At 02:43 AM 3/6/2004 -0800, Ravi Kumar Munnangi wrote:
Users,
Iam working in a LAN with a number of systems. A system has X hardware address and another has Y hardware address. I want to set the hardware address of the second system also to X.
OK. Whether you can do this or not is a hardware issue ... some NICs support it, but others do not. In any case, read the man page for "ifconfig" (or maybe "ip", if your system does not use ifconfig) to see the Linux part of this.
I also want to set the primary IP addresses of both systems to 172.31.19.30.
OK. As you probably already know, you do this the same way you would set the machine to ANY IP address. The exact details are a bit distro spcecific, and you don't mention which Linux distro you are using ... you may use "ifconfig" directly, "ip" directly, or "ifup" in conjunction with a config file (usually /etc/network/interfaces).
So when I ping to 172.31.19.30, both the systems should respond.
Yes, because ping is pretty good at handling multiple addresses, and icmp is "connectionless". Whether more complex and TCP-based protocols like http will also work in this setup is less clear to me ... I'd wonder if anyone has ever tried it (except as an accidental misconfiguration).
The topology of LAN we are using is star. All systems are connected to a Switch. Does the settings change when we are using a bus topology?
Probably not ... but I am not certain what you mean by "bus" in this context. Are you referring to some sort of ring network (like an old-style thinnet LAN)?
Do I have to write a start up script? How should I write?
These are distro-level questions ... and to some extent kernel level. Post a followup with the usual details on your setup ... what distro, what version, what kernel version, what NICs and modules ... and I or someones else might be able to answer this one.
My actual goal is to start 2 web servers on two systems with same hardware address and same primary IP address but with different secondary IP addresses. So when a request comes from a client, the request has to be seen by both systems. I will write some mechanism by which only one will respond finally.
Before you spend too much time on this approach, you want to think through the "some mechanism" piece. Hand waving won't do here, and I don't see a sensible way to create a mechanism that will work.
My best *guess* is that you want to do one of two things --
1. Load balance -- in normal operation, have each server handle roughly half of the traffic.
2. Failover -- in normal operation, have one server handle all of the traffic, but have the second ready and waiting to "hot swap" in if the first should fail.
Either of these needs can be achieved in other, more conventional ways than what you propose to try. The details depend on the content of the Web site ... how much of it is dynamically generated, and how it is generated, how often its static content changes, how you intend to keep the two machines' versions of the Website in sync, and the like ... and may involve a third host (or even a fourth, if a large, shared database is involved).
- To unsubscribe from this list: send the line "unsubscribe linux-newbie" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.linux-learn.org/faqs
