Networking :: Bonded NICs At Host OS - Ubuntu Server - Bridged To Virtual OS - UServ - In VMware Server
Jul 19, 2009
Just something that struck me while working on our virtual servers today.
I have bonded 3 NICs at the host in Ubuntu Server 8.04 LTS. They are using mode 0 for Round-robin. Point is to increase the speed/performance of all the servers, but mainly the fileserver. The fileserver is a virtual server running Ubuntu Server 8.04 LTS on VMware Server 2.0.
1) I noticed the NIC in the slave OS reported link speed as 1000 and Im unable to change it as the NIC (virtual one) doesnt support it. Does this not really matter, as the NIC doesnt exist, and it will run at higher speeds anyway? Or do I have to remove the bond on the host, bridge all 3 interfaces from the host to the slave OS, and then make a bond in the slave OS?
2) While at it, does mode 0 only increase performance on data being sent from the host or does it also increase the available incoming bandwidth?
I have an Ubuntu 10.10 host running in VirtualBox 4.0.8 r71778, and I need to setup a LDAP authentication network between two VM's: a Maverick-based distro and Ubuntu Server 11.04.
Screenshot of the two running perfectly fine:
[URL]
Using NAT, I can have the two VM's up and running on VBox, but they get the same IP (10.0.2.15).So I googled it and quickly found out that I needed to change the VM's network interfaces to Bridged Networking. So I tried that. Here's two screenshots of the default configurations:
1- Ubuntu Maverick based distro:
[URL]
2- And here is one of the Ubuntu Server:
[URL]
(I'm Portuguese, that's why there are mixed languages in the UI)It's pretty stupid to explain like this, but whenever I try to start one of those two using Bridged Networking, I get a black screen full of errors (I can't take a printscreen of that, so I took a bunch of pictures - yeah, it's lame, I know ). Then, I have to press the power button to shut down my host, and restart the pc.Here they are:
[URL]
Where I think the problem is:
[URL]
My host is currently running Linux kernel 2.6.38-10.I have dkms, linux-header-generic and build-essentials installed (and the VB extension package too)And my host is a laptop, connecting to the internet using the router my ISP provided me.How can I solve this? Is this some kind of bug regarding 2.6.38-10 kernel?I don't have other pc's to test my LDAP authentication, so I needed the two VM's to have different IP's.
I'm having trouble getting Fedora 15 to work with my em/bond/br configuration as it did in Fedora 14. I've got a bonding interface over em1 and em2, and then individual bond vlans (bond0.2, bond0.10, etc...), I then have a similarly named bridge interface (br0.2, br0.10, etc...) for kvm. Regardless of if I assign bond0.2 or br0.2 the IP address, I am unable to ping the gateway, but the native untagged VLAN (bond0, br0) has no issue at all.
To try to isolate this issue down to the switch or Fedora 15, I went with a plain em setup and configured em1, and em1.2 which works as expected. This issue has been plaguing me since Fedora 15 came out and I'm about to trash it and go back to Fedora 14 (Which didn't have this issue), but some of the new kvm features have kept me trying to get over this incomprehensible hump.
My brother has a Ubuntu server attached to a LAN with Windows PCs. I set up an openVPN tun service on the server, let's call this VPN1, so that I can connect remotely from my Ubuntu desktop. The server has one NIC and the LAN has a router that is the gateway to the internet.
My brother would like to remotely access his windows network when roaming with his Windows laptop.
Therefore, I would like to set up a second VPN service, this time tap, that is bridged with his LAN. Let's call this VPN2.
The LAN subnet is 192.168.1.0/24. The internet gateway is 192.168.1.1
The NIC has a fixed IP set by the router of 192.168.1.150
How do I set up VPN2 so as not to trash VPN1? That is, if I bridge eth0 with tap0 how will that impact VPN1's tun0 that is using eth0? Do I need a separate NIC for the VPN2 bridge?
I completed the installation of VMWare Server 2.0.2 onto a CentOS 5.4 64-bit distribution. There is a VMware virtual machine file on the server in question, and I want to activate it through the application console.
I did not see anything readily apparent that would facilitate the importing of the virtual machine file. Does anyone have a procedure that can accomplish this task?
I have a WindowsXP virtual machine which I need to isolate from the host machine completely (have the host act as a bridge but not be visible on the IP layer at all.) It still needs to have Internet access. Obviously it has to be able to contact the router but I would like to be able to block port 80 (or even just block all SYN packets addressed to the router.) I also want to allow port forwarding from the router to the virtual machine. I can use basic iptables but this is way beyond me.
The host OS is probably going to be Debian Lenny but this is not built yet so if someone can recommend a different distro which is as lightweight as possible but will support VMWare, iptables and tcpdump then that would be great. I was thinking of Slackware but I have not used it in ages and from what I can remember their is no real package management.
