Im trying to make an ubuntu server box my entrypoint to my networking. Meaning itll function as a server, a firewall, and a gateway. so i already installed dhcp3 and a dns server.
I have 2 ethernet cards in it. So now i wonder, should i the second card into a router's modem/wan port and make the router a switch? or should i plug it into one of the routers lan ports?
I have a LAN crosswire with rj45 connectors and I want to connect two laptops for data transfer as well as for socket prgramming to check my chat client. I am new to socket programming means also new to networking. Please tell me the basics to connect two computers(setting Ip addresses,gateway and other stuff related to networking) so that it helps me also in my programming and transferring data between computers. If it is not possible to describe everything here then give me a refernce to any book. I am using Ubuntu 10.04 on both computers. Please give me a solution to connect two computers for data transfer and socket programming.
View 2 Replies View RelatedI absolutely love Lubuntu on VirtualBox. I installed it to my Compaq Mini CQ-10 without issue. However, there are no drivers for the Wireless or the NIC. Ok, so I go to Broadcom and download the drivers. But, when I try to install with the Makefile, Terminal tells me that make is not installed! Now, how do I install my drivers without make?
View 4 Replies View RelatedI installed UBUNTU 10.10 alongside windows 7. Now the problem is: If i boot into UBUNTU first, ethernet does not work. But it works if i boot windows first and restart back into UBUNTU. Which means i have to boot windows first to make ethernet work on UBUNTU every time i start my PC.
View 9 Replies View RelatedI just fresh installed Fedora 11, and tried to connect via ethernet. I connected the cable and connection was started (in the network managed the two gray circled turned green) but then instead of connecting me the connection was aborted. I am behind a router if it helps and made sure eth0 is enabled. PS this is the output of "service network start":
Quote:
[root@Gal-PC subsys]# service network start
Bringing up loopback interface: [ OK ]
Bringing up interface eth0:
[code]......
I have 2 (some future machines will have 4) ethernet ports. I want to have them configured such that if any gets unplugged, as long as at least one of them is plugged in, it can reach the network (even if via a different IP address), and it can be reached (at least when trying a working IP address). I tried this for /etc/network/interfaces:
Code:
auto lo
iface lo inet loopback
auto eth0
iface eth0 inet static
[Code]...
The GUI for network configuration of Fedora is marvellous such that the configuration is almost fool-proof. But how can I make the connection by hand in the command line mode? It goes okay except the very last step. When I disconnect the eth0 interface from the right-hand side of the desktop GUI, I tested how to bring it back by command line but I failed. When disconnected, the ifconfig still shows the eth0 interface, with just the ip address portion changed.
I tried "ifconfig eth0 192.168.1.xx" to give it an ip address but the connection is still down even if the ifconfig shows an ip address for the eth0 interface. Then I tried "/etc/init.d/network restart" and "ifup eth0" and also "route add defaut gw 192.168.1.1" but none of these could accomplish the same work as a single click on the GUI to connect. I am very curious about how to do it in the terminal.
I cannot connect to my Ubuntu Server. Currently I'm trying via my MacBook through ethernet. A physical connection is apparent; however, the computers cannot communicate. To resolve this problem is important because I just need basic SSH functionality. While connected through ethernet, my MacBook cannot ping my Ubuntu box. Here is a screen shot of the Mac's network setup, a ping attempt, the ifconfig's of both machines.
[Code]...
I've set up 10.04 Server so that I could install directly to a command line, due to the fact that Desktop was crashing during the install every time. The installation goes fine, except for the networking portion. DHCP fails every time I try it. So I set up a static IP as an alternative. Once installation completes, none of the network-oriented tools (ping, telnet) work. I've tried pinging my router and I get 'Destination Host Unknown'. This is true if I change to a DHCP oriented setup as well.
