Posts made by chad-bisd

Have you checked the wiki for working or problematic devices?

You can setup your Fog server as DHCP and PXEBoot. The IP address given during PXE boot will only be used for that function, and when the regular operating system loads, it will switch to it’s configured static IP address.

Just setup your DHCP scope to be outside the normal IP range used by your static network. I’m assuming you have a pretty simple networking, no VLANS, just a few switches, most likely unmanaged.

If you have a more complex network, it can still be done, just more things to consider when configured it all.

did you check to make sure the ftp service is running? You can login to the FTP service using the information at:
[CODE]define( “STORAGE_FTP_USERNAME”, “fog” );
define( “STORAGE_FTP_PASSWORD”, “-somethingencrypted-” );[/CODE]

Also, there are some minor issues with Fog 0.32 and Ubuntu Server 12.04 LTS. Just minor kinks being worked out. I continue to suggest those new to fog stick with Fog 0.32 on Ubuntu Server 10.04 LTS. There is better support within the Fog forums due to its age and the number of folks currently running that combo of OS/Fog.

I think if the base install on the real hardware has all of the drivers loaded and available, you can deploy that image to a virtual machine and sysprep it to remove all the hardware specific information, but leave the drivers still available for when you deploy it back to real hardware.

This would be my experiment:
[LIST=1]
[]Deploy Fog image to virtual machine
[]Clone virtual machine
[]Update clone (software, windows updates, etc)
[]run sysprep /generalize to remove hardware specific info
[]Upload to [B]new[/B] image definition within Fog
[]Deploy new image to real hardware
[*]See if the devices are recognized and the drivers found and auto installed
[/LIST]
Not having done this yet, I can’t tell you how it would work. If you do the experiment, let us know your results. At some point, I’ll be doing this to update images instead of hauling machines out of the storage closet I hope.

The intarweb ate my reply earlier, so I hope it enjoys having indigestion.

For mainly unicasting, set the storage node (dl380) up to have the majority of the clients, since it has a faster disk subsystem (I’m assuming it’s got 4-6 drives in raid 5). Go into the storage node settings and change the defaultMember (should be your HP machine) to have a queue size of say 1, and change the other storage node (the DL380) to have a queue size of 10.

For mainly multicasting, the computer with the most RAM would probably be the best to use for the source of your images.

It sounds like PXE boot is working, because you are seeing the Fog menu, correct?

What is failing is when control is transferred to the Fog kernel which tries to load and start processing whichever Fog task you have called, such as registering, imaging, etc…

Have you tried different Fog kernels through the Kernel Updates option in the Fog server Web UI? Sometimes newer kernels work, like those based off 3.2.x or 3.3.x, but sometimes things work better going back to older kernels, such as 2.6.37 KS. Try a few and see if you get better results. If not, you can also work on compiling your own kernel and the forums are here to help either way.

[code]sudo service tftpd-hpa restart[/code]

or you can [code]sudo service --status-all | grep tftp[/code] to find the service name

search the forums for recent posts on tftpd-hpa, someone has posted a way to make the script sleep/wait on the LAN interfaces before it tries to start tftpd-hpa.

Do you get the main Fog Menu where you can try to register the host and then fail, or does it fail to even display the fog menu?

You can also just push the arrow up or down key and it will stop the timer and give you all day long to read the menu. I set mine short so the end users are not slowed down too much when booting, and to keep the students from trying to mess with the fog menu.

common issue. something to do with tftpd-hpa trying to start before the interface has an IP address. Workaround for now is to stop and restart the tftpd-hpa service after boot up.

If you want something that works out of the box for now, downgrade to Ubuntu 10.04 LTS Server for your Fog server. I’m assuming there is work being done to make it so tftpd-hpa service will wait until the interfaces have IP addresses before it starts, but I cannot be sure as to when that will be fixed.

unless you are just tied to fedora for some reason, install ubuntu 10.04 Server LTS and you can follow the instructions in the wiki article. I do not use Fedora currently, and I have no idea how to customize dnsmasq to do what it needs to do without the Ubuntu LTSP stuff.

InstallRite has no support system or forums?

Have you considered a Fog snap-in to run post imaging that might execute whatever commands you need to enable the 2nd monitors?

Go to the sourceforge page for fog and go through the downloads. There should be kernels listed there. Start testing out the most recent kernel and move your way backwards through them. Without access to their configs, I can’t tell which supports what hardware other than trial and error.

Please post your support requests in the Technical Help forum. This area is reserved for developers of Fog looking for help.

For a storage node to work, it needs two things. A valid connection to the main mysql database so it can check it’s status (and maybe other things). This should be setup when you choose the “S” install type and complete the setup wizard. In the wizard, it asks for the username and password for the main mysql database, usually username is fogstorage and password is something like “fsXXXX” where XXXX is between 0000 and 9999 randomly.

You only have to update your mysql configuration if the mysql engine does not allow connections from remote hosts, such as on ubuntu 10.04 LTS.

The problem you seem to be having is the ftp connection to from the MAIN fog server to the Storage Node. The main fog server PUSHES the images to the storage nodes, so it has to know a valid username and password for each different storage node for the file transfer to work. The storage node determines the password, but the username is always “fog”.

Example: On your [B]storage node[/B] the /opt/fog/.fogsettings file has a password of “1234abcd”.

When you login to the Web UI of Fog on the [B]Main[/B] Fog server and define the [B]new storage node[/B](not defaultMember, but a new one), the Management Username is “fog” and the Management Password is “1234abcd”

Do not change the Management username or password for the defaultMember storage node.

Once you have the correct Management Username and Management Password in the storage node definition, you can restart the FOGImageReplicator service and watch the log file to see if you continue to get an error.

Yes, it should work without issue IF you have your network configured properly. Fog works just fine for multiple subnets, even with ProxyDHCP (with minor modifications to base config).

The thing that may trip you up is making sure your layer 3 devices forward dhcp broadcasts from the subnet of the client to the subnet of the DHCP server. Which, should already be setup if you can get a DHCP address on a client in a different subnet.

I didn’t look through the code enough to figure out if it already tries to update the /opt/fog/service/etc/config.php file with the mysql root password. I kind of assumed that the installer doesn’t know what the password is, at least on Ubuntu because the installer prompts you for it when you install the mysql packages.

I had to change an existing server to have a password on the mysql root user account, and had to scour the wiki, forums, and google to find both locations it’s needed.

The installer currently warns you that if you set the mysql password for the root user to a non-empty value, that you must update the /var/www/fog/commons/config.php file, when in reality, you have to modify the /opt/fog/service/etc/config.php ALSO.

I want to make Fog easier to use in general and adding an additional note to the install script should be easy before 0.33 is released.

If at some point in the future the installer can update the config files for you, we’ll be even closer to perfection.

Silly question: Do you have the right IP address for your storage node “defaultMember” in the StorageNodes definition in the Web UI? I think this is stored in the mysql database and may not be pointing to the right server since you are recovering onto a new installation.