Posts made by george1421

This one may actually work a bit better since it should connect you right to the FOG menu

set fogserver:ipv4 <ip_addr_fog_server>
set next-server ${fogserver}
chain http://${fogserver}/fog/service/ipxe/boot.php?mac=${net0/mac}

Funny I was just discussing this the other day with a colleague. I haven’t had a chance to put together a POC yet on this. But the general idea is to create the ipxe boot image on the usb device and then chain to the fog server using the command like Sebastian posted. On your USB boot stick you want to enter something like this: (Understand this just came out of my head, and may be filled with bugs).

set fogserver:ipv4 <ip_addr_fog_server>
set next-server ${fogserver}
chain tftp://${fogserver}/undionly.kxpe

That “should” tell the ipxe client using the boot media to chain to the fog server. Then the fog server will take over from there.

@need2 said:

I’m late to the party, but no I never got WDS and FOG to play well together.

@FlowLive was just able to do this. Its a bit down in the thread, but he has FOG acting as the PXE boot server then you can select the WDS environment from the FOG Menu.

I went the other way around and setup FOG to be able to boot the MDT litetouch image so that you can deploy through MDT from the FOG boot menu. Two different ways to have FOG and MDT/WDS coexist.

The normal PXE boot sequence is that it waits for the dhcp server to provide the dhcp IP address and then requests the dhcp option values of 66 (boot server IP) and 67 (boot file). For option 66 you should enter the IP address of your FOG server (not the conical name) and for 67 you should enter undionly.kpxe. Those settings will tell teh client what to do next. If you see some messages about iPXE then you have connected to the FOG server.

Lets start out by getting some background information.

I see you are running Ubuntu 14.04. What version of FOG are you running are you running the stable 1.2.0 or the SVN trunk release (pre 1.3.0)? If you are running the trunk release what is the number in the cloud (upper left corner or page) on the management console?

Now that is out of the way. Have you ever setup PXE booting before (FOG or something else)?

What did values did you set for DHCP option 66 and 67?

Here is a step by step for rhel 6 distros
https://forums.fogproject.org/topic/6074/extending-a-file-system-on-redhat-base-distros

I can’t help you with the fogstorage node issue, but I can with the space issue.

Actuality what I’m going to recommend will sound counter intuitive, but its the right move. Traditionally with *nix systems you place the /home and /opt directories on their own partitions/disks. These two partitions could potentially grow out of control filling up the root partitions ( / ) basically killing the *nix systems.

What I’m going to recommend is that you do the following.

1. Create a new VMDK file for your storage drive
2. Create a filesystem on that vmdk and mount it on /mnt/itemp
3. Stop FOG and all of its services
4. Move the contents of the /opt directory to /mnt/itemp
5. At this time /opt will be empty with the contents not under /mnt/itmp
6. Now unmount /mnt/itmp and mount that new vmdk on /opt (this will restore the contents of /opt but the files will now be on the new vmdk
7. Now for the last bit. Make a directory under /opt/fog called images
8. move the content of /images to /opt/fog/images (this will place all of your big files on this new vmdk file
9. Now that all of the contents are in /opt/fog/images delete the /images folder and create a symbolic link from /images to /opt/fog/images

You may need to check the file permissions before you start and after you’re done to make sure things are sane. But IMO this is a better setup than placing both the /images and /opt/fog/snapins on the root partition. In this setup the /opt directory (vmdk) can go to 100% and it won’t take down the OS.

But if you want to extend your current VMDK I did a write up a bit ago on how to do this with rhel 6 based system.

I guess a question that is still not clear to me is when you ran the install.sh on the current trunk, the installer “should” download the latest client. Did that step complete as expected? Or was there an error saying that needed to manually down it?

IF your fog server sits behind a proxy server you need to update wget (what the installer uses to get the msi) with your proxy server settings.

[Edit] Well if you followed Wayne’s instructions and it worked, then something in the installer or svn trunk is broken. The wget in Wayne’s instructions should have thrown an error if there was a proxy issue [/Edit]

I’m going to guess that its not possible to change the display order. There is no concept of display order in the database so I assume the display order is being set based on the record ID where boot from the hard drive is record ID 1. I would not recommend messing with any record ID to alter the display order.

You will receive much bad karma if you do go ahead with this plan.

@FlowLive said:

Now I read that it’s not possible to change the menu order of things, WDS being all the way down, can’t I just bump it up a little?

Where it should be, way at the bottom. Where do you think the competition would put it, front and center??

I’m glad/happy you got that wds thing fixed out. From what I’ve read those forward / backslashes are a PITA to get worked out. It looks like you got the right combination to get it done.

