@Tom-Elliott its no big deal. once this summer is over, adding new hosts gets more rare. It was just this being our first roll out, there will be a lot of machies to touch. All good though, rather have FOG do what it needs correctly instead of me tearing it up.

@Roger-Saffle I’m just guessing that the ca was updated at some point either knowingly or unknown. I’d suggest updating the image that has the 0.9.10 client on it.

I wanted to thank you guys for your help. I updated to the Trunk (version 8179) and its working out just fine, its able to see the new PC’s and can register/image them without any problem. Thanks again.

That seems to have worked!

@adukes40 I guess if I was more on top of my game tonight, I’d have seen the 500 page and said to look at the Apache Error logs, and I imagine we would see some errors about memory being exhausted… I’m glad you figured it out.

Sorry to bring this up again, but having the same issue. Different fog server however. using build 8179.

I created a small group of hosts ~25 and went to the group page, checked the box and it didn’t autofill the form, thought OK, remembering this thread that didn’t necessarily matter. Then went to a host and saw the box was checked to join and the old information that was in there was now gone. I then tried to update the individual host by clearing the fields and hitting the join box, still didn’t autofill, checked the defaults in Settings and that was still there and correct. I believe this was working on SVN5675.

Browser is Chrome 51. (Now that I think about it I think my chrome updated last week sometime)(perhaps chrome issue?)
Tried MS Edge, Works as expected.

Can someone verify my results?

@9intelligence cool.

@dmoseley What’s typically happening if the fog client is running, is that during the OOBE setup the fog client may do an unscheduled reboot which kind of whacks the OOBE setup. Starting the service at the end of the setupcomplete.cmd makes sure that windows is fully setup before the fog client connects the device to AD or renames the computer.

Ubuntu 16.04?

@Tom-Elliott Thanks very much for the quick response Tom, I figured that was the case. So there’s no way to specify a next-choice pxe option (like chaining another server), just using something from “Exit to Hard Drive Type”?

Yes, sorry, we are on Fog version 1.2. Yes, I want to display the “Full Inventory and Registration” task for all Hosts registered or otherwise in the PXE menu.

Our naming convention is Buildingname-room-systemnumber,

example would be ZES-204-S-01

You guys are awesome!!!
OK, So here is what was happening.
WDS was still running on an old server (I am trying to decommission but it has our lunchroom software)
Stopped WDS and Boom!!! Now I have PXE boot.
I still have to move my datastores for image locations but now can at least boot into PXE.
Lots of tweaking and I am sure questions. But I feel I am moving in the correct direction.
Thanks!
Rob
Please mark as closed.

@Sebastian-Roth Updated to fail in the case we can’t create the mount directory OR the partition cannot be mounted. Because I write the data to a log, I am also able to output the “reasoning” behind the failure too to try to help people hone in on the issue.

v0.11.0 is released and has replaced GreenFOG with PowerManagement. See the News post for more details.

I have a feeling that the way I’m going about installing the latest trunk might be causing the issue.

I’m using a Dell Optiplex 3010 as the fog sever that only has one NIC and intend to put the system on a network that has not internet connectivity.

When I install fog 1.2.0 (which works for some reason) I’m switching between DHCP and manual Ethernet configurations throughout the process to complete the install.

Example: I’ll set a manual Ethernet configuration of 192.168.1.1 etc, start the fog installation and choose my settings, then since the install reaches out to the internet during installation I’ll switch it to DHCP to grab and install the files. When it wants me to update the schema I’ll put it back to manual 192.168.1.1 to do that until the install completes.

It’s a dumb method I know, but works for 1.2.0 and what I have.

I’m trying the same thing for trunk and I guess this method I’m trying doesn’t work with it.

I’ve ordered a second NIC for this 3010 and will attempt to install it without screwing with the Ethernet settings during install to see if it will proceed correctly this time.

I’ll report back when the NIC arrives and I attempt to install Trunk again.

@Oleg said:

ok. That means we need to wait now until it’s solved in partclone.

Yes. As soon as there is a new partclone version out (please inform us if you hear about it first) Tom can include it into the init’s and hopefully we should be fine then. But I don’t know when that will be.

@alorsung What version of FOG are you running? On the login screen, it’s displayed in the top left.

Hi,

Mwebb_Sureline’s boss here.

I’ve cobbled this together to do what was needed.
It is rather quick and dirty I’m afraid, could probably call “doInventory” and use the results from there to populate the service tag / serial number rather than directly running the dmidecode in the script.

I’ve seen a few people on the forums trying to get something similar working, so here is the code.
Thanks go to @george1421, I modified his script in the post he linked to make this.

(Using SVN rev 5698) I’ve copied fog.man.reg and setup a new custom job for this, the code replaces the hostname input section from fog.man.reg directly, e.g. remove lines 156 to 167 and paste this in.

Whilst this works for us, you use this at your own risk.

while [[ $res != "#!ok" ]]; do echo " * Retrieving serial number or service tag for this computer to use as hostname " getserialno=`dmidecode -s chassis-serial-number`; getserialno="${getserialno%"${getserialno##*[![:space:]]}"}"; # Remove trailing whitespace # get the serial no from the baseboard if bios doesn't give it up if [ "$getserialno" = "" ]; then getserialno=`dmidecode -s baseboard-serial-number`; getserialno="${getserialno%"${getserialno##*[![:space:]]}"}"; # Remove trailing whitespace fi if [[ "$getserialno" = "" || "$getserialno" == "None" ]]; then echo " * Unable to retrieve tag or serial number, please enter hostname manually. Just hit return to leave blank (will use MAC address)" read getserialno getserialno="${getserialno%"${getserialno##*[![:space:]]}"}"; # Remove trailing whitespace fi host="$getserialno"; if [[ ${#host} -gt 15 ]]; then host=${host:0:15} echo " | Truncated to 15 characters: $host" usleep 2000000 fi echo " * Hostname set to: $getserialno" host=$(echo $host | base64) res=$(wget --post-data="host=$host" -qO - http://${web}service/hostnameloop.php 2>/dev/null) [[ $res != "#!ok" ]] && echo "$res" done

@Arsenal101 To ensure all goes smoothly on the next upgrade, edit the /opt/fog/.fogsettings file.

Look at the snmysqluser= snmysqlpass= and snmysqlhost fields and make sure they’re set properly.

In your case:

snmysqluser='root' snmysqlpass='' snmysqlhost='IP/Hostname of FOG here'

@LJedi Could you please upgrade to the latest trunk and see if the d1.fixed_size_partitions is now created properly in your case?

Thanks Tom for looking into this and fixing it!