You ask really good questions

It looks as if a new tftboot directory was created as there is a tftboot.prev directory as well as the tftpboot one. That indicates to me that the new server was made from scratch, and the images that were common between the two were manually copied over.

That being said, in the pxe.config directory are entries aside from the default that carry the MAC addresses of my computers in question as well as a reference to the image 5800R2. I’ve removed these files and will try my luck with PXE again.

This is already progress from where I was, so even though you’re just asking me for more information you are indeed helping me so thank you very much.

Edit: That did the trick! Thank you!

Bingo. After my adventure to the MDF I can safely confirm that the 5800R2 image is nowhere to be found on the storage node, by the way.

My apologies for not being more thorough. Both the old server and the new one are/were .32, and the new server uses the same IP and storage node as the old one.

I’m going to guess that your next suggestion will be to see if I can’t hunt down this 5800R2 on the node someplace, so I’m going to pop in to the MDF and have a look right now :rolleyes:

We had an image on fog server 1 called 5800R2.

Fog server 1 died.

We now have an image on fog server 2 called 5800

Any time I PXE boot a specific handful of computers, they boot to immediately upload to the image 5800R2. This server does not exist, but the machines look like they complete this upload successfully. All I want is for them to deploy this new 5800 image though on fog server 2.

Has anyone seen anything like this before? I’ve even zero’d the hard drive and flashed the BIOS on these machines, and they still PXE boot directly to this image upload.

[quote=“jflanagin, post: 30035, member: 24486”]So I did a start/stop tftpd service on the FOG server and now im not getting the time out but its still doing what the video does above.[/quote]

So I’ll go one farther. My T540p worked just fine yesterday, and now does exactly what your video does no matter what. I haven’t made any changes, so my thought was spanning-tree; however, I thought that only mucks with ipxe? This happens before that comes in to play!

Also, I’ve tried Tom’s June 17th unofficial kernel, and that isn’t helping either.

[B]Updated: [/B]I tried a fresh T540p out of the box, and it works just fine! Unfortunately, that’s not the one I’ve been prepping an image on for the whole day so it doesn’t do me much good. My thought is that maybe the 2.13 BIOS update broke it somehow?

[B]Updated Again: [/B]So I took another fresh one out of the box, PXE booted it, and that worked. I restarted, and the thing can’t PXE boot anymore it does the same thing in your video. Nothing has been done to this computer other than look at the FOG menu. Either I missed a switch when I turned on portfast, or this model Lenovo has a rockstar for a NIC.

So I solved the problem, and thought I’d let everyone know what it was since I’ve been on here for a couple of days now banging my head against a wall.

As it turns out, an ambitious young would-be network technician decided to go ahead and change what switch my office is on in the MDF down the hall. That switch hasn’t been used in forever since it was so unreliable. I can’t even ssh in to it. I put my office back in the appropriate switch and everything works just like it ought to.

As far as I can tell this has been my issue since square one, and I just wanted to come thank everyone for the help and input along the way to this Mad Hatter’s Tea Party of a solution (chaaaaaaange places!).

Thanks again!

Last issue and I’ll be out of all of your hair I promise! I’ve got a brand new VM installed with Ubuntu 14.04 and Fog 1.1.1. Here are the settings I chose when installing Fog:

[INDENT=1]Here are the settings FOG will use:[/INDENT]
[INDENT=2]Base Linux: Debian[/INDENT]
[INDENT=2]Detected Linux Distribution: Ubuntu[/INDENT]
[INDENT=2]Installation Type: Normal Server[/INDENT]
[INDENT=2]Server IP Address: 10.6.116.243[/INDENT]
[INDENT=2]DHCP router Address: [/INDENT]
[INDENT=2]DHCP DNS Address: 127.0.1.1[/INDENT]
[INDENT=2]Interface: eth0[/INDENT]
[INDENT=2]Using FOG DHCP: 1[/INDENT]
[INDENT=2]Internationalization: 0[/INDENT]
[INDENT=2]Donate: 0[/INDENT]
[INDENT=1] [/INDENT]
Once everything was installed I updated my database, and restarted mysql, apache2, and tftpd-hpa. Maybe it’s something, maybe it’s nothing, but upon restarting apache2 I get the message:

