Okay. So.
I got around to testing again.
The same problem remains.
Using ipxe.efi - the LAN Boot works just find, but when exiting from the fog menu or just letting it time out the system tries to boot via refind.
It finds a hard drive (after changing to uefi_deep_legacy_scan active) but then there is an error: Failure booting legacy (BIOS) OS
The harddrive is a nvme SSD drive.
I edited the php.ini file as suggested here: https://forums.fogproject.org/topic/6800/php-fatal-error-allowed-memory-size-of-134217728-bytes-exhausted-fog-1-2-0-missing-fog_memory_limit/2
in order to get rid of the above error.
Now that error doesnt appear anymore in the error_log
But nothing else does either and the web gui still doesnt work 
I just noticed the server time is off by an hour, maybe that plays into this?
Edit: Naw ā¦ Same problem, trying to check the scheduled tasks freezes the web GUI
I get - before I start the scheduled task even, this error:
[Fri Sep 15 07:38:47.258945 2017] [:error] [pid 4691] [client 192.168.43.88:34096] PHP Fatal error: Allowed memory size of 134217728 bytes exhausted (tried to allocate 32768 bytes) in /var/www/html/fog/lib/fog/fogcontroller.class.php on line 261, referer: http://fog.lfdw.local/fog/management/?node=host&sub=edit&id=82
After I start a cron job the web gui says it saves fine, but I get the same above error message in /var/log/httpd/error_log.
Then when checking in tasks -> scheduled tasks that stops the web gui from working but in the log file nothing else happens.
Since RC9 when I go to the tasks menu and want to open the scheduled tasks, the Web-Frontend stops working.
No more page opens, even after waiting for a very long time no reaction, reopening the web UI in a new tab doesnt change anything, no fog page loads.
Deleting Cookies and Clearing Cache helps, login is possible again, as soon as I open scheduled tasks the same thing happens again.
Problem with my server? Or with the RC?
I just noticed that there are two places in FOGs configuration where you set what exit type is chosen, maybe I had a mixup there since I never registered it as being two separate settings for bios and efi.
Iāll check that on monday and report back if I may.
If you have any other input Iām very thankful.
Thanks for the hint with the TPM - for some reason I had it fixed in my mind that wouldnt work. Just checked, it does work.
Regarding the other issue:
I converted (using mbr2gpt) the disk of a windows 10 installed client to gpt. But alas, still the same error messages along: Boot from SAN device 0x80 failed: No such device and so on as stated above
I just dont get it ā¦ when in my BIOS Settings I have only UEFI set it happens, compatibility mode activated or deactivated it happens ā¦ only way to work around it is to not autoboot from LAN, then it starts (with UEFI exclusively on - no legacy).
Off-Point: Thanks for the link, Iāll have a look into it. Is there an easy way to find out what PXEClient: I will have to enter there for our hardware? We are using mostly Lenovo Laptops (X and T versions) and mostly HP Desktops, though there are a few older HP Laptops as well.
On Point: I suppose Iāll have to go and set up a legacy (bios) and an Uefi Image as well.
In general, I would prefer to use UEFI since it allows me to use the TPM in order to Bitlock our drives. We have to have the bitlocking functionality activated.
Using BIOS we would have to use a password with Bitlocker, something I would like to avoid since most systems have the ability to boot using EFI.
I guess the reason for not completely switching to UEFI is that I want to keep clients that might not be able to boot via UEFI functional and still use FOG with themā¦
Hi
Thanks for your help.
I am using our exisiting Windows DNS Server as DHCP.
I removed the uefi_deep_legacy_scan and I already had the scanfor setting activated (sorry for not mentioning it).
No change though.
So do I get it right that I would have to install our Windows 10 Clients as UEFI Images in order to get it work with rEFInd? Or would that throw up other problems then?