The router sees that the machine is there, as indicated by the slowly blinking connection light meaning that it's hooked up to *something* but there's no activity. Also, networking did work within Windows before I nuked it, so I know it has the ability to operate correctly. I'm at a loss, mostly due to my own newness and ignorance of how to start tackling this within this environment.
i already have an eth0 configured to automatically get its info from dhcp.
but i wanna configure eth1 to be able to serve dhcp and dns. i havent configured either dns or dhcp server on the server box since i have not configured the serving interface.
so far my interfaces file is:
Code:
# The loopback network interface
auto lo
iface lo inet loopback
# The primary network interface
auto eth0
iface eth0 inet dhcp
[Code]....
I am using Ubuntu 10.04 Lucid and VMware 2.0 server. I have a an XP virtual machine that is needed for work. When I am at home and I place my cat 5 cable to my cisco catalyst switch, my XP VM recognizes the ethernet card on my laptop and bridges fine with ubuntu and works fine and I can get to my network. When I take my laptop to work and I go to a site and I plug it into a World Wide Packet 311 management port on the switch, my XP Virtual Machine does not see the network no matter what setting I set it too(100 full, 10 full, 10 half and etc). It simply says that the network card is disconnect. Ubuntu see the network fine and I can ping and talk to the network without any issues but the VM does not.
View 2 Replies View RelatedI have been working with Linux for about 10 years and I have a good understanding of IPv4 networking and routing but not so much detailed knowledge of implementation or specific networking packages.
I have a Linux system running a custom-built (Linux from Scratch) distribution and using 2.6.15 kernel. My system has 4 Ethernet controllers (eth0, eth1, eth2, and eth3). I have intentionally left out support for forwarding/routing of packets as this machine is not intended to be a router and I specifically do not want it to be able to so. My intent for the machine is to function as a server to 4 different and separate networks without allowing traffic or hosts on one LAN to access or adversely affect those on the other LANs.
My main question: is there a way to completely isolate the connections to these LANs and if so, how might I achieve it, and through what packages?
I have been having some issues:
1) Default gateway. With only a single default gateway, network requests from a LAN that is not directly connected to one of the 4 NICs and for which there may not be a static route defined, may send a response to the default gateway which is on one of the LANs connected to a different NIC than the one that I received the initial request on. Is there anyway to isolate outgoing messages to only go out on the NIC that the corresponding request was received through?
2) Default gateways. I have a requirement to allow DHCP on these interfaces. It is possible that the DCHP servers on each of the 4 LANs may provide a default gateway, but I don't see how I can make use of that. I know that there is a way to configure multiple default gateways via the iproute2 package (which is what I am using) but I don't know how to make sure that the correct default gateway is used for requests that were received on a specific interface.
My laptop was working fine on wireless till the userinterface changed and it defaulted to ethernet and now it won't let me go back on wireless How do i disable ethernet?
View 4 Replies View RelatedAfter using fedora 11 for a month or two now the ethernet gave out on me tonight. However if I switch to my ubuntu or windows XP install it runs fine. I made no recent changes to network connections, and no installed programs that should effect it.
kernel version: 2.6.29.6-213.fc11
default connection: eth0(and only connection)
ifcfg-eth0:
# Networking Interface
DEVICE=eth0
BOOTPROTO=dhcp
HWADDR=00:1F:C6:DB:05:0C
ONBOOT=yes
TYPE=Ethernet
IPV6INIT=no
USERCTL=no
NM_CONTROLLED=yes
NAME=eth0
UUID=5fb06bd0-0bb0-7ffb-45f1-d6edd65f3e03
DNS1=192.168.1.254
PEERDNS=yes
PEERROUTES=yes
I have tried, restarting, older kernel, restarting services, and clearing the DNS Cache. The specific error I get is that "Firefox can't find the server", and most other applications return similar. I can however ping websites still. All other computers on the network are running fine, and booting into another system the internet will work. Just not for fedora.
Cost effective (the people want cheap) solution to increase server storage space with as little impact on throughput (what I want) as possible. I have a server that sends instructions to 20 clients to perform certain tasks. The clients send gigs of data back to the server for storage.
BACKGROUND INFORMATION
Network: Gigabit Ethernet
Clients: (20) HP, Solaris, Linux
Linux Server: Dell R805 PowerEdge (QUAD Core Athlons 16 gigs of ram, 4 gigabit ports.
