@Wayne-Workman said in DHCP failed - undionly.pxe:

It’s really easy to do.

@george1421 said in DHCP failed - undionly.pxe:

@tigerpride said in DHCP failed - undionly.pxe:
FWIW: Only the dhcp server needs to be 2012 the reset of your domain can remain on 2008. But you will have to migrate sooner or later. 2008 server is already 8 years old.

If you don’t want to upgrade and can isolate the uefi and bios clients to different dhcp scopes you can do it that way too. Just create different scopes with different boot files for each scope. Its not really clean but will work. In my case we only pxe boot when we image machines, the default target computer will boot from the hard drive and only boot pxe when we select it from the boot menu.

Probably is a snap and I realize it needs to be done, but the thing is, I’m just week 4 into my new position and environment. DHCP is done by the primary DC. I could move it to the secondary (which is 2012) but I don’t know what’s all dependent yet. I’d rather not risk it at this point since my users will return in a week and a half. I’ve a ton on my plate as is, and the main LAN + VOIP system potentially tanking is not really how I’d like to remember my start here.

To be honest I’m waiting for Server 2016 to come out of RC to set up a test tertiary and see about winking off 2008 if that’s to my liking.

@forte647 said in DHCP failed - undionly.pxe:

My company recently switched to Probook 455 G3s, and I had the same issue after the switch. Something about the NIC in these laptops isn’t playing nice with undionly.pxe? I switched my boot file to ipxe.pxe and it worked. So maybe try using ipxe.pxe and see what happens.

You got it! That works. No dhcp failed or kernel panics! I’m grabbing the stock image as we speak, then setting up a fresh W10 image to base a capture from for the rest of this lot.

TYVM! Thanks for all the help everyone!

Follow the instructions from Tom. I occasionally have to reblank the root password for the db.

sudo -i
mysql -u root
ALTER USER 'root'@'localhost' IDENTIFIED WITH mysql_native_password BY '';

Worked to fix my FOG 1.4.4 on Ubuntu 16.04.2 LTS

@Wayne-Workman said in DHCP failed - undionly.pxe:

It’s really easy to do.

@george1421 said in DHCP failed - undionly.pxe:

@tigerpride said in DHCP failed - undionly.pxe:
FWIW: Only the dhcp server needs to be 2012 the reset of your domain can remain on 2008. But you will have to migrate sooner or later. 2008 server is already 8 years old.

If you don’t want to upgrade and can isolate the uefi and bios clients to different dhcp scopes you can do it that way too. Just create different scopes with different boot files for each scope. Its not really clean but will work. In my case we only pxe boot when we image machines, the default target computer will boot from the hard drive and only boot pxe when we select it from the boot menu.

Probably is a snap and I realize it needs to be done, but the thing is, I’m just week 4 into my new position and environment. DHCP is done by the primary DC. I could move it to the secondary (which is 2012) but I don’t know what’s all dependent yet. I’d rather not risk it at this point since my users will return in a week and a half. I’ve a ton on my plate as is, and the main LAN + VOIP system potentially tanking is not really how I’d like to remember my start here.

To be honest I’m waiting for Server 2016 to come out of RC to set up a test tertiary and see about winking off 2008 if that’s to my liking.

@forte647 said in DHCP failed - undionly.pxe:

My company recently switched to Probook 455 G3s, and I had the same issue after the switch. Something about the NIC in these laptops isn’t playing nice with undionly.pxe? I switched my boot file to ipxe.pxe and it worked. So maybe try using ipxe.pxe and see what happens.

You got it! That works. No dhcp failed or kernel panics! I’m grabbing the stock image as we speak, then setting up a fresh W10 image to base a capture from for the rest of this lot.

TYVM! Thanks for all the help everyone!

@Wayne-Workman

That is useful information to put with the instructions on the wiki. I come from zero FOG experience, I was basically handed a broken 0.34 server on 12.04 that did not meet my organization needs any longer (someone deleted all the images too). I decided to stick with it and get it Win10 deployment compatible. Dist-upgrading it broke it even more so I might as well go blank slate.

@george1421 said in DHCP failed - undionly.pxe:

@tigerpride Simply, yes it would.

BUT there is an option if you are using a windows 2012 server for dhcp services or a linux dhcp server. Both have a process you can setup pretty easy. The process is outlined here: https://wiki.fogproject.org/wiki/index.php?title=BIOS_and_UEFI_Co-Existence

Argh, the DHCP is on Server 2008 still and by the looks of the wiki the setup won’t work. I don’t feel comfortable migrating it to 2012 quite yet. I’ll have to call in some help and get approval to get that going. I may just have to bite the bullet and experiment with WDS instead.

The good news is, undionly.kpxe seems to work for my non-uefi clients. Was undionly.pxe a typo from me?

@george1421 said in DHCP failed - undionly.pxe:

@tigerpride said in DHCP failed - undionly.pxe:

I can get to the Fog menu when using pxelinux.0 for dhcp bootfile, but the inventory will fail with a kernel panic.

This is understandable and expected, pxelinux.0 is not supported since 0.3x version.

When I switch dhcp bootfile to undionly.pxe, I get a DHCP Failed error.

Really what I would start with is undionly.kpxe if that doesn’t work then try undionly.kkpxe Understand this is for bios/legacy mode computers. If you are running a uefi based computer then you want ipxe.efi.

I’ve uefi and non-uefi in my environment. Would setting one or the other affect the ability to pxe boot from either?

Everything is in the documentation, but it’s presentation is a bit awkward.

https://wiki.fogproject.org/wiki/index.php?title=Ubuntu_16.04

1. Installation from scratch

This describes the installation of the server pretty well. I’m not sure why partitioning is considered tricky, I just used the default guided for my VM and used the whole disk. I picked bare install (no LAMP).

2. Update / Dist-Upgrade

This wasn’t applicable to me.

2.1 MySql Password behavior.

This was an issue, but at this point mysql wasn’t installed for me yet. I had to fix the root password after install and then it was good to go.

2.2 Customized my.cnf

Not applicable.

2.3 PHP Issue

Not an issue.

2.4 MySQL 5.7 Issue

Wasn’t an issue.

3. Installation

This is missing all steps to clone from git. I used the tutorial for the trunk and accidentally installed FOG with default settings a few times, which of course failed the php install until I got here:

–> php_ver=‘7.0’ php_verAdds=‘-7.0’ ./installfog.sh -y <–

This was the most important bit, but it’s only one line at the bottom of the page. I must’ve missed it a dozen times as I was switching back and forth between forums/wiki pages/documentation. My installs kept failing with the regular install instructions.

Making the subheaders smaller to make it more clear that the issues are a sub of the Update / Dist-Upgrade will go a long way.

Hello,

I just got Fog 1.3.0-RC6 running on Ubuntu Server 16.04.1 (the install manual wiki page really could use some better directions).

I want to inventory and set up a stack of new HP Probook 355 G3s, but I’m having problems PXE booting these.

I disabled secure boot and enabled legacy boot.

I can get to the Fog menu when using pxelinux.0 for dhcp bootfile, but the inventory will fail with a kernel panic.
When I switch dhcp bootfile to undionly.pxe, I get a DHCP Failed error.

Is there anything I should be configuring differently?