Update!
So I upgraded to Ubuntu 18.04 which did not resolve the issue. I then updated FOG to 1.5.9.71 which must have been release today 
. After completing the FOG update, the services are running and I was able to successfully deploy 2 machines using multicast. I rebooted the server and verified that the services started after a reboot. Thanks for your help and working on this awesome software!
@tom-elliott Yes sir, I believe it is complete. During my version changes/installs it all seemed to go normally. We were running 1.5.5 I believe and after purchasing some new computers (Dell Optiplex 5080’s) it was having a driver error when booting. I updated to, I think it was, 1.5.9.62 and it fixed the issue and was able to capture and deploy single machine images. I only noticed the problem when we attempted to multicast. I then went to 1.5.9.68 to 1.5.9 and back to 1.5.9.68 to try and resolve the issue. We can still deploy one machine at a time and have full access to the web interface.
Edit: I do remember watching the most recent install more closely. When it was at the starting services portion they all said “OK” from what I can remember. Would updating my version of Ubuntu help you think?
@sebastian-roth I am running Ubuntu 16.04 LTS
@Tom-Elliott I’m not super versed in Linux so I was just looking at other threads trying things that may have helped others. The command was “systemctl restart /lib/systemd/system/FOG*.service” which was what you had posted in the thread I linked.
I just tried “systemctl restart FOGImageReplicator.service FOGImageSize.service FOGMulticastManager.service FOGPingHosts.service FOGScheduler.service FOGSnapinHash.service FOGSnapinReplicator.service”. Services still say “failed” when I check them.
Thanks for all the replies!
Hey all. I have run into an issue with my FOG server. Some services won’t start on reboot and I cannot start them manually either. I am currently on 1.5.9.68 but was also having the issue on 1.5.9. When I run “systemctl --failed” I get the following:
root@FOG:~/fogproject/bin# systemctl --failed
UNIT LOAD ACTIVE SUB DESCRIPTION
â—Ź FOGImageReplicator.service loaded failed failed FOGImageReplicator
â—Ź FOGImageSize.service loaded failed failed FOGImageSize
â—Ź FOGMulticastManager.service loaded failed failed FOGMulticastManager
â—Ź FOGPingHosts.service loaded failed failed FOGPingHosts
â—Ź FOGScheduler.service loaded failed failed FOGScheduler
â—Ź FOGSnapinHash.service loaded failed failed FOGSnapinHash
â—Ź FOGSnapinReplicator.service loaded failed failed FOGSnapinReplicator
I ran across this other post that looked identical to my issues. I followed the instructions to update the Systemd scripts but get this result:
Failed to restart lib-systemd-system-FOGImageReplicator.service.mount: Unit lib-systemd-system-FOGImageReplicator.service.mount not found.
Failed to restart lib-systemd-system-FOGImageSize.service.mount: Unit lib-systemd-system-FOGImageSize.service.mount not found.
Failed to restart lib-systemd-system-FOGMulticastManager.service.mount: Unit lib-systemd-system-FOGMulticastManager.service.mount not found.
Failed to restart lib-systemd-system-FOGPingHosts.service.mount: Unit lib-systemd-system-FOGPingHosts.service.mount not found.
Failed to restart lib-systemd-system-FOGScheduler.service.mount: Unit lib-systemd-system-FOGScheduler.service.mount not found.
Failed to restart lib-systemd-system-FOGSnapinHash.service.mount: Unit lib-systemd-system-FOGSnapinHash.service.mount not found.
Failed to restart lib-systemd-system-FOGSnapinReplicator.service.mount: Unit lib-systemd-system-FOGSnapinReplicator.service.mount not found.
Any help would be appreciated!
@andythemayor Lol, my bad. I thought you were OP. I didn’t read…
@andythemayor I’m not an expert and maybe someone else can chime in, but FOG doesn’t boot to any type of Microsoft Windows environment. As far as I can tell it is booting to Linux. Where you have “test.pxe” in your DHCP server settings you need to have “undionly.kpxe”. If you read through the wiki that @Wayne-Workman posted it should get you sorted out.
@jgallo said in Custom Full Registration Menu:
FYI if you jump to 1.5 you will need to modify items in the fog.man.reg file as I learned.
@tom-elliott said in Postinitscripts and storage node replications:
@jgallo In particular, to fix this for you. Edit your “curl” lines. CHange the parts that say
http://${web}...to just state${web}...(… = all the other data).
I upgraded to 1.5.2 recently and everything seemed to be working smoothly. Today I tried to add a new host to update an image and would get an error saying “Unable to register host: (/bin/fog.man.reg)”. I rebooted FOG, reinstalled 1.5.2, and tried looked around here to see if anyone else had a similar issue with not a whole lot of luck at first. I then went back into my post history when I was asking about customizing the Full Registration Menu and found the answer. See below. Thanks @JGallo for posting in that thread and thanks @tom-elliott for helping him originally.