Selected directories are mounted (NFS) by clients. One directory to be mounted by the clients contains executable files. Mount commands are sent from the server (rsh) to the clients. Yes I know (ssh) but this is how the software was originally written so let's go with it.
The clients then receive commands to execute the binaries contained within the mounted directory. The binaries basically create an image of the clients filesystem including special applications. The client then sends the files and directories to the server which stores this �image�. This process can be reversed to restore a client to the initial baseline image.
In a nut shell, 20 clients are sending gigs of data to the server for storage. I need to attach a NAS (gigabit) to extend the storage capability of the server. The server would mount (NFS) to the hard drives on the NAS. The NAS, configured as a �direct attached storage� device (DAS), means the storage device is connected directly to the server and not the network (which makes it a DAS). This would provide direct storage expansion for the server. However, this configuration cannot be a bottleneck that significantly hinders performance.
Clients <---data---->Server<----storage---->DAS/NAS
1) Will a low end NAS (gigabit port) work? (NetGear 1TB ReadyNAS Duo, RND2110-100NAS) or something with a higher rated throughput (QNAP TS-239 Pro II)
2) Low end won't work, go with �?
3) Does the processor on the NAS play any role that would determine the performance of this configuration given that the server would mount directly to a directory on this device (DAS/NAS)?
I have a server with two ethernet ports. I configured eth0 to be static, set at 10.1.10.148. I plugged in another router into the other ethernet port in order to configure that router. I configured eth1 to use dhcp. Using /etc/network/interfaces rather than gnome network manager. When I did this, I lost internet connectivity (internet routes through eth0 of course)
- Why did I lose internet connectivity?
In order to recover internet activity, I had to disconnect the new router on eth1 of course, and do sudo ifdown eth1. That wasn't enough however. After rebooting numerous times and pulling out my hair, I finally tried configuring eth0 as dhcp, rather than static, and this fixed the problem.
- Why didn't sudo ifdown eth1 solve the problem? What information was saved between reboots that somehow remembered that I plugged in the new router? Because my thinking was if /etc/network/interfaces was identical, and the network topology was identical, after a reboot everything should be restored, but it wasn't.
I am trying to make my pc as a small web server. so i followed the steps below
1) Configure the router - Active Port Forwarding
Private IP Protocol Type Public Start Port Public End Port Connection 192.168.1.3 ALL 78 78
[Code]....
However, when i am trying to use public ip. it goes to the router's configuration page.
I have two PCs:
- PC1 with Ubuntu 10.4
- PC2 with DOS & Windows 3.11
They are connected by a serial null-modem cable. I used the PC2 as serial terminal, but now i wanna connect it to my LAN. How can I make my PC1 a SLIP server?
I have a Synology home server at home. I have a laptop with Linux Mint 8 (Helena) installed. I can access my server's shared files using LAN through Places>Network>Windows Network>Workgroup>Server>Shared files. I can also access the files through a web browser. Samba is installed. The problem is when I use the wireless lan of my laptop, I cannot access the server and the files anymore locally even through a web browser.
I tried Connect to Server through Windows Share but what I get is "Could not display "smb://server/".".
I have the following setup:
Location A:
Internet -> ISP 1 -> Firewall 1 (external IP address) -> Web Server 1 (internal IP address)
Location B:
Internet -> ISP 1 -> ISP 2 -> Firewall 2 (external IP address) -> Web Server 2 (internal IP address)
I want to find out if I can failover to another location without having to make changes on the DNS server. Just by redirecting traffic from ISP 1 router to ISP 2 router.
i have followed multiple guides to make my home server use a static ip address. no matter what i do it always reverts back to its dhcp address after about 2 minutes. how do i get the static ip address to stick, or be permanant.
View 9 Replies View RelatedThe cable internet I'm using runs on Dynamic DHCP IP addresses. I changed it to a static IP address in my router settings, but it keeps changing. This means that I can't connect to my home server from a remote location.
