OpenSUSE :: SSH Script - Make Changes To Several Computers At Once In A Private Network
May 17, 2011
I'm trying to make changes to several computers at once in a private network. I have ssh keys setup so that I can ssh from the main computer to all the other computers without having to enter my password. So I wrote a bash script like for computers in `cat computer list` do ssh $computers #type changes here exit done
but instead of executing the commands on every computer, it opens a shell on every computer for me to type into, just as if I had used ssh myself and not in a script. how to get this to work? I want to be able to add lines at the comment like "echo "test" > /testFile.txt" and be able to see the changes on each machine, in this case see testFile.txt on every machine.
im trying to connect two computers on lan.One computer has: VMWare Workstation and has Opensuse 11.3 mounted in it.The other computer has: VMWare Player and has Opensuse 11.3 mounted in it.Both computers are connected to a switch with cables.I have followed this guide in both computers:Depanati singuri calculatorul!: Opensuse 11.3 - configure local networkin order to setup a network.In one computer, if i go to: Computer---Network---Network folder, i only see one machine. When in fact i could see both of them right
I am new with IP tables stuff and i have a problem.... i have a pc Contain a fedora OS and i want to make a small network (4 PCs Contain XP OS) and using the pc of fedora OS as a firewall i want to Prevent the ping (i think it called(ICMP)) in the privat network and prevent one of the PCs from Browsing Internet(prevent port 80 and 81 as i think) and i still don't know how to make the Internet go Through the firewall to the private network...
I have openSUSE 11.2 installed and i need to create a gateway server that allows virtual private network connections. I want to play with my friends some lan games, but we are in different networks, so i want to create this gateway server so we can connect with VPN clients to this server and play freely.
i am totally new to Opensuse and to filezilla. I formerly used winscp - a windows-client for ssh.now i have a linux box and want to connect to the server via ssh ing filezilla.Can i use the same keys as i used in WinSCP - where should i store them and the third question.I read something about a certain so called certificate-file that has to be created!?
My ISP has for a long time had a broken forward/reverese DNS so that my ADSL connection with static IP address resolves to a completely different IP address on a reverse query. This has not been a problem until I upgraded a remote server from 10.3 to 11.3 last weekend and now ssh connections from my ADSL connection to it using public/private keys are being rejected with the following message in /var/log/messages (IP addresses changed): Aug 10 12:00:32 penguin1 sshd[1270]: Address 83.175.246.243 maps to 83-175-246-243.static.dsl.aupex.com, but this does not map back to the address - Possible Break in Attempt!
But if I log in interactively with username & password, the connection succeeds. I've changed the StrictModes setting in sshd_config to 'no' but this hasn't resolved the problem. Obviously something in 11.3 is being stricter about this IP mismatch than it was in 10.2 (and no, the server is not using a firewall). There must be something I can change to make sshd more permissive? I've tried before to get my ISP to fix their problem but no luck. This needs to be sorted as a server at my home (which does not run SuSE) retrieves backups from the remote OpenSuSE 11.3 server every night using scp and these are now failing.
I have Ubuntu on one machine and open SUSE on the other. Both connected with an ethernet cable to the same router. Both connect to the internet ok. I did manage to set up 2 machines both running Ubuntu with file sharing but now I have switched one to SUSE I seem to be getting into a mess. Ubuntu sees the SUSE machine but does not open any folders. SUS does NOT see the Ubutu machine. I have looked for a beginners guide to setting up file shaing with SUSE without success. (I like SUSE better because I can use my Audiophile sound card "out of the box")
My computer shares an internet connection using an ADSL router.There are other three machines.I have set up a Apache server for learning purpose and I want it to be inaccessible from anywhere else including the PCs in the network.When I enter my ip-address assigned in the network (192.168.1.1xx) from other computer,I get my ppages and I dont want that.
How can I block HTTP requests from other computers?
i'm trying to automount a samba share on computers startup, everything works just fine when the network adapter is controled by traditional way ifup but when i'm switching to NetworkManager doesn't work anyway. i'm hiding the share information from the common user by using the /etc/samba/cifstab as i read on intrernet, in the cifstab i put //server/share /home/user/share cifs sername=user,password=pass,uid=1000,_netdev .could anybody give me a hint or tell me where to look?
