Tag Info

Hot answers tagged

4

I think it's actually more debated than you make it appear. There is an admittedly old, related Linux FAQ: http://www.tldp.org/HOWTO/VPN-HOWTO/ I've used a PPP-over-ssh-over-ADSL for more than 12 years, and it never failed me, so from my experience I'd dare to say that the doomsayers probably largely exaggerate. TCP over TCP is probably a bad idea with RTC ...


3

As you can see, connecting back to your system isn't so simple. The essence of your problem is that even you can do the DNS registration (which makes eligible for the servers to find eachother), the actual ip connection between your machines is further impossible. There are multiple solutions for your problem. The simplest were if you have some type of ...


3

You'll have to block the outgoing traffic to this ip address not the other-way around. Use the OUTPUT keyword and the -d flag like this: iptables -A OUTPUT -d $ip -j DROP


3

There are two different ways of doing anycast: routing-based and on a single subnet. The routing based one can be done for both IPv4 and IPv6. The single-subnet way cannot be done with IPv4. Routing based anycast is done by announcing the same IPv4 or IPv6 prefix into a routing protocol from multiple routers. All of them announce they have a direct ...


1

Given the lack of answers, i am posting now a solution that is not very elegant, but simple: Run another OpenVPN instance on TCP for "bad clients" proto tcp and lower the TCP MSS on the client, e.g. iptables -t mangle -A POSTROUTING -p tcp --tcp-flags SYN,RST SYN -o ${OUT_DEV} -j TCPMSS --set-mss ${PATH-MTU-MINUS-40} An advantage of this solution is ...


1

It's not unusual on a small network to provide both of these functions from the same server. It's not a problem unless you had reason to expect DNS to be provided by a different system, in which case check the configuration of the DHCP server.


1

If your DNS and DHCP services are on the same server, that is not a problem. If these are on different machines, it could be that you have an IP address conflict in which case you must change the IP of one of your servers. About the DNS service itself, do you have anything in your Event Viewer ? What does an nslookup return ?


1

Each of your compute nodes will need a default route of 192.168.1.2 (head node eth2). You will then need to configure IP forwarding on the head node to allow the compute nodes access through it. You don't need to worry about the managed switch for this - it is "invisible" for the task you're trying to accomplish. You can also, if desired, enable masquerading ...


1

The First IP is the IP that has been resolved by the DNS-Server. The second IP should be your Router that tells you that the TTL expired in transit. So when the destination is not reachable. The last router gives you an error.


1

Yes, you can do this, it's not intuitive though. Also, you should have been able to find the documentation through Google? Here's the document: http://www.sonicwall.com/downloads/supporting_multiple_firewalled_subnets_on_sonicos_enhanced.pdf Example #1 Default NAT Mode with secondary subnet Create Static ARP entry for the gateway address of ...


1

You can create a sub-interface on your LAN interface (X0, usually) and use VLAN tagging to separate the traffic. In Network > Interfaces there is a "Add Interface" button.


1

Have you done any research on the topic? http://www.f5.com/it-management/solutions/global-load-balancing/overview/ While it is not trivial to implement... it certainly is not impossible. Generally speaking, it sounds like a Geographically sensitive DNS server would be your best bet. There are patches to allow Bind to do this in a split horizon setup. ...



Only top voted, non community-wiki answers of a minimum length are eligible