its okay thanks for your help

0_1500471038531_091f047d-6905-4f8b-81ae-16f6e29ee5ec-image.png

Thank you.

@ewilk2600 You have to do it from terminal on the Ubuntu Server.

Your database will update to the newer format automatically (you just have to navigate to the FOG management URL when it asks you to and click upgrade), but it will make a backup first, just in case.

@sudburr I’m not sure what is going on or what you’re trying to do - but virt-manager typically only connects to hosting systems via SSH, not the guests. Virt-Manager provides direct console access to guests, but SSH to the guest is not involved in this, it’s via ssh to the host.

@sudburr The reboot isn’t necessary, but otherwise great job on the post below - it’ll help many folks in the future.

I’m using mostly ubuntu and all my images are set to “single disk, resizable” .

the order you should go in short:

on the desktop pc:
setup the desktop you want to clone
enable pxe boot in the bios of your desktop pc

on the fog server (gui):
create new image (single disk, resizable) (image management)
register host pc with the server(there are multiple ways to to that, check the wiki 🙂 )
assign your created image with the entry of the desktop pc (select host image in host management)
click on capture image at the entry of your desktop pc
if you don’t run the fog client on the desktop pc, you should reboot your pc so it does the magic 🙂

deploy a image:
is mostly the same as above: create new host, assign image, click on deploy.

you should also look for the “fog client” - running it on the desktop pc is very usefull! 🙂

I hope I did not forget anything. 😉

Thank you for the response.

I ended up going back and comparing the file to another one I downloaded on another computer, and they were different sizes. Moved the new one in and it was able to mount fine.

@notaspy If you force this system to fail, then boot the fog server up.

Run the following commands.

ps aux | grep dhcp this should show you if the dhcp server is running in memory. netstat -an| grep 67 this should show you that the dhcp server is listening on its data port. You should see something like this udp 0 0 0.0.0.0:67 0.0.0.0:*

Lets start with those. That will give us an idea if the dhcp-server is running and / or listening.

Edit: It looks like ubuntu 16.04+ is based on systemd so these commands will be useful. Don’t use them until we can understand what’s going on here.

sudo systemctl restart isc-dhcp-server sudo systemctl status isc-dhcp-server sudo systemctl enable isc-dhcp-server

@VijayBomble This thread is extremely old and I have no idea what your particular problem is. based on the message of “on this switch it works, but that one it doesn’t” leads me to think a configuration issue on the switch that’s not working. There’s not anything that I can do to figure out what the problem is.

@george1421
I thank you, George for you help, for me it’s ok to solve this topic.
cordially

@george1421 im very sorry to make you confused with word “cobbler”, yes, it is an opensource project

if akin to FOG. later, i will try FOG to compare it.

but now, i think my problem is how use dhcp + dnsmasq.proxy mode + ipxe install rhel67(old kernel without efi_stub) with uefi client make it general.

or how gurb2 auto looking for grub.cfg @ ipxe environment.

i was confused that, why grub can’t use anyother args like — grubx64.efi --net_bootp --net_default_server=xxxx --configfile=xxxxx/grub.cfg, or like ipxe support ${mac(ip):hexraw} to make grub.cfg be general. that’s will be perfect to support network install.

how FOG do if someone need install rhel67 with general cfg @ dnsmasq.proxy mode & uefi client(for now all hp gen9 server use efi.)

thank you.

Perfect !
I substitute the conf files and replaced “Fog IP Server”,now it works !

Solved !
Regards

in /etc/environment, i marked
http_proxy=“http://xxx:3128”
https_proxy=“https://xxx:3128”
ftp_proxy=“ftp://xxx:3128”
np_proxy=“localhost,127.0.0.1,@IPfog”

and ./installfog.sh works

thanks a lot !

Solved :
The http proxy is not correctly configured.
Edit /etc/wgetrc and add this line :
no_proxy = <ip server fog>

Explanation :
The step “Ensuring node username and passwords match” lunch wget request to local ip.
This is the same issue for update halt at step “backing up database”

I had that happen before when one of the repo’s were being blocked by our network filter.

@FoSTerCyber This issue only happens after upgrading, not on fresh installs it seems.

@mlerch so just a hint here: the fog settings are kept in /opt/fog/.fogsettings file. That is where fog knows what the user account and password should be.

Worked with @keith75140 and we seem to have gotten things fully operational.

While it is true the OP used the fog user as an admin user, we were able to get things working properly. PXE Booting is now working as well after working with the network team to get things setup.

We had a couple of issues but we were able to work around them.

You’ve tried “EFI Exit” with rEFInd? This really doesn’t seem to be a problem with LVM or disk information, directly.

It really really feels like it’s trying to boot across incompatible types.

Please try network booting using EFI by enabling the Network Stack and PXE Network Stack. Then try setting the “EFI Exit” to rEFInd.

This will mean you need to use one of the EFI labeled boot files, but this can typically be automated.

@kleanthis Well, you have to write the scripts, so if they’re good to go, then yes.