@BSZAdmin Possibly this issue has to do with the computer having been turned off for a fair amount of time before you get the request to move it to another department?? See here: https://www.prajwaldesai.com/workstation-trust-relationship/

The more I read about this the less I think this is caused be the fog-client. But we’ll see. Please try things as suggested and we’ll take it from there.

@Wayne-Workman
@george1421

I figured this one out… I was having problems with security. One of Windows 10’s recent updates (as in the last 3 months or so) was blocking connections to anonymous shares. It allowed them but they needed to be authenticated. For those of you who stumble across this post…

Add your FOG server to your domain and configure Kerberos. After you do this you will need to configure your driver share. I have included the way I did mine and it works. This configuration is for using an existing domain controller. In our case, we use Windows Server 2012 to do domain control.

Do the following:

yum install samba*
yum install krb5-libs krb5-workstation

Confuguration Kerberos /etc/krb5.conf

[logging]
default = FILE:/var/log/krb5libs.log
kdc = FILE:/var/log/krb5kdc.log
admin_server = FILE:/var/log/kadmind.log

[libdefaults]
default_realm = YOURDOMAIN
dns_lookup_realm = false
dns_lookup_kdc = false
ticket_lifetime = 24h
renew_lifetime = 7d
forwardable = true

[realms]
DOMAIN.TN = {
kdc = dns-name-your-domain-controller
}

[domain_realm]
netbiosnameyourdoamin = NETBIOSNAMEYOURDOMAIN
netbiosnameyourdomain = NETBIOSNAMETYOURDOMAIN
Configuration samba /etc/samba/smb.conf

[global]

workgroup = DOMAIN
password server = dns-name-your-domain-controller:88
realm = NETBIOSNAMETYOURDOMAIN
security = ads
idmap uid = 16777216-33554431
idmap gid = 16777216-33554431
template shell = /bin/bash
winbind use default domain = false
winbind offline logon = true

log file = /var/log/samba/log.%m
max log size = 50

passdb backend = tdbsam

load printers = yes
cups options = raw

[homes]
comment = Home Directories
browseable = no
writable = yes

[printerdrivers]
comment = All FOG Printers
path = /PATHTODRIVERS
browseable = no
guest ok = no
writable = yes

Add dns name in /etc/hosts

Install and configuration ntp server (It’s important for use Kerborose authorization)

yum install ntpd
edit /etc/ntp.conf
server ip-address-your-ntp-server prefer
Create ticket
kinit account-admin-for-active-directory@NETBIOSNAMETYOURDOMAIN
Add server in domain

net ads join -S dns-name-your-domain-controller -U account-admin-for-active-directory
Create keytab for Kerberos

net ads keytab create -U account-admin-for-active-directory
Edit file /etc/nsswitch.conf

passwd: files winbind
shadow: files winbind
group: files winbind

Restart samba and windind

Test
net ads info
wbinfo -t

@kkptrck said in Skipping partition restore (MBR Only):

/dev/sda1 : start= 64, size= 1076291, type=27

I have missed one important thing here when you first posted this. The start sector 64 is quite unusual (usually it’s 63 or some other sector but I have not seen 64 yet) and I can imagine this being the cause of your issue. Did you manually do the partitioning?

@bogle No model specific drivers are copied to the target computer during image push and then inserted into the windows environment using pnputil utility.

https://forums.fogproject.org/topic/11126/using-fog-postinstall-scripts-for-windows-driver-injection-2017-ed

I probably should update it for 2019 / Win10 specifically, but everything is in there.

I do have MDT add in the WinPE10 standard drivers so the golden image has at least the core network and disk drivers, but most of the windows 10 built in drivers are good enough to get to at least the point where the pnputil program is used.

@george1421 I got some tricky ways to find if dnsmasq is using the configuration that we created. In the directory /etc/dnsmasq.d/, we created the configuration file, but if there is no other file called README, I think dnsmasq did not use the configuration file. But after I updated dnsmasq, there is a new file called README. And it works!!! I am not sure if my thoughts was right. But it got working in my case.

@george1421 Done, hope that’s more obvious now.

I added some more checks and a more appropriate error message so people will find out more easy in future versions of the fog-client.

@Sebastian-Roth Thanks for your help! That fixed the issue.

Thanks for the insight! I copied one of my build images to a test-copy, then made two changes. I then ran sysprep /generalize /reboot /unattend and allowed the VM to sysprep and reboot to see what happens. I didn’t capture it to run it through FOG, but I imagine the results will be mirrored once I do.

I noticed that your unattend file didn’t contain a reference to SetupComplete. Therefore, I edited unattend.xml to remove the <Component> block that contained <AutoLogon> and <LogonCommands> (within which was contained the reference to SetupComplete.cmd).