We have been using FOG for about 6 months now and all has been running smoothly…well at least until yesterday lol. We received a new demo unit (Optiplex 5055) to test with and I wanted to make sure that it would work with FOG. After messing about in the Windows 8-esque BIOS, I could not for the life of me get it to see the M.2 SSD, SATA SSD, or USB when in Legacy mode to boot from. So I was left with having to re-enable UEFI and start from there. My quest led me to this Wiki to get me setup with PXE booting both UEFI and Legacy computers on Server 2012 R2. I followed all the steps and no dice, UEFI still would not PXE boot. I tried all the different UEFI architectures and still nothing. So after fiddling about I noticed that the screen capture in Step 7 was a little different from what I was seeing. I had an “Append wildcard()” box as shown and a “Prefix wildcard()” box. So ticked both boxes and finished the setup. Upon PXE booting the new machine with UEFI it worked!!! So I just wanted to share my experience in hopes that it would save someone some frustration and hair pulling. On a side note after deploying a test UEFI image, I was getting the error “Failed to set disk guid”. I found out this is related to GPT which seems to be used by default when using UEFI instead of MBR. After updating from 1.4.4 to 1.5.2 the error went away and so far everything seems to be working.
After checking the log I saw that it was still looking at eno1. I found where I missed changing the interface under the default storage node under Storage Management. After changing the Interface setting from eno1 to enp7s0f0 all is good. Hopefully someone else will find this useful.
I finally received the parts to upgrade my FOG server to use a 10 Gigabit NIC. After putting in the new NIC and optics I also changed the IP to be on our server vlan instead of a test vlan. First I set the new static IP on Ubuntu with the new NIC. I then went and changed the IP in: /opt/fog/.fogsettings, Web Interface -> Storage Managment, Web Server -> FOG_WEB_HOST, TFTP Server -> FOG_TFTP_HOST, and in /tftpboot/default.ipxe. I then changed the interface from eno1 to enp7s0f0 in the Web Interface under NFS Server -> FOG_NFS_ETH_MONITOR and Multicast Settings -> FOG_UDPCAST_INTERFACE. I then reran the fog install and rebooted for good measure. Most everything seems to be working normally, I can access the web interface and can PXE boot and clone an image to a client. The only thing I noticed that was not working was when I cloned a client, the bandwidth GUI on the home page did not show my bandwidth utilization like it did prior the network swap. Did I miss a setting somewhere?
FOG 1.4.4
Ubuntu 16.04
My environment is complex. I plan on using a combination of groups for some computers and also not grouping other computers at all. For the computers that won’t be grouped, I plan on using the multicast task under Image Management for cloning. After the multicast session has completed, it reboots the computer. I have searched the forums and read through a few different wiki articles and couldn’t find what I was looking for at first. What I wanted to accomplish was shutting the computer down instead of rebooting it after the multicast session was done. I found this post that was asking for something similar where @tom-elliott replied. I just wanted to say that this solution worked for me. From my understanding this is a global setting so it will shutdown after any task has completed and I am fine with this. The purpose of this post is dual fold in that I wanted to say thank you, and also to link this solution to the multicast preference I wanted. This way if anyone else searches for multicast shutdown upon completion, hopefully they will find this. (See below for setting to add into the WebGUI)
FOG: 1.4.4
Ubuntu: 16.04
@tom-elliott said in New PXE menu entry: Deploy Image & Shutdown:
@george1421 It definitely is the easiest way, but if it’s specific to “quick imaging” goto:
FOG Configuration Page->iPXE Menu Configuration->fog.deployimage
InBoot Options:add:shutdown=1If you want it for ALL tasks (will not impact debug of course):
FOG Configuration Page->FOG Settings->General Settings->FOG_KERNEL_ARGS and add:shutdown=1
@jgallo Oh I see now. I needed to modify the fog.postinit file to call patch.fullreg. So now the last line in fog.postinit reads
. ${postinitpath}patch.fullreg With that final modification, my custom registration menu now works! Thank you for all the help, this community has been great and I am really liking FOG.
I apologize in advance for my noobness. Ok, so I have the edited fog.man.reg.fix and put in the postinitscripts directory along with patch.fullreg. Do I need to restart a particular service for these to take effect?
@george1421 Ok cool, thanks. I’ll run through the tutorial then. I just wanted to make sure before I started making changes so that I didn’t break anything.
Re: Custom Full Host Registration Menu for 1.4.0.rc2 and later
So I was going to remove a few menu items that I don’t need in the registration menu and was reading through this tutorial. When I went to go look for /bin/fog.man.reg, I found that my /bin directory does not have fog.man.reg. Has something changed for 1.4.4 or am I looking in the wrong place? (I’m on Ubuntu 16.04)
@tom-elliott Great! Thanks, will do.
@tom-elliott The IP did change at some point. It was assigned by DHCP when I first installed Ubuntu but I created a reservation for it on the DHCP server afterward with a different IP. I just cannot recall if I changed it before or after installing FOG. Since I will be needing multicast in the future, I did some reading and found where it said to change the hosts file (located at /etc/hosts) IP entry for my server hostname from 127.0.0.1 or 127.0.1.1 to my current static IP and then restart the multicast service (sudo systemctl restart FOGMulticastManager). Now the multicast log is showing “No Tasks Found!”. Success! Thanks again for all your help!
Edited to give more info on the multicast fix.
@tom-elliott That did it! I wonder why even after rebooting the server it didn’t work? On a side note I was looking at the Multicast log and it also showed the master node error message. So on a hunch I changed the IP on the storage node from my internal static IP to 127.0.0.1. After that the Multicast log said “No Tasks Found”, which I’m taking as a good sign as that it can read the storage node. I changed the IP back to the static IP on the storage node and then ran the command to restart the service and it now displays the image file sizes for the two new images. My question is should I leave the static IP on the storage node or set it to 127.0.0.1? Thank you all for the help!
Current Status:
Image Size: On Server - Functioning
Multicast Log - Showing “This is not the master node”
