The bzImage kernel updates did the trick! Thank you so much for your help!

Okay, so we’ve actually ran into an issue with the updates.

When going to Kernel Update in the FOG web interface settings, I am getting an FTP error:
“Type: 2, File: /var/www/html/fog/lib/fog/fogftp.class.php, Line: 707, Message: ftp_put(): Could not create file., Host: XX.XX.XX.XX, Username: XX”

We tried to get around this by manually FTP-ing into fog and copying the file ourselves, which we were able to do with no problems but we still get the error. We tried all of the steps on this page, but nothing works.

The other option is to manually upgrade, as described here, but that will automatically upgrade to the latest version, which I believe is 4.17.0. Is there a way to manually upgrade the kernel to a specific version?

Is there a reason why I should only upgrade to 4.15.2? Can I update to the latest kernel?

Your explanation makes sense, thank you for your response!

1. I am running 1.4.4 (which I now see is not the most recent)
2. The kernel version (bzImage version) is 4.11.0
3. The network device is an Intel 1219-LM

Thanks for your help!

We recently got a new shipment of HP EliteDesk 800 G4 desktops, but imaging is not working. Before we had 800 G2’s, and they still work fine (I just successfully deployed an image to an 800 G2 to make sure it wasn’t the FOG server).

I am confused because on network boot, it grabs the right IP and even enters the FOG pxe interface, but if I do anything (deploy image, perform host registration) it goes to a black screen, displays about 20 ACPI Errors/Exceptions, initializes the random number generator, and says “Failed to get an IP via DHCP! Tried on interfaces(s):
Please check your network setup and try again!”

Then after a while it says “Press Enter to continue:” After pressing enter it displays this and restarts:

It shouldn’t be having DHCP issues, because before it even boots into FOG it displays the MAC address of the computer and displays the correct IP.

Any ideas as to what the issue might be? Are there certain NICs that just aren’t compatible with FOG?

Thanks for the reply! I’m going to get working on these.

Much of the software that is installed on our images requires licensing after deployment in order for it to work. I have made a couple of .bat scripts that will do that silently. Is there a way to have them automatically run after deployment of the image or upon logging for the first time?

I am looking for whatever solution people have found for this, whether there is some FOG feature that will do this or some workaround in Windows itself.

Thanks in advance!

I came in today and our fog machine is having all kinds of problems:

1. None of our department’s computers are able to boot into the fog network interface. On boot, the computers are able to find the right server IP. iPXE initialising devices… returns ok. tftp://FOGIP/default.ipxe… returns ok. But then, http://FOGIP/fog/service/ipxe/boot.php… returns “Input/output error (http://ipxe.org/1d0c6439)”. This prevents it from booting into fog.

2. When looking at what the fog machine actually displays on screen, it says:

[     8.967992] [drm:intel_set_pch_fifo_underrun_reporting [i915]] *ERROR* uncleared pch fifo underrun on pch transcoder A
[     8.968006] [drm:intel_pch_fifo_underrun_irq_handler [i915]] *ERROR* PCH transcoder A FIFO underrun

When I Googled this error, I found multiple places that said to run sudo fsck -f /dev/sda2, which is actually what we had to run a few days ago when fog went down, but this time it’s not working.

3. Maybe this is the root of the problem, but our web interface is also down.

Any help to fix these issues would be much appreciated! Thanks in advance!

@Wayne-Workman said:

@mecsr Can you run find / | grep /dev/.mntcheck please to locate where your image directory is?

Also, how big is the image you’re trying to capture, In terms of space used?

The image I’m trying to capture is at most, 300 GB. My image directory is just at /images.

