4 years studying IT and a simple networking problem almost bowled us out…

We have an Airtel broadband connection for some months now. But sharing the connection between the roomies was getting difficult. So we got a Netgear wireless router hoping that now I can kill the bandwidth when my mate is playing AOC 😀

Happily, I setup the wireless router with the internet now getting to us as <Airtel Router>–><Netgear router>–><The home network using the ethernet ports and wireless access>. I did the same setup at my office and it worked happily. But the same thing wasn’t working out in the house no matter what I did. The comps at home cudn’t even ping to the Airtel router. We were thinking that the router has come with some kind of firewall…giving all kind of theories….thot that it the router software might have got crashed. Even started up ethereal to see what is happening at the network level.

While propounding the theories, we figured out that the new router has come up with its own dhcp and dns server and it allocates the addresses starting from (it reserves 1.1 for its own interface….and of course this is configurable once we can get the router up and running). Similar configuration of the dhcp and dns servers existed in Airtel router as well, so Airtel router also gave to itself and allocates ip’s after that. So, we configured the airtel router to give itself 1.4 and allocate from 1.5 onwards. This gave us a network with all hosts having unique IPs, yet the computers were not able to talk to the airtel router.

Next day, it struck that we are trying to configure a router over here (not a switch or hub), so it must have different networks/subnets on each of its side. That was it….configuring the new router’s dhcp to allocate up in 172.16 network resolved the issue.

What is more frustating about this story is that during the process of figuring out the problem we looked for the troubleshooting…which stressed that the routers should be powered on and connected in a specified order. Moreover, that clearing the browser cache can resolve the problems. We were sure that it was not possible, yet we were following the instructions as any layman would….there goes what I have learnt in those 4 years….haven’t yet moulded my brains to apply the concepts immediately.

This entry was posted in daily chores, tech. Bookmark the permalink.

One Response to 4 years studying IT and a simple networking problem almost bowled us out…

  1. P@R1 says:

    Interesting and informative posts for those who need to setup home n/w.

    DHCP does fire some neurons. The kinda work which I do requires my development environment to have certain windows services to be up and running all the time and to come daily and start them again seems a herculian task for me so I prefer not to reboot my machine
    One fine day i decided to reboot the system only to find that I am unable to login back. I ended up calling the IT helpdesk and learning a few things about Domain Controller and DHCP.

    It was deja vu for me as it had happened a number of times with me.In my previous company it also used to happen. As if after reboot the system was kicked out of domain.
    Later on I found out and could relate the networking knowledge to it.
    Companies which have a DHCP server have their own DHCP lifecycle whose primary job is to lease ip addresses to clients on network.
    Internally the DHCP server maintains an address pool from which it assigns addressed to clients.

    This what we all know? However at times accidentally the client is kicked out as well. Its my perception that a typical dhcp service is installed on a dc(Domain Controller), DC is widely used
    to authenticate on systems. After some discussion with the IT guy I could conclude that in volatile networks the object references for some network clients are accidentally compromised which form our first case of anomalies which cause domain alienation
    In some cases stale addresses are left in systems which cause conflict and throw certain systems out.
    In a leased environment such a scenario can occur when systems are typically taken off network while their expired ip address is reallocated to some other machine.

    In any case the solution lies in updating the domain information by first taking the client off the domain and after that re-adding the client to the same domain again.

    I was back into network in no time. In fact I kinda knew the solution all along but one needs domain admin passwords to do some major grinds here.

Leave a Reply

Your email address will not be published. Required fields are marked *