Is there anyway to run my ubuntu server on a DHCP IP address without connecting through my router 192.168.1.xxx?
I have been working in Linux for the past two years.And working in squid for the past 6 months.The following is my Setup
ETH1 - ISP1 (Active)
ETH2 - LAN (Active)
ETH3 - ISP2 (Inactive)
Using Iptables I forward 80 port through squid (3128).Other ports just get connected in the proxy machine itself since it has IP_FORWARD enabled. All my client machines browsers enabled with "Use proxy 3128". I just want to make all the port to pass through squid server.
I have a client's linux server where someone added a script which sends spam mails from the server's local loopback address (127.0.0.1) all my seniors worked on it and couldn't trace the script on the server this issue was going for a week, now I have an idea to bring down the loopback interface of that particular server(ya, that was not a permanent solution but just had an idea), but my support manager told that it will bring the whole server down(so far as i have learn't loopback address is used only for testing the NIC card's functionality), is that really bring the server down.?. How ever i tried this using a test machine all the connections works fine even after i bring down the loopback interface. As my client's server is a production server i can't test it there without clear knowledge, can any one help me out whether will it bring down the server or will it cause any side effects on it(i know that you guys are experts so i am asking it here)?
View 5 Replies View RelatedI'm trying to make a DHCP server work on my CentOS 5.4 box, but I have placed it on another subnet than the one that it is going to serve. I use a helper address on my router to make the the DHCP requests being routed to the DHCP server on the other subnet.
dhcpd.conf file:
ddns-update-style interim
ignore client-updates
ubnet 10.160.195.160 netmask 255.255.255.224
[code]...
I'd love a hint or two on the following problem. I've set up iptables rules to forward all connections to port 3306 to a non-standard mysql port on a remote server. This works, except that I need to deal with the loopback interface in a special way and I'm stuck.
Code:
iptables -t nat -A PREROUTING -p tcp --dport 3306 -j DNAT --to 128.XXX.XXX.XXX:3197
iptables -A FORWARD -p tcp -d 128.XXX.XXX.XXX --dport 3197 -j ACCEPT
iptables -t nat -A POSTROUTING -j MASQUERADE
Since locally-generated packets will never hit the PREROUTING rule, you'll need to setup a near identical rule using OUTPUT to make it work. Here is what I've tried:
[Code]...
When my pc is connected directly to the LAN modem, my server can go online, ip connects to the domain and everything is cool.
When my pc connected to the router which is connected to the modem, I can't make my server go online. It asks for some linksys authorisation.
How can I teach my server to ignore my router and go directly to the modem. Unfortunately I can't just connect my server to the modem couse 3 more PCs are connected with the hub.
I am not exactly newbie with GNU/Linux, but definitely new to SSH/NX stuffs. I followed the below tutorials for installing FreeNX so that I can access my desktop like TeamViewer in windows.
https:[url]....
https:[url].....
But I don't see the folder NX like the one mentioned here.Then edit the file /usr/NX/etc/server.cfg
I don't see a folder called NX to proceed. Also when I use nomachine client to connect from Windows, it couldn't connect and I get the below errors.
nxssh: <host-name>: no address associated with name.
how to make connect to freenx server from windows using nomachine client?
Is it possible to make this automatic, that is, how can I make the server active during the machine boot up?
View 4 Replies View RelatedI'm trying to find a command to tell nslookup, "I want you to try to resolve hostname X using DNS server Y, and if the lookup fails, just output failure do NOT fail over to some other DNS server as a backup and use it to try to resolve the same hostname."I'm trying to follow the man page instructions for doing this, but it doesn't seem to work. The man page for nslookup on my system (CentOS 5.5) includes the line:
"[no]fail Try the next nameserver if a nameserver responds with SERVFAIL or a referral (nofail) or terminate query (fail) on such a response."But if I try using that option, the output seems to indicate that after the lookup failed on ns1.afraid.org, nslookup failed over to 208.67.220.220 (the first nameserver listed in my resolv.conf -- an OpenDNS nameserver) and used it to resolve the hostname instead.
