KK20, thank you VERY much for this! I saw this answer in other places, but I wasn’t sure if it applied to my FOG issue. I was having this issue imaging some Optiplex 9010 all-in-one units…

That is most likely a problem with permissions. Please search the forums, the Wiki or Google. There should be a lot of threads regarding this problem.

OK, so the lack of response on this thread leads me to believe there is nobody else out there that is having this problem and nobody who has a clue how to fix it. However, this problem still persists for me but I have had a bit of a break-through today (!) and thought I would share it with the community on the off chance that it’ll be useful to somebody one day, and if not that then just for bragging rights! 😛
So I had another biggish order to process today (10 machines) and they were all Dell GX620s, PDs, 2GB and Win 7 Pro. I set up the fist machine, registered it on Fog and renamed the Host to “Gems 1”. (I always rename the Host name from the mac address to a reference of the customer) Selected the image I wanted to deploy and clicked update. Decided to do a quick image as I will probably just use this one machine to load all of the HDDs for the order so I didn’t create a task for it. Selected Quick Image from the PXE boot menu and I got an error saying “Quick Image Failed… Invalid Host Information”. So went back onto Fog Management and created a task for it instead. I get the DREADED HASH AGAIN! :mad:

Getting a bit fed up of this now I went back to the host information on Fog Management and decided to tweak it a little. First thing I did was rename the host from “Gems 1” to “Gems0”. Then I deleted all of the “Host Description” (Usually Fog will put in the date and time the host was created by default) so it was completely blank. Then in the Host Kernel, I pointed it to an old Kernel that we used on an old Fog server that never had this problem (Kitchen Sink 2.6.31.1) and finally in Host Arguments I typed in “nomodeset”.
Now, the last two things I mentioned (Alternative Kernel and nomodeset) I have tried many times before and it got me nowhere but I thought I’d try them with the other things I did just for good measure. Clicked update on the Host, killed the previous task for it and then created another one. PXE booted the machine and it imaged fine!! 😄

YES!!!

Now I’m not completely sure which variable actually did the trick but it worked. It could have been the name change, deleting the Host Description, or it could have been a combination of the whole lot. The next time I get this problem with a machine though I will try one method at a time and try to figure out what is actually causing this.

Will post as soon as I get answers!

I went through the same problem and found that the script service/common/functions.php use this function getIPAddress to get the correct IP address of my machine:
function getIPAddress()
{
…
exec( “/sbin/ifconfig | grep ‘[0-9]{1,3}.[0-9]{1,3}.[0-9]{1,3}.[0-9]{1,3}’ | cut -d’:’ -f 2 | cut -d’ ’ -f1”, $arR, $retVal );
…
}

The problem is that my default installation language is PT-BR (Portuguese), so this command line does’t work because of character space added at the start of the line:

ifconfig

eth0 Link encap:Ethernet Endereço de HW e8:39:35:9b:b2:de
inet end.: 172.17.10.106 Bcast:172.17.10.255 Masc:255.255.255.0
endereço inet6: fe80::ea39:35ff:fe9b:b2de/64 Escopo:Link
UP BROADCASTRUNNING MULTICAST MTU:1500 Métrica:1
RX packets:5497666 errors:0 dropped:0 overruns:0 frame:0
TX packets:16486831 errors:0 dropped:0 overruns:0 carrier:0
colisões:0 txqueuelen:1000
RX bytes:360971401 (344.2 MiB) TX bytes:23808915837 (22.1 GiB)
IRQ:18

lo Link encap:Loopback Local
inet end.: 127.0.0.1 Masc:255.0.0.0
endereço inet6: ::1/128 Escopo:Máquina
UP LOOPBACKRUNNING MTU:16436 Métrica:1
RX packets:2294 errors:0 dropped:0 overruns:0 frame:0
TX packets:2294 errors:0 dropped:0 overruns:0 carrier:0
colisões:0 txqueuelen:0
RX bytes:678626 (662.7 KiB) TX bytes:678626 (662.7 KiB)

/sbin/ifconfig | grep ‘[0-9]{1,3}.[0-9]{1,3}.[0-9]{1,3}.[0-9]{1,3}’ | cut -d’:’ -f 2

172.17.10.106 Bcast
127.0.0.1 Masc

So you have to change this script line to represent your default language.

I found the problem!

I put a PCI network card and it works.

That is the problem is in the Ethernet card onBoard.

The solution is to put PCI ethernet cards on all machines, the problem is that they are about 20 machines.

Does anyone know where the settings FOG move to change some parameters?

\ Gilberto Ferraz

The only other way to massively remove hosts is to access the database directly and remove them from there. It’s easy if you want to just remove all of them:
[CODE]TRUNCATE TABLE hosts;[/CODE]
Or if you have some known host value that you can leverage; such as all hosts currently assigned imageID 3:
[CODE]DELETE FROM hosts WHERE hostImage = 3;[/CODE]

If you’re not used to messing around in MySQL then you might want to consider installing phpmyadmin, which will give you a web interface for managing the database.

I don’t know if this is any help to you now, but someone else may benefit:
I was getting GRUB errors with DelFX710 thin clients running Windows 7 WES (window Embedded). These thins had the most recent updates applied and would be unbootable. I found some fresh out of the box without the updates (same thin client hardware). Perfect images no problem. The defining difference (I think) is that the non-updated ones used GRUB 1.5. The new ones GRUB 2.

For what its worth.
cheers

Hi,

I have done the above quoted steps already,but still it is not working.
I came to know that if I give exact broadcast address in wol.php file,its working.
But its not possible to hard code all subnets there.
can u pls help?

Regards,

Avatar

As I mentioned previously, you will need gPXE-1.0.1+ in order to boot your EFI based systems from network. another option is to get the live pxe version of Clonezilla and create an entry on your pxe menu which points to your Live Clonezilla. With this in place then you can upload/download EFI, EUFI, EFT, EXT, NTFS, FAT based file systems.

Try a different Kernel for that machine. You can setup multiple kernels so if one kernel works on one model, but not another, you can fix that. The newest kernel necessarily work on all computers that the old one did from what I’ve seen.

Hi there. Sorry for the massive late reply - this hasn’t fixed the issue. There were no processes waiting or clogging up the system. :-z

When i type ‘fog’ and hit enter whilst on that screen, it loads for a second then reads;

[code]“Cannot read “/proc/partitions”. (No suck file or directory). Then, you must use the command line to run Partition Image. Type ‘partimage–help’ for help.” [/code]

then;

[code] You haven’t accss to /proc/mounts.[/code]

Any ideas?

The youtube video above is still happening (even when I’ve tried reformatting the laptop, re-uploading it to a new image/host - and when i try using an existing image that works on other laptops of the same make/model).

I got it to work, was an error in our autolog program, thanks for all of the help

It does not image it. It just say database updated. And computer client just reboot again.

I would like to apologize in advance for this not being in the “hardware” section, sorry!

Try the following fix:
[url]http://fogproject.org/forum/threads/cannot-create-tasks-task-manager-blank.3823/#post-9593[/url]

Let us know what you come up with for the PXE boot menu wont you? I’m intrigued and interested in doing similar things.