[INDENT=1]Could not reliably determine the server’s fully qualified domain name, using 127.0.1.1.[/INDENT]
[INDENT=1] [/INDENT]
I figure that’s exactly right since my DNS server is elsewhere, but I’m putting it in here just in case.

Everything seems just peachy. I can log in to the web interface, and update my kernels, and we should be good to go, right? I boot up a machine to test, and I get the following:

[INDENT=1]CLIENT MAC ADDR: XX XX XX XX XX XX GUID: XXXXXXXX XXXX XXXX XXXX XXXXXXXXXXXX[/INDENT]
[INDENT=1]CLIENT IP: 10.6.4.7 MASK: 255.255.255.128 DHCP IP: 10.6.112.80[/INDENT]
[INDENT=1]GATEWAY IP: 10.6.4.1[/INDENT]
[INDENT=1]TFTP.[/INDENT]
[INDENT=1]PXE-M0F: Exiting Intel Boot Agent.[/INDENT]
[INDENT=1] [/INDENT]
I have no idea what I’ve done wrong, but if someone could help me out I don’t think I have the vocabulary to describe how much I’d appreciate it.

[quote=“NickConrad, post: 30590, member: 24656”]I’m sure not! I’ll check right now.[/quote]

You are right as always: 127.0.1.1

[quote=“Tom Elliott, post: 30589, member: 7271”]Are you sure it’s 127.0.0.1 and not 127.0.1.1?[/quote]

I’m sure not! I’ll check right now.

[quote=“DigitalSauce, post: 30571, member: 24673”]Hello,

When I went to check the state of my fog server today I connected through the web browser. I was prompted with the Database Schema Installer / Updater. Thinking nothing of it I clicked to have it update. Unfortunately now I get prompted with it every time I connect. I never get to log in. If I reboot my sever I have a small window of time where I can connect and log into the web management, but it will quickly boot me off to update Schema.

Any help would be great.[/quote]

[CODE]sudo service mysql restart[/CODE]

This resolved the same issue when I saw it yesterday.

[quote=“Tom Elliott, post: 30586, member: 7271”]What do you mean by:?[/quote]

Sorry to be unclear. When going through the fog installer it asks if you’d like to use fog as your dhcp server or if you have another one in mind. I said to use our existing dhcp server. The next question is if you want to input an IP for a DNS entry on the DHCP server. I wasn’t sure what that meant, but saw a tutorial saying to say yes to this question and leave the default option for the IP address at this step. That default address is 127.0.0.1.

So I installed a fresh copy of Ubuntu today (14.04)

Installed a fresh copy of Fog (1.1.1)

I told it to use my AD DHCP server and loopback at 127.0.0.1

Everything works just great, until I PXE boot a client to start imaging. I get as far as

TFTP. . . . .
PXE-E32 TFTP Open Timeout

At which point it exits PXE and boots Windows normally.

My [I]thought [/I]was that I hadn’t set option 66 and 67 in my DHCP server correctly. I went to my DHCP server. The IP of this fog server is not in any scope, so I went to IPv4 Server Options instead of Scope Options and set 66 and 67 (to my fog server’s ip address, and undionly.kpxe resepectively).

Again, TFTP works in every fashion other than over PXE boot.

Did I set my options wrong?

so with TFTP off, the message I get at PXE boot is the expected TFTP . . . . . until it finally times out. This is making me think the issue is specific to TFTP. I went and checked my username / password, and for some reason the web interface was dropping part of the password for the user I provided. I changed the user to something that has the same access but a plain text password. The web interface likes that just fine, but I get the same

TFTP.
Exiting

Screen when booting over PXE. Are there other usual suspects in the TFTP settings I should be cluing in on?