In the past I found some great help on this forum, so here goes. Bare with me because it's a long story. I'll try to be as complete as possible. I've installed and configured OpenLdap on a virtual machine with ip 192.168.39.134. I've added 2 users via LAM. In the ou WikiUsers and the domain is wiki.local.
I've then created another host with ip 192.168.39.133 with mediawiki installed on it. Then I added the extension LDAPAuthenthication. In the LdapAuthentication file I added this code (only the last paragraph is mine, I added the others to show it's location in the script):
I know I'm close because I can't register any new users or accounts on the mediawiki site. Although I could before I added the LDAP service. This is indeed all just to test and get to know how LDAP works. That's why it's all virtual in VMWare. I did not really configure anything on the LDAP, i just installed it and chose a domain (wiki.local).
I installed apache server on my Fedora 12. Its showing the test page 127.0.0.1, but when I give the address 97.168.234.76 (0r www.chaalu-kam.com), it tells that the requested url could not be retrieved.
I have made the following additions to the /etc/httpd/conf/httpd.conf file
Code:
In addition to this, I have added this to my /etc/hosts file
Code:
I wanted to test the Apache server on my laptop only by putting an html file in /var/www/html and naming it as index.html. I want this index.html page to be displayed when I put www.chaalu-kam.com in my browser. What should I do? My firewall is off too
Bit of an odd one, this. I've migrated a website from my old server to a new machine. Both servers run Ubuntu + Apache2. Both only serve a single site, apart from the default site.I've flipped the domain name to the new IP address.The trouble is that after moving the virtual host config over into sites-available, with the necessary link in sites-enabled, Apache attempts to serve from the default web root (/var/www) rather than the actual site content (in /var/www/technology). So for example, an attempt to browse.
apache virtual host to limit the concurrent connections of virtual hosts? Taking into account the host of each virtual user's home directory can also have more than one subdirectory, which should be restricted to a subdirectory. Is beyond the control of the operation of these sites in a subdirectory. Best local restrictions or limitations to the overall situation.
I'm having an issue with setting up the virtual hosts on my web server. I have 2 virtual hosts (example1.com, example2.com). example1.com works but example2.com is sent to the index file of example1.com. I did some searching on google and it seems the problem might be with my /etc/hosts file.
First virtual host that the second is also directed to...in sites-available/sites-enabled (note port 80 is blocked by my isp so I use 8080)
Code:
Second virtual host file
Code:
And my hosts file
Code:
# The following lines are desirable for IPv6 capable hosts
Also I'm using a dyndns.org...would that make a difference?
Im using fedora 13 x64 install from DVD. Has anyone else found problems using vmware 7.0.1 or 7.1. My bridged networking will not pickup a dhcp address, and even if I configure static my network will still not work, yet NAT works fine. Even configuring bridged to use a dedicated device doesnt work. Workstation worked fine on Fedora 12 but this seems to be a reoccuring problem in every other release.
I have Linux server with VMware Server 1.0.8. When I creating a virtual machine, I can not "send" a PING to the virtual machine. In the virtual machine, I have installed CentOS 5.3 (32bit). In the virtual machine I have defined the addresses IP. So, whey I can't "sent" a ping to the virtual machie? I have to set something on the Linux server?
I currently have one of our clients set up to use a routed VPN for their 5 laptops to connect to the server remotley. And this works brilliantly. They are about to bring on a remote office that will need a VPN connection back to the main office, so I was going to set up a bridged connection between the two sites (and possibly more sites in the future).
So my question is whats the best way to go about this? Can I have one instance of OpenVPN running with tun0 set up for a routed connection to the laptops and add a second tun (tun1) to the config that will be for the bridged connection between the sites? Or am I going to have to run multiple instances of OpenVNP, one for the routed and another for the bridged?
If routed and bridged have to run in seperate instances, will I have to add another instance for each new remote site that needs a connection? Can a bridged config connect to multiple sites, or have multiple tuns in the one config?
Not sure if this post belongs here, but here it is.
Host: Win server 2003 Guest: Ubuntu server
Host: IS able to ping guest. Firewall is OFF. NOT able to access guest (which is a web-server) at browser. Guest: NOT able to ping host. Running a web server, you can check the website: (pegajosa.com) is running under that virtual Linux server.
Problem: guest needs to access host's sql database and/or any resources.
I've set up bridged networking so that I can have KVM virtual machines that are accessible from outside the host. I can access both the Host and my VM from other machines on the local network, and from the VM I can access the internet but from the Host I can only access my local network. Since I can access the local network and the same issue applies regardless of whether I use host names or IP addresses I suspect it's not picking up the gateway properly.
