problem has existed since upgrading from 9.355-1 to 9.405-5

    We fixed it by changing the openvpn binary back to 9.355-1 - this seems to work! But is a bad solution, because it breaks the idea of having an appliance. This might also void warranty - but when dealing with critical VPN connections being broken what do you do.... right?

    Premium 24/7 support - no solution by Sophos - They saw "wait for the next update" but no date is given. So the answer from Sophos seems to be "Wait for the next update"

    For reference, this is where it breaks on ours:

    The problem seems to be when the ip route script is called. The route it adds is a /10 that seems wrong to us. I've debugged what happens and this is the moment openvpn crashes with an error but Sophos seems to be unable to find the bug. A change in arguments in openvpn maybe?

    Version 9.405-5 (broken)

    2016:09:01-14:16:09 redacted openvpn[26247]: /bin/ip link set dev tun0 up mtu 1500 
    2016:09:01-14:16:09 redacted openvpn[26247]: /bin/ip addr add dev tun0 local 10.242.2.6 peer 10.242.2.5
    2016:09:01-14:16:09 redacted openvpn[26247]: /bin/ip route change dev tun0 10.242.2.4/10 proto 41 src 10.242.2.6   <<< error here!!!

     Version 9.355-1 Before upgrade (working) (2 weeks ago).

    2016:08:11-06:02:43 redacted openvpn[28719]: /bin/ip link set dev tun0 up mtu 1500
    2016:08:11-06:02:43 redacted openvpn[28719]: /bin/ip addr add dev tun0 local 10.242.2.6 peer 10.242.2.5
    2016:08:11-06:02:43 redacted openvpn[28719]: PLUGIN_CALL: POST /usr/lib/openvpn/plugins/openvpn-plugin-utm.so/PLUGIN_UP status=0
    2016:08:11-06:02:43 redacted openvpn[28719]: /bin/ip route add xxx.xx.0.0/16 proto 41 dev tun0 via 10.242.2.5

    Dear Sophos: Get your stuff together.

Children
No Data