[quote=“Tom Elliott, post: 30474, member: 7271”]Do you have another DHCP server on your network, or is FOG the only one?[/quote]

I have my DHCP service running on Active Directory as well, though it doesn’t have a scope for the IP that my fog server is on

Just to see what would happen, I made a symlink like Tom had suggested, but I made it to pxelinux.0.old. No change when booting a client over PXE

DHCP Server Configuration file.

see /usr/share/doc/dhcp*/dhcpd.conf.sample

This file was created by FOG

use-host-decl-names on;
ddns-update-style interim;
ignore client-updates;
next-server 10.6.116.243;
filename “undionly.kpxe”;

subnet 10.6.116.0 netmask 255.255.255.0 {
option subnet-mask 255.255.255.0;
range dynamic-bootp 10.6.116.10 10.6.116.254;
default-lease-time 21600;
max-lease-time 43200;
option domain-name-servers 10.6.116.246;
option routers 10.6.116.246;
filename “undionly.kpxe”;
}

[quote=“Junkhacker, post: 30464, member: 21583”]if that is a direct copy, i see the issue

undionly.kxpe should be undionly.kpxe[/quote]

not a direct copy, so just a dumb typo. Good catch, though, you made me double-take for a second between screens :eek:

[quote=“NickConrad, post: 30457, member: 24656”]the full error was “pxelinux.0: no such file or directory” which is why I figured it was that.

As for undionly, it’s in tftboot as a .kkpxe, .kpxe, .kpxe.INTEL, and .pxe.[/quote]

Tom, the following is my dhcpd.conf located in /etc/dhcp3, did I goof somewhere along the way here?

[INDENT=1]use-host-decl-names on;[/INDENT]
[INDENT=1]ddns-update style interim;[/INDENT]
[INDENT=1]ignore client-updates ;[/INDENT]
[INDENT=1]next-server xxx.xxx.xxx.xxx.[/INDENT]
[INDENT=1] [/INDENT]
[INDENT=1]subnet xxx.xxx.xxx.xxx netmask xxx.xxx.xxx.xxx {[/INDENT]
[INDENT=2]option subnet-mask xxx.xxx.xxx.xxx;[/INDENT]
[INDENT=2]range dynamic-bootp xxx.xxx.xxx.xxx xxx.xxx.xxx.xxx;[/INDENT]
[INDENT=2]default-lease-time 216000;[/INDENT]
[INDENT=2]max-lease-time 43200;[/INDENT]
[INDENT=2]option domain-name-servers xxx.xxx.xxx.xxx;[/INDENT]
[INDENT=3]option routers xxx.xxx.xxx.xxx;[/INDENT]
[INDENT=3] [/INDENT]
[INDENT=2]filename “undionly.kpxe”;[/INDENT]
[INDENT=1]}[/INDENT]
[INDENT=2] [/INDENT]

[quote=“Tom Elliott, post: 30454, member: 7271”][code]sudo ln -s /tftpboot/undionly.kpxe /tftpboot/pxelinux.0[/code]

Will create the link for you. I’m already aware that pxelinux.0 doesn’t exist. What that command “no such file or directory” is telling you and maybe me is that /tftpboot/undionly.kpxe doesn’t exist on your system. This might very well be your issue.[/quote]

the full error was “pxelinux.0: no such file or directory” which is why I figured it was that.

As for undionly, it’s in tftboot as a .kkpxe, .kpxe, .kpxe.INTEL, and .pxe.

[quote=“Wolfbane8653, post: 30452, member: 3362”]This shows that the file exists and it is accessible to the network without firewall conflicts. Next I would direct you to your option 66 and 67 settings but Tom is already on that. Soooo…keep with him.[/quote]

Will do, and thanks for your help thus far