How do I go about allowing the VM host to access the internet while still having bridged networking so I can access my VMs?
What I've checked so far: Ping Google DNS (8.8.8.8 ) From Host: Destination Host Unreachable From VM: Suceeds
[Code].....
The issues are only with accessing things outside of my subnet. I can access other machines on the same subnet from the Host just fine.
I just had one hell of an experience trying to get openSUSE 11.2 to run on VMware server 2.0 on a Windows XP host. I am posting some instructions on how I finally got things working.
1) Download the openSuse iso image (and add-on CD if needed)
2) Download and install the VMWare server
3) Before you try to start the server console, you will need to configure the web access to allow you to log on. go to press Start->Programs->VMware->VMWare web access->Configure Tomcat
4) on the logon tab check logon as: option
5) add your account there and set the password
6) Now in your web browser (I use firefox) access the console via: [url] [url]
7) Before you start to create a virtual machine, you will want to set up the server to have access to your iso's. The server will only allow VM's access to specific directories through "datastores". You can either move the iso images to what has been specified as the default datastore... or create another datastore that points to the where you have the iso's.
8) Now we can create our virtual machine. This is simple, just press the button and follow the prompts.
a) at some point it will ask about an CD-Rom, you will want to use your suse DVD iso image. Select the option and then navigate the datastore that contains the image. Once found click and continue.
b) Network interface has 3 choices:
Bridged - the virtual machine will appear on the network as any other computer. Allows the VM to host server applications. User needs to have control over the network to ensure DHCP requests are handled (or static IP is assigned).
NAT - VMWare server will manage a virtual router, the VM is then connected to this router. On this router, the hosts internet is used as the WAN. Users do not have to have control of the network, but will have to setup port forwarding in VMWare to host server applications
Host Only - This means that the virtual machine can only talk to the host. The only way to get networking is to have a proxy through the host machine. This is setup on the host independent of VMWare. Users do not need control of the network, but control over the host. I doubt you could host server applications though.
9) With the VM created we can start it up. We can open a window to it through the console tab of the VMWare server console. When I first started it, it was stuck looking for an OS once the console finally rendered the graphics. A reset, an F2 at the post and selecting the appropriate boot media for boot got the installation started.
10) Install Suse as you would any other suse installation. I added a separate partition for swap, but kept the rest on one big one. Others may still want to maintain the separation of home and root
11) ---- Important Error/Solution ----
Once the installation has completed many will notice that the mouse or keyboard does not work! This was the source of my biggest head-ache. Some may think it was a problem with VMWare tools... I tried that to fix that and found the tools were working properly.
I just installed (n this 11.2 system) vmware server (v 2.02), to have access to a few small systems. One of them is a 11.1 guest which I just upgraded to 11.3, successfully (almost).
I have a problem, though: from the host I can not ping/ssh the guest.
Guest to host works fine (including names).
Code:
The firewall is down on both sides. I don't see anything with iptraf in the guest. The IP addres of the guest is correct, unless I'm too tired to see.
Since a few weeks I cannot open a console in vmware server 2.0.2 anymore. I don't know exactly when this started because I use vmware only occasionally (I have only 1 program (an old Ashlar DrawingBoard) which does not run properly in wine). VMware server and the console have been working before without problems, but now I get the message "Cannot connect virtual machine console." and it times out. This is on Opensuse 11.1 with VMware Server on the same pc, Firefox and the guest is Windows2000.
I have tried to disable all extensions in firefox 3.6, disabled the firewall, even created a new user (on the same pc) with a clean firefox, all without any result. However when I connect from a different computer which runs Internet Explorer under windowsXP, then I can get the console without problem.
So, somewhere there is something wrong between Firefox and VMware server.
I'm running CentOS 5.5 x64 and have just installed VMWare Server 1. I am able to create a new Virtual Machine, but when I go to power on the VM in order to install the OS (windows server 2003), either the host crashes/reboots (with "Run in Debugging Mode" turned on) or the VM simply does not start (with that Debugging box unchecked) and in the log there is an error along the lines of VMware Server unrecoverable error (vmx)
I have 4 domains registered through godaddy. I have a dell poweredge box with one static IP hosted somewhere. I want the 4 domain to resolve to four different sites. I have already created 4 different sites in apache with four different aliases and has enabled it.
eg alias /a /var/www/a alias /b /var/www/b alias /c /var/www/c
I tried with godaddys forward with masking option. It works but I can't hide xx.xx.xx.xx/a in the links. Whats the best approach? One limitation is I cannot use the webhosts name servers. How do I set this up with name based virtual hosting? Do I setup a DNS server in the box?
