[quote=“pmonstad, post: 24126, member: 17422”]Read somewhere in the forum about replacing ftp/NFS with http. Is this something which is going to be realized in an upcoming release?[/quote]

Please don’t hijack threads, yes http transfer is slated for 0.33b and its being implemented with iPXE

Did you already tried the kernel from Tom?
Here is a post from him with a explanation how to install the kernel.
[url]http://fogproject.org/forum/threads/cant-pxe-boot-to-fog-server.10107/page-2#post-23967[/url]

Tony Byrne,

There is a forum resource that has all the pertinent information needed to changing your compression level already. I will post to that here.

You will find, however, that the upload time will be dramatically increased with higher compression levels. However, the deploy times should actually decrease. This is because there is less data being sent to the system. So, while uploading images will take a long time, deploying should actually be faster.

Ultimately, you state that you don’t care about the speed.

What version of FOG are you using?

In 0.33b (while it is only in beta mode it’s still pretty good) has options to adjust the compression level from the FOG Web GUI under FOG Configuration->FOG Settings->FOG_PIGZ_COMP.

In 0.32 and below, you will have to edit the init.gz to change the compression level.

-0 is the worst compression (most disk space used on server) while -9 is the best compression (less disk space used on the server).

In 0.32 and below, it defaults to -3.

To edit the compression level open a terminal on your fog server as root.

Perform these commands:
[code]cd /tftpboot/fog/images
mkdir tmp
gunzip init.gz
mount -o loop init tmp
sed -i ‘s:PIGZ_COMP=“-3”:PIGZ_COMP=“-9”:’ tmp/bin/fog
umount tmp
gzip -9 init
rmdir tmp[/code]

The sed -i line is where it’s changing your compression. It looks for the instance of PIGZ_COMP=“-3” and changes it to PIGZ_COMP=“-9” so if you want it less, make the edit to the second PIGZ_COMP phrase and make it less.

Step by step of the commands are:
change to the images directory
make a directory within the images directory
decompress the init.gz file.
mount the extracted file to the tmp folder you just created.
edit the relevant part of the proper file for compression.
umount the tmp folder
compress the init.gz file.
remove the tmp folder.

Hope this helps.

I may remove the need for the \ all together. I mean, you’ve already got the domain in place, so why not use that, then just have a setup like:

domain: test
ou:
domain username: user
domain password:ao723597u32l4kjfldgpjdas;fjkasupiou2j5kasjfk;

Then when the call to send it comes along send it as:
$Host->get(‘ADDomain’).‘’.$Host->get(‘ADUser’)

What OS are you on?

In ubuntu you use updaterc.d to disable/enable startup stuff.

apache, tftp-hpa, vsftpd, nfs-kernel-server, nfs-common, portmap/rpcbind, FOGImageReplicator, FOGMulticastManager, FOGScheduler, mysql

In redhat types you use chkconfig <servicename> off

httpd, xinetd, vsftpd, nfs, rpcbind, mysqld, FOGImageReplicator, FOGMulticastManager, FOGScheduler

This is a known issue with the 12.04 Ubuntu derivative, the services are started before an ip address is assigned to the machine.

There are a number of posts ON THIS FORUM that explain how to rectify the issue and delay the start up of the job so that you don’t have to start them manually after each reboot.

Anyone actually fix this “Unable to find a valid task ID based on the clients mac address of” as I have tried the “@” fix to the file at line 133 and still have a loop on a machine with a previously entered MAC address. Help!

Ok,

I have installed tftp following the instructions above.

When I go to theTask Management tab on my FOG server and click UPLOAD under the host I want to use I get the following error.

“Unable to connect to tftp server”

My apologies

I use start up scripts to active my Windows and my services. I set up my programs to install silently and I use a file to answer the questions. This could include your key to your program. I don’t know that this is the proper way to handle the situation, but I get pretty inventive with what I know how to do.

You need to be good at scripting before you attempt this.

Your findings, Lee, is about what I’d expect. This, of course, if the deploy is done (well maybe not now) from a unicast stand point rather than a Multicast.

Compression is a fickle thing. Uploading anything and compressing before doing so always take far more time than re-obtaining the data even if the re-obtaining is highly compressed. This is because the CPU is actually doing much more work finding what can and how much it can be compressed.

On a download, the less data you have to transfer, the faster it transfers.

Wouldn’t say it was large but have 600+ PC/laptop network, as always questions are welcome. What may be a good start is if you said more about your network/setup and plans

Fog can hold all the inventory as well. Makes thing very easy down the road.

If you know the MAC’s and Hostnames, you could create a CSV that contains the MAC and Hostname. I forget all the bit’s and bobs, generated from your current inventory to make importing them that much easier.

It doesn’t cost anything, I just have to figure out where the bug is and squash it.

Thanks! I’ll start working on that, and I’ll let you know if it works or if I run into any other problems.

Thanks again!

The VOIP phones do run on the same subnet as the network. I just made it an isolated network.

Thanks!

This sounds like a job for the Location Patch created by Lee Rowlett, Maybe he can chime in here!