[vpn-help] windows OK, linux does not connect
Fred Odendaal
fred.odendaal at gmail.com
Fri Jan 3 23:10:39 CST 2014
> Emre Erenoglu erenoglu at gmail.com
> Thu Jan 6 18:32:36 CST 2011
>
> On Fri, Jan 7, 2011 at 4:11 AM, Matthew Grooms <mgrooms at
shrew.net> wrote:
>
> > On 1/6/2011 5:47 PM, Emre Erenoglu wrote:
> >
> >> Dear Shrew Users,
> >>
> >> I have a strange problem. I'm using Shrew Soft client on my XP
> >> successfully, everything is working fine.
> >>
> >> I'm exporting the same configuration to my Linux system, it seems to
> >> connect fine since I get the "tunnel enabled" message and the tap0
> >> interface gets an address, however, the "security associations"
> >> "established" shows "0" and after some time "failed" startes to
> >> increase. Status shows "connected" and remote host shows the IP.
> >> Transport used is NAT-T / IKE / ESP. Fragmentation and Dead Peer
> >> Detection shows disabled although I enabled them in the config.
> >>
> >> I tried to search internet, saw settings about rp_filter, so I
set the
> >> following sysctl values and rebooted.
> >> net.ipv4.conf.default.rp_filter = 0
> >> net.ipv4.conf.all.rp_filter = 0
> >>
> >> Still no luck. My iptables is empty, there are no other firewalls
on the
> >> system. Do you have any idea why this Phase2 negotiation is
failing? I'm
> >> pasting the logs below. Please note that I changed the shown IP
> >> addresses by hand, so don't mind them unless necessary.
> >>
> >>
> > Your phase2 negotiation is not completing successfully. As a
result, you
> > don't have an IPsec SA to send traffic with. The kernel is sending an
> > ACQUIRE message appropriately, and the ike daemon is attempting to
negotiate
> > phase2 but is failing to get a response from the peer.
> >
> > BTW, what is 1.2.176.8? ...
> >
> >
> > ii : creating NONE INBOUND policy ANY:0.0.0.0:* -> ANY:1.2.176.8:*
> > K> : send pfkey X_SPDADD UNSPEC message
> > ii : creating NONE OUTBOUND policy ANY:1.2.176.8:* -> ANY:0.0.0.0:*
> > K< : recv pfkey X_SPDADD UNSPEC message
> > ii : created NONE policy route for 0.0.0.0/32
> >
> > If I recall correctly, these NONE policies get created is when
there is a
> > route to the peer, usually a default gateway. However, your next hop
> > shouldn't be at 1.2.176.8. Its not even close to 192.168.1.150. Do
you have
> > static entries in your route table for something?
> >
> > -Matthew
> >
>
> No,these are addresses I made up myself not to disclose server
addresses to
> a public mailing list. However, if the key to the solution is them,
I can
> send them intact. As far as I saw, those addresses were OK, one was the
> address assigned to me, other was the vpn server address.
>
> There was one thing in the logs:
> ii : received config pull response
> ii : - IP4 Address = 1.2.176.8
> ii : - Address Expiry = 0
> ii : - IP4 Netmask = 255.255.240.0
> ii : - IP4 DNS Server = 1.2.1.13
> ii : - IP4 DNS Server = 1.2.1.199
> ii : - IP4 Subnet = ANY:0.0.0.0/0:* ( invalid subnet ignored )
>
> Could the last ignore be an issue? Maybe I can test the same in windows.
>
> Any other clues?
>
> --
> Emre
> On 01/03/2014 11:06 PM, vpn-help-request at lists.shrew.net wrote:
I was having the exact same problem on 64-bit Fedora18 with the 64-bit
version 2.2.1 client. The same configuration is working on the Windows 7
version 2.2.1 client.
Even though my Fedora18 Linux OS is 64-bit, I switched to the 32-bit
version 2.2.1 client and it works.
Fred.
More information about the vpn-help
mailing list