@FlowLive

Looking and the chain load link that you sent. I did see this reference

  set netX/next-server ip.of.wds.server
  imgexec tftp://${netX/next-server}/boot/x86/wdsnbp.com

What I think I might have missed is in my guess is setting the network adapter 0:

set wdsserver:ipv4 192.168.1.88
set net0/next-server ${wdsserver}

Its interesting that the chain does work (kind of) that it does gram the wdsnnp.com file, but there is something in the referral that its missing to continue.

FWIW: All of the 7010 SFF Dells we have have “Intel 82579LM Gigabit Network Connection” according to our inventory tool.

Just thinking out loud here, but if target device doesn’t support PXE net booting in UEFI mode, might it support USB booting in UEFI mode. The thought is build a ipxe boot environment on a usb flash drive and boot off that. I’m pretty sure that ipxe boot environment can be setup to boot off usb flash. The question would be does the hardware support this…

^^^^ (just using a graphic to isolate ideas)

On some hardware platforms you need to enable the legacy ROMS to get support for both USB and PXE booting in UEFI mode. I don’t have any lenovos here to test that idea on.

@Jbob said:

@george1421 Centos 6 appears to be missing some of the modules I require. I’ve been trying to hunt down the exact dependencies I’d need, or if need be, a new policy with different rules / modules.

That’s why we are going through this. No worries on my part. We’ll get them worked out.

@Jbob said:

@george1421 there is a DHCP section in the post. I will include DNS as well. Right now I wish to keep those options separate. Eventually the installer should automatically configure the firewall based on installation preference.

Sorry I missed that. I copied the top sections and skipped the bottom.

No luck compiling for me on Centos 6.7

[root@centtest SELinux]# make -f /usr/share/selinux/devel/Makefile fog.pp
Compiling targeted fog module
/usr/bin/checkmodule:  loading policy configuration from tmp/fog.tmp
fog.te":50:ERROR 'unknown class service used in rule' at token ';' on line 3260:
allow unconfined_t init_t:service enable;
#============= unconfined_t ==============
/usr/bin/checkmodule:  error(s) encountered while parsing configuration
make: *** [tmp/fog.mod] Error 1

During the install of FOG, it asks to make fog a dns and dhcp server but those ports are not listed in the script.

For clarity, I took and rebuilt a clean centos box. I set the firewall rules and then installed the latest SVN trunk. I just remembered that I need to set the selinux policy since it is centos defaults. But any way the plan is to apply your settings to a clean install install your policies and then install the latest SVN trunk.

Ok a couple of things that hit me right away.

The instructions above for rhel is for Centos 7.X and newer. The firewalld function is not available on Centos 5 and 6. Centos 5 iptables is a mess so I wouldn’t even recommend installing FOG on Centos/rhel 5. So for Cento/rhel 6 you must use the “Other” instructions.

The second thing I ran into using the Other instructions is the first line for adding the modules to the iptables-config. The last IPTABLES_MODULES entry config wins. So after running that first line and restarting iptables I only had one new module loaded “nf_conntrack_netbios_ns” The product of that line looks like this in the iptables-config file.

IPTABLES_MODULES=“nf_conntract_tftp”
IPTABLES_MODULES=“nf_conntrack_ftp”
IPTABLES_MODULES=“nf_conntrack_netbios_ns”

The correct syntax should be

IPTABLES_MODULES=“nf_conntract_tftp nf_conntrack_ftp nf_conntrack_netbios_ns”

I have the screen shots if you need them but I think that info should get you pretty close.

@FlowLive said:

Took longer than expected, had the FOGFTP error and had to fix that.

Now I try the option in the menu but unfortunately all I got is a blinking cursor

crud, I just noticed that the forum ate some of my command line. Lets try it as code. You have to pay attention to the double forward and back slashes.

Menu Item: wds.BootIt
Description: Launch WDS Boot x86
Parameters:
set wdsserver:ipv4 192.168.1.88
set next-server ${wdsserver}
chain tftp://${wdsserver}/Boot\\x86\\wdsnbp.com

The slowness in the forum is driving me a bit nuts today. Its taking me almost 12 minutes to post this.

The typical workflow for a fully configured but blank FOG server is this

1. Before you build your reference image, pxe boot your reference image computer (VM) into the FOG pxe menu and select full registration. Enter the computer name and any other parameters necessary and submit the registration to the FOG server.
2. Reboot the reference image computer and then build your reference image.
3. Once your reference image is built then sysprep and power off your reference image.
4. From the FOG menu locate the reference image computer that you registered before and setup a upload task to capture the image from the reference computer. Don’t forget to check the box to power off the reference computer after the capture is done.
5. PXE boot the reference computer and the upload will begin automatically. Once the upload is done the reference computer will power off. Set this aside for now.
6. PXE boot your first target computer into the FOG menu and register it. Once it is registered you can select quick image to download the captured image to the first target computer.
7. Fine tune and repeat.