@Tom-Elliott Here’s the output of df -i

Filesystem        Inodes   IUsed     IFree IUse% Mounted on
udev             2035414     522   2034892    1% /dev
tmpfs            2038213     691   2037522    1% /run
/dev/sda2      243122176 9464955 233657221    4% /
tmpfs            2038213       1   2038212    1% /dev/shm
tmpfs            2038213       7   2038206    1% /run/lock
tmpfs            2038213      15   2038198    1% /sys/fs/cgroup
/dev/sdb1      244195328 9354922 234840406    4% /mnt/mist
/dev/sda1              0       0         0     - /boot/efi
tmpfs            2038213       4   2038209    1% /run/user/1000

I haven’t had too much time to look at this since yesterday morning, so I haven’t really had an opportunity to run any of the suggested commands on the server yet - I will get around to that very soon. But I do think that there is an update on the situation - last night I started a computer imaging with one of the images that we already have had uploaded for quite some time. It’s worked before, but this time, on imaging, the image almost completes and then fails at the last moment. I have yet to see what the error message says.

I’m starting to to think that the suggestion of hard drive issues might be correct.

@Wayne-Workman Ok, here’s what I’ve got:

Filesystem      Size  Used Avail Use% Mounted on
udev            7.8G     0  7.8G   0% /dev
tmpfs           1.6G  9.4M  1.6G   1% /run
/dev/sda2       3.6T  3.0T  431G  88% /
tmpfs           7.8G     0  7.8G   0% /dev/shm
tmpfs           5.0M     0  5.0M   0% /run/lock
tmpfs           7.8G     0  7.8G   0% /sys/fs/cgroup
/dev/sdb1       3.6T  2.9T  566G  84% /mnt/mist
/dev/sda1       511M  3.4M  508M   1% /boot/efi
tmpfs           1.6G     0  1.6G   0% /run/user/1000

Mod edited to use code box.

@Wayne-Workman That sounds like it might be the case. The VM for the image I was trying to upload is now giving me this message, which seems to be in line with what you’re saying:

@Arrowhead-IT Install/upgrade actually just ended up telling me that an upgrade wasn’t needed, so that’s good. I now have access to Web management. I updated FOG two days ago, so I think I should be good in that respect.

I’ve been working on understanding FOG better since our FOG expert left here for better places, and it’s been going alright - a bit of a bumpy road, but not too bad. However, last night I tried to upload my first image (not the first for our FOG server, but for me, yes) using a Virtual Machine. It was going a bit slower than I expected, but seemed to be working.

And then I came in this morning and saw this:

I also noticed that I couldn’t log in to the FOG server for management or anything else. It appears that either this or something else turned off our server. I was able to turn it back on, but upon logging back in to the web management GUI, I get the following message:

I would appreciate some insight into what kind of course of action I should take, and what I can do to perhaps prevent this from happening again. Thank you!

I’ve actually had this exact same problem with a realtek USB ethernet adapter that comes with the Dell M3800.
Sadly, the only thing I have found that works is another adapter. And I’ve even tried compiling a custom fog kernel that has every available network driver enabled. Didn’t work so I went back to the ones that are provided.

One of the dell’s came with a different adapter that works though. It has a SMC LAN7500 chip, that one works and it works without a hitch.

It’s possible that hp might exchange just the adapter for one with that chip if they have one. Sadly the one that works doesn’t actually have any labels or part numbers on it so I can’t find it anywhere on the internet to show you where to get it.

So in the the latest version of for 1.02 I was able to successfully upload and deploy an image with multiple partitions including an extended linux partition that had a partition for root(/) /home and swap. It also has the first partition for windows and a partition after the extended partition for shared space.

I was testing an updated version of this image on the latest svn and ran into a problem. It looked like it uploaded fine and all the data was there. But when I download the image, to an identical computer model even, it gets to the first partition of the extended (sda5 within sda2) and it says that it isn’t a partclone image and just skips it. All the other partitions download perfectly fine. But linux is pretty useless without the root partition that has all the installation on it and what not.

For funzies I did try formatting the partition that doesn’t get downloaded and copying the conents of that partition from the original image, and I had to change the uuid of the partition to match the old one in the grub config, which did get the grub boot screen to pop up but it didn’t actually boot all the way into linux, it does still get to windows through grub, so that’s good.

So I’m wondering if anyone has any thoughts on this.

Oh, and I kept this updated one separate from the old extended partition image and that one still downloads fine, but I’d rather like to not have to update everything on windows and linux again you know.

Thanks,
-JJ

I’m also wondering if anyone has other helpful tricks they use on their fog