I set up a server that has dhcp going. This works 100% as I get ipaddresses. My question is how do these internal computers connect to the internet? There is a proxy that needs to be passed as well...
Is this just a simple case of portforwarding or is this a completely different thing?
Should I be able to transfer files between computers wired to the same router? Is the router bridging function used to transfer data between wired computers? I just got DSL and I'm new to networking. I have openSuse 11.4 on two computers plugged into a Belkin router. The internet connection to my ISP works from either computer. If I ping $HOSTNAME from either computer, the hostnames are different, but the IP address is the same which is that of my ISP. I only have the one router, but it has a bridging option. I don't know if it can do both bridging and routing simultaneously.
I had a google of this but can't find anything useful. I use networkmanager to configure my wireless card. Currently this only works when I'm logged in to KDE. If I log out the system loses the network connection. Is there a way to make it persistent using NetworkManager?
My goal is to have users(Unix Users like root on host1) use evolution /sendmail to send and receive emails between users( Unix users either from host1 or any other host on n a private network. The network is a private network with a few servers without any connection to Internet and no DNS. They resolve one another by /etc/hosts. with smtp configured am able to send and receive emails on one machine(host1) from unix user abc on host1 to user xyz on host1 and vice versa. the commad i used to test is host1@abc #echo "this is a test mail" |mailx zyx@host1 But when i tried sending from user abc on host1 to user xyz on host2 its giving error for DNS unable to resolve hostname.
the commad i used to test is host1@abc# echo "this is a test mail" |mailx zyx@host2 create a simple sendmail config so that i can send mails between multiple hosts using unix usernames within one network. I can use one of the server as mail server and rest as clients. I dont want to configure any other softwares for pop or imap access. I want to keep it simple and use evolution on the clients to send emails between one another on the private network without internet.
Well to make this straight forward as possible I want a program like The Hotspot Shield In Windows and Mac, Its uses's VPN (Virtual Private Network). To view Hulu (And other stuff). A link: [URL]. Or can I use it wine (I don't really think so, by the way I didn't test it out Because its blocked in my country).
We have to connect one PC in private network (campus) with other PC (mostly a modem in our case)in public network. Connection should be peer to peer like and we have to use C coding for establishing connection between this two systems. Is it possible if we use port forwarding or is there any other way?
1. I have at work a regular LAN with many PCs, each with a DNS-registered public IP. Therefore I am able to address each of these PCs by their fully-qualified names and, for instance, initiate ssh sessions to any of these computers just by typing "ssh <name_of_machine>" from a terminal.
2. Within the aforementioned LAN I have just created a private network with some clients, which access the LAN through a router (a D-link DIR-825). We have created this private network for many reasons, but most importantly because we need to guarantee that the hosts in this network will remain networked among them even if the LAN goes down for any reason (which unfortunately happens often). But we still need to have access to the hosts in the private network from the LAN.
3. I am able to define port forwarding rules in the router in order to access certain services on the private network's clients. For example. I am able to access (by ssh) hosts "H1" and "H2" on the private network from a client on the LAN by defining rules for forwarding ports "P1" and "P2" on the router's public IP to TCP port 22 on the private IPs of "H1" and "H2", respectively. Then I would access each of these hosts from the LAN by using:
4. The problem with the port forwarding approach is that it is not easily scalable. For instance, If I wanted to enable ssh access to each host in the private network, I would have to define a port forwarding rule for each machine, and then REMEMBER all these port rules when initiating a ssh session from the LAN in order to point to the right host. And the problem gets worse when considering more services in addition to ssh.
5. The ideal solution would be to be have a means for addressing each host in the private network individually, in much the same way in which I address the hosts in the LAN (which have DNS-registered names). For instance, in order to access hosts H1 and H2 as in the previous example, i would like to be able to just type
I guess I can say that what I need is some kind of combined DNS-ing and routing that allows me to communicate with the hosts in the private network from outside of it in a transparent way.
The question is: what are any possible solutions for accomplishing this? I have searched the web and found stuff about things like VPNs, reverse-proxies and NAT servers, but I really can't understand if any of these could serve to solve my problem (BTW, isn't my router doing some sort of NAT-ing already? could I just add some DNS-ing in some way?)