I changed the default for bios mode back to SANBOOT, the error I get is:
Boot from SAN device 0x80 failed: No such device
and then twice:
Could not boot: No such device
So, we are trying to get FOG running with UEFI in order to be able to use the TPM for Bitlocker on our Windows10 systems.
I got as far as being able to boot to the FOG menu from the PXE Boot by changing the boot file to ipxe.efi from ipxe.kpxe - before that there were troubles with the computer reverting back to the boot selection screen.
Deploying via FOG works now.
We have LAN Boot set to first priority - which would be great to have in order to make it easier for us to manage our systems via FOG without having to manually boot via LAN.
During boot I get the rEFInd menu and then when trying to boot from the harddisk - that it finds and shows after I changed some settings in the refind.conf (uefi_deep_legacy_scan) - I get the error: āFailure booting legacy (BIOS) OS.ā
Any idea on where I could look to solve this?
The other options in fog configuration (GRUB, SANBOOT, etc.) dont work either.
I get a long list of errors, all along the line of: āACPI Error Namespace lookup failure [CPC1]ā when trying to deploy an image.
Thank you so much!
No, it is not 
Is there a better way for me as an āaverageā user to report things like that?
Running Version 20
So far everything seemed to work fine, update went without a hitch.
But when I go to the groups list and try to open the details about a specific group - no matter which one - i get a HTTP 500 error.
All other detail views (in Hosts, Images, ā¦) work just fine.
When I go to:
http://fog.lfdw.local/fog/management/index.php?node=about&sub=settings
(this is saved in chrome as a visited link and comes up when I enter fog.lfdw and then autocomplete)
the login page shows up, but clicking on āLoginā doesnt do anything. The page doesnt change.
If i manually go to http://fog.lfdw.local/fog/management the same login page shows up and clicking on login logs me in and brings me to the dashboard.
Cleared Cache (Strg+F5 to reload) - same āproblemā.
EDIT: Upgraded to working branch - Version 20, same thing happens.
EDIT2: This happens from ?node=about and ?node=groups. ?node=host, ?node=image and ?node=storage work fine.
Workaround for the group thing:
From āList all Hostsā: selecting and adding Hosts to a group works.
Some of the āconfirmationā buttons donāt do anything. For example:
Opening the Web Interface via: http://fog.lfdw.local/fog/management/index.php?node=group&sub=membership&id=6 doesnt work - the page just sits there. When I truncate the url to ā¦/fog/management and try it works.
When trying to add Hosts to a group from the groups membership menu the button āAddā again doesnāt do anything. Havenāt found a workaround yet.
Tried with both Internet Explorer and Google Chrome. No scripts being blocked in either from the browser side.
Thanks for your feedback.
Cleaning the databases didnt solve it, will wait for the change to come to one of the higher stable branches.
I have two āincompleteā entries in my Pending Hosts list:
No matter what I do, I cannot seem to approve - or delete - the entries here. I do get a confirmation when I try:
but when I check the Pending Hosts list again, the two entries are still there.
Does that indicate a problem? Or is there a way to delete these entries?
Things I have tried:
Reset Encryption data for all my Hosts
Another thing that shows up in the log:
Several of the modules (for example: TaskReboot or the HostnameChanger) list:
Middleware::Response Module is disabled on the host
Where would I even do that? On the server - as far as I can see from the web-gui - the modules are activated. I also deactivated and reactivated them to see if that would kickstart anything - no success.
EDIT:
Okay ā¦ I just saw that one of the two hosts I deployed the image to had the modules deactivated in the Service Settings. I have no idea why, it is enabled globally and the other machine does not have them deactivated.
Now in the fog.log the Modules are shown with a response success on the one machine (the one that had the services disabled) but the āUnable to get subsectionā message still appears on the other machine.
EDIT2:
The computer that had the services disabled just restarted - so in that case it seems everything works.
EDIT3: After reading some more I tried resetting the Encryption Data and aw and behold - everything seems fine now. Thanks for reading along if you have
Sorry if Iām being complicated - just trying to figure this out