I do have a dns host name for my box which is publicy accessible. Some my.ca.examplehost.net
I'm using Apache 2.2 to host multiple subdomains using a single SSL certificate (a wildcard certificate e.g. *.mydomain.com) and, yes, it works! Everything seems to be served correctly and the browsers are pretty happy.
And you can also have the non-SSL sites (virtual hosts on port 80) on the same IP. (That's covered elsewhere)
For those that want similar functionality here's my discovery...
My configuration is like this:
ssl.conf:
Code:
I would have to say that I don't believe that this will work if you are not using a wildcard SSL certificate and having anything other than subdomains under that wildcard.
It is somewhat limited in scenario where this is useful, but for a set of company websites that should be under SSL, this can be tremedously useful when you have a single IP.
I'm running XAMPP 1.7.2 on Ubuntu 8.10 (Linux dt19.im.local 2.6.27-14-generic #1 SMP Tue Aug 18 16:25:45 UTC 2009 i686 GNU/Linux) and am using the PHP 5.3.0 Apache module as standard. For one virtual host I'd like to use PHP 5.2.X as it is part of a project which has a lot of legacy code which is not compatible with PHP 5.3.0. The virtual host configuration block and the applicable directory directive are as follows -
Code:
Checking phpinfo() output on the above virtual host (or using the default virtual host directive and accessing it via http://localhost/[SNIP]/[SNIP]/phpinfo.php rather than [url] shows PHP 5.3.0 is running. After applying minor tweaks such as adding ScriptAlias or SetEnv options the problem persists. I've Googled for a good while and have checked the permissions and the like and tried the advice of other users (XAMPP or otherwise) either resulting in PHP 5.3.0 being used or a HTTP 400 bad request/invalid URI error. I've stuck with the configuration above as this is correct according to the PHP manual.
FYI cgi-bin/php-5.2.6 is a soft symbolic link to /opt/lampp/bin/php-5.2.6 (I've added the FollowSymLinks option to the cgi-bin directory directive in httpd.conf). I've tried installing php5-cgi from the Ubuntu repos and setting it up in a similar way, to no avail. I've also tried copying the executables into the cgi-bin directory, pointing the Action line directly to bin/php-5.2.6 and dropping the -c /opt/lampp/etc/php.ini-pre1.7.2 option in the Action line. I've even tried commenting out the LoadModule lines for PHP which results in a HTTP 400 bad request/invalid URI error. This demonstrates the fact that the PHP CGI use is being ignored.
I've checked httpd.conf and the extra/httpd-*.conf files and ensured all required includes are being loaded. I know that it's probably something stupid on my part which is causing this! Given that I've tried PHP CGI builds in the Ubuntu repos I don't think this is an XAMPP-specific issue.
One of our Apache servers, version 2.2.3, is running three virtual hosts on it. However, while two are accessible, gray.mgh.harvard.edu, www.plastimatch.org, the third, cmrol.mgh.harvard.edu cannot be reached by any remote host.
I am at a loss as to why this is occurring. To that end, I am including the httpd.conf file as an attachment in the hopes someone can see what I missed, and help correct this issue.
I'm trying to figure out why Apache is ignoring SSL Cipher statements when placed inside a Virtual Host. Specifically I'm trying to disable SSLv2 and only allow SSLv3 or TLSv1
In httpd-ssl.conf I have the following
Code:
And when I try and connect, I get the expected results.
Quote:
In my virtual host statement I have
Code:
The only difference between the two is the httpd-ssl.conf has -ALL in the CIPHER, and the virtual host entry has ALL. However if I try and change the ALL statement in the virtual host entry to -ALL I get the following error in my logs & get no content.
Quote:
Why it's not working in the virtual host statement? Unfortunately as many of you are probably aware it's impossible to gain PCI certification with SSLv2 enabled.
i have linux server application based on centos 5 that run httpd daemon from /usr/local/pf/conf/httpd.conf which is generated from the /usr/local/pf/conf/templates/httpd.conf.apache22 file
the problem is every time i run my linux server application i am not able to access phpmyadmin unless i stop my linux server application which is stop it httpd daemon then start "service httpd start"
i have been told by forum moderator to create an alias for my phpmyadmin in Virtual Host but it stil dont work, i am not able access my phpmyadmin site
this is what i have done to solve this problem - open /usr/local/pf/conf/templates/httpd.conf.apache22 file then look for
<VirtualHost *:%%admin_port%%> - add Alias /phpmyadmin "/usr/share/phpmyadmin" (Failed) - add Alias /phpmyadmin /usr/share/phpmyadmin (Failed) - add Alias /phpmyadmin "/usr/share/phpmyadmin/" (Failed)
[Code]....
why i still cant open my phpmyadmin site ? or why i cant run phpmyadmin together with my linux server application because it use the same daemon (httpd daemon)