I run a server that is connected to several other boxes in a private network (192.168.0.0). I have had no problems previously, but upon a reinstall of Debian squeeze I have no connectivity to the private network: a ping of other addresses on that network fail. Ping and connection to outside world is fine. This box is configured similarly to another on this private network which connects successfully. I have quadruple-checked all my basic information. I post my ifconfig below of the malfunctioning device, then the ifconfig for equivalent nic on a box connected succesfully to the private network.
Questions:
1.Could it be a cable problem? (I don't see how since the cables have not been changed I do not believe since my reinstallation).
2. What about the difference in the last lines of eth1 below and eth2 of the successful box. Is it significant that the bad eth1 reads Memory:fc3a0000-fc3c0000
I have 2 different networks: the first one is gateway machine (eth0), and the second is a private machine (eth1). So, I've configured the iptables and forwarding stuff and when I try to ping [URL]... on the gateway machine, it works, while it doesn't work on the private network.
I have users using Windows XP, Windows 7, Linux (Fedora) and Mac. They all are in a single private network and all access internet through a Linux (RHEL5) system in which Squid acts as gateway. The same is true with my branch offices too except that private network is different and gateway system uses Fedora 9 instead of RHEL5. All the branch offices are connected through point to point leased lines with the head office for file transfer.
My requirement is this: I have a web server located at head office. Presently I am able to access this server from my branch offices through internet. I would like to access this server from branch offices through leased lines. This too I am able to access if I do routing in users system. The file transfer is taking place through one to one system at two ends by creating static routing in those systems.
Suppose that I am in a private network of computers (say each having ip addresses 192.168.. ). Some of the machines( we have no information regarding their ip, name and no physical access to the servers) in the network are connected to internet and they run an http proxy at some port say 3128. Is there a program which can be run on Windows or Linux which will give me the list of machines(ip addresses and ports if possible) acting as proxy servers?
I have 2 different networks: the first one is gateway machine (eth0), and the second is a private machine (eth1). So, I've configured the iptables and forwarding stuff and when I try to ping google.com on the gateway machine, it works, while it doesn't work on the private network. Note: I am using VmWare 7. I need your quick assistance about this issue.
i dont know much about setting up networks at all. i searched through this site and others, but have not yet been successful in finding the information i need.i have a laptop running 11.2, kde 4.4. this laptop is connected to a wireless router. now, i also have a desktop running winXP, without wireless, which i would like to connect to the network. where the desktop is located, i can't connect to the router via ethernet, but i would be able to connect it to the laptop via ethernet.so, is there a way to connect this XP desktop to my op.su. laptop so that it will have net access?if so, could someone please help me, or direct me to the appropriate information source?
i was wondering if anyone knew how to make a Grub network boot cd. I would like to run openSUSE off of my server and not on my hard drive because i don't have a lot of space for the linux partitions.
I set up a squid transparent proxy and I have a problem with an iptable rules. I have a rule to redirect all request to port 80 to go on port 3128. To do so, I'm using this iptables command :
This command is working like a charm. The only problem is, for some unknown reasons, this rule will be dropped at some point. I did not manage to identify what is causing this to happen. It occurs during night, but I have nothing about that in my log files. messages / firewall / ...) The only way I managed to reproduce this 'dropping' is this one: I type the command like as root. The command is effective and working fine. I open yast, I go to the firewall module, the I do a simple "save changes and restart firewall" (without changing anything). As soon as this process is finished, the iptables rule is gone.
-How can I make this rule permanent ?
-Is there a place where I can launch a script executing this rule, after the yast firewall module is 'touched' or something ?
I want to enable Pretty URLs in Concrete5 and Joomla CMSs. So far I have- enabled pretty urls in the CMS- created .htaccess files (as they showed)- changed AllowOverride None to All in /etc/Apache2/httpd.conf <directory>- restarted ApacheBut the pretty urls still don't work, When I click any link on either site it gives me a page not found error. If I use the "non-pretty" URLs, all the links work. So I guess Apache either doesn't read the httpd config file or it doesn't read the .htaccess files or something else. Has anyone got this working? What else did you do?