I moved unattend.xml to the Panther directory at your suggestion. When I ran sysprep, I tested it without defining a file, and sure enough, it picked it up with no problem.

After allowing sysprep to do its job, I logged in as the local admin (since I didn’t have FOG to join the domain for me), and no command windows popped up to try running SetupComplete as the local user. SetupComplete didn’t appear in the Startup tab of Task Manager. FOGService was set to Automatic and was Running. In other words, the things that weren’t supposed to be happening no longer were, and the things that were supposed to happen WERE! 🙂

 

By the way, slightly on the topic, I know the wiki suggests restarting the computer with a command in SetupComplete after enabling the FOG Client, but Microsoft’s documentation suggests this is a bad idea:

Warning You cannot reboot the system and resume running SetupComplete.cmd. You should not reboot the system by adding a command such as shutdown -r. This will put the system in a bad state.

My SetupComplete file contains two lines:

sc config FOGService start= auto
net start FOGService

It seems to still restart if needed, so maybe the shutdown line can be removed?

 

Thanks to @fry_p, @george1421, and @Sebastian-Roth for your contributions in helping me figure out this strange issue!

In addition to what @Shad0wguy posted, your golden image should be developed on an isolated network, or at least one that doesn’t have internet access for just that reason. You don’t want windows updates to help you “fix” your system while you are preparing for it. If you are doing this in audit mode, for sure disable the windows update service until just before sysprepping it. I myself use MDT to build the golden image so one of the task sequences is to update the windows updates from our WSUS server and not to use direct internet updates. That way the updates are applied and completed in a known state before MDT exits. We will typically sysprep and capture with FOG right after MDT finishes.

But its great you are documenting the problem here and the potential solution. Because if you have the issue there are 100 other folks that will have it in the future. So thank you for helping them out too.

@seahunter Well we can take the fall back approach and build a FOS usb boot drive to bypass iPXE altogether. I’m pretty sure that bzImage will work if we boot from a usb flash drive.

Here are the instructions. Read all of my posts including the caveats in taking this approach: https://forums.fogproject.org/topic/7727/building-usb-booting-fos-image

Also look at the forum chat bubble in the upper right corner of the page for a few additional hits.

@duanekaeb Thanks for the pictures. What I posted was the command parted -l - the parameter being a lower case l (like lion). But using the print command you still got us exactly the information I was hoping to see.

Ok, so the partition layout is fine from my point of view. So now I am wondering if C: is simply almost full and that is causing the problem. Reviewing the pictures again I saw something that I have not noticed before: Needed relocations: ... (16179 MB)

While I am not an expert on the ntfs3g-lib implementation I am really wondering if this is just “too much for it”. It tries to shrink the partition and for that to be able to do it would need to relocate/move roughly 16 GB of data. While it’s not nice that it fails to do so I can kind of understand it’s having a hard time to do so.

I really wonder why defrag didn’t help you with this. Probably a bit different on the SSD, hmmmm. Not sure what else we can do for you.

Yes, I’ll try to get to those. I am under a time crunch right now so I can not get to that right away.

I deployed the image that gave me the error after it was created, and it deployed successfully.

@Russell18 said in Cannot Install One Or More Boot Critical Drivers:

Does this help in way?

Truthfully, no.

FOG doesn’t care about drivers at all, your issue is with WinSetup/OOBE.

From a windows standpoint, how are you installing the drivers into your windows golden image? Are you deploying a universal image for all hardware or do you have hardware specific images? Are you sysprepping your image before you capture with FOG? If you have a single golden image did you install all drivers for all possible models you are deploying to? Have you done the Shift-F10 thing and look at the windows setup logs in c:\windows\panther directory to see what driver/issue its having troubles with?

As you might see there are more questions than answers. But at this time the issue during deployment is after FOG’s job is done, your issue is within WinSetup/OOBE.

Thank you for your help. Removing the client was the solution. I looked into and as before I didn’t need any of the functions it assisted with. Thank you for your help.

The drive still doesn’t resize correctly, but I will wait for the release of 1.5.7. At the time thought the two may be related. Looked through the reported bugs and didn’t see one that specifically called out that issue, but I will keep looking.

Thanks again for all of your help!!

+Craig

@bravetoad Anything after that? Interesting that it says “Response Invalid Host”. From what you wrote so far I suspect the host to be registered in the FOG web UI already, right?

Do you see anything in the apache or PHP-FPM logs. See my signature on where to find those.

It looks like a sysprep /generalize into audit mode before capture solved the problem. Thank you for your help!

@Fazyl Is this your first contact with FOG? Just wondering if you ever had an image that captured and deployed properly?

Please post the contents of the files d1.partitions, d1.minimum.partitions and d1.fixed_size_partitions which you find in the /images/... directory on your FOG server.