FOG .33b PXE Boot woes
-
Hi all. Ive been a long time fog user, i think since .27 and love the heck out of the software. I am trying to get .33b setup on a machine running Ubuntu 12.04.4. It is connected on an isolated network for testing using a simple linksys home router for the DHCP server and dnsmasq. When i boot up an HP 8300 elite client it is able to connect to my fog server but when it hits TFTP it stops and exits the boot agent immediately. When i do the same with an HP8000 elite i get the following:
TFTP.
PXE-T01: File not found
PXE-E3B: TFTP Error - File not foundI have verified that the TFTP server is running on the fog server and frankly i am stumped.
I’ve followed all of the instructions from [url]http://www.fogproject.org/wiki/index.php/Using_FOG_with_an_unmodifiable_DHCP_server/_Using_FOG_with_no_DHCP_server[/url]
and i have searched this forum extensively to see if i can solve my problem. Any insight would be greatly appreciated! -
This post is deleted! -
Did you also follow the Chainloading iPXE using dnsmasq setup?
right now it is telling you it can’t find the boot file we need to make sure that you have set up DNSmasq properly to accept the undiony.kpxe file and that it is the correct location.
Also if possible, please include a screenshot. From the sounds of things it’s not finding the pxelinux.cfg/default file unless it fails before trying to load the files, this is where the screenshot will help.
-
This post is deleted! -
I did follow the chainloading instructions under the link i provided earlier.
My /tftpboot folder:
fogadmin@FOG2:~$ ls /tftpboot/
boot.txt ipxe.krn pxelinux.0.old snponly.efi undionly.kpxe
default.ipxe memdisk pxelinux.cfg tftpboot vesamenu.c32
fogadmin@FOG2:~$ ls /tftpboot/pxelinux.cfg/
default default~ pxelinux.cfg[ATTACH=full]650[/ATTACH][ATTACH=full]651[/ATTACH]
[url=“/_imported_xf_attachments/0/648_default pxelinuxcfg.jpg?:”]default pxelinuxcfg.jpg[/url][url=“/_imported_xf_attachments/0/649_ltspconf.jpg?:”]ltspconf.jpg[/url][url=“/_imported_xf_attachments/0/650_default pxelinuxcfg.jpg?:”]default pxelinuxcfg.jpg[/url][url=“/_imported_xf_attachments/0/651_ltspconf.jpg?:”]ltspconf.jpg[/url]
-
I meant a screenshot of the machine failing to load the pxe boot info.
Did you edit the dnsmaq.conf file? I see it open in the editor there, I recommend leaving those settings alone and only editing the ltsp.conf
-
found it.
dhcp-boot should be
[code]
dhcp-boot=/tftpboot/undionly.kpxe
[/code]After you make the change, restart the dnsmasq service,
That is probably my fault I will make adjustments on the wiki.
-
Oh lol sorry.
HP 8000
[ATTACH=full]652[/ATTACH]HP 8300
[ATTACH=full]653[/ATTACH][url=“/_imported_xf_attachments/0/652_hp8000.jpg?:”]hp8000.jpg[/url][url=“/_imported_xf_attachments/0/653_hp8300.jpg?:”]hp8300.jpg[/url]
-
[quote=“Paul Freeman, post: 25137, member: 23545”]Oh lol sorry.
HP 8000
[ATTACH=full]652[/ATTACH]HP 8300
[ATTACH=full]653[/ATTACH][/quote]See the post above this one, I believe the problem to be in the ltsp.conf file for the dhcp-boot: value.
-
Hmmm, made the change and still the same result.
-
restart the service:
[code]sudo ufw disable
sudo service tftpd-hpa restart[/code] -
Here is what my /var/log/syslog file reports for that last boot attempt.
[ATTACH=full]654[/ATTACH]
[url=“/_imported_xf_attachments/0/654_var log syslog.jpg?:”]var log syslog.jpg[/url]
-
you are using eth1 not eth0… these are very important pieces of information to include. are you actually using eth1? If so you need to edit your default file to reflect that information, try net1 in place of net0, if not an error has been made and I would re-configure the server again
-
Yes this is correct I am using eth1. I have been looking through all this for so long and it never dawned on me that this would need to be changed in the default file. I definitely do not pretend to be an expert on linux and its various functions, but have been able to poke by so far.
-
Seems like i may have other issues going on. On boot the tftpd-hpa service is running but when i try to restart the service it hangs. When I get the status of the service it comes back as tftpd-hpa start/killed. Also after running ufw disable, the command completed successfully however upon a system reboot the firewall service is again active.
-
make sure you didn’t set DNSMasq to be the tftp server in the conf settings.
[code]
sudo apt-get purge ufw
[/code]
Usually the disable command stops the Firewall from running, if it is still running after you have disabled it, that will get rid of the firewall for you.also why are you restarting the tftp service? you need to restart dnsmasq.
-
TFTP Is not enabled in the dnsmasq.conf or ltsp.conf. I think at this point I may reconfigure the machine. I did create a clone of the drive before installing fog so that i could roll back. Would you recommend that I just use eth0 to make the setup more straightforward?
Thank you both for your assistance.
-
[quote=“Tom Elliott, post: 25142, member: 7271”]restart the service:
[code]sudo ufw disable
sudo service tftpd-hpa restart[/code][/quote]I was restarting as per this post by Tom. I will remove the firewall and try again.
-
It doesn’t matter which NIC you use but you need to be consistent, I have only ever used one NIC so I am not much help in setting up or configuring the other NIC
but it can’t be too difficult, just remembering to supply the new NIC information anytime you run across one mentioning the OLD NIC.If I remember correctly someone raid bonded two NIC and was getting some insane speeds… Maybe I should look into this…
[quote=“Paul Freeman, post: 25150, member: 23545”]I was restarting as per this post by Tom. I will remove the firewall and try again.[/quote]
ahh, I see, sorry I missed his post that makes sense now haha
-
Well after purging UFW and a reboot i get something slightly different when trying to boot a client. Rather than immediately stopping at the TFTP boot sequence the HP8300 pauses there for a minute as though its trying to connect. Though still does not connect…
The HP8000 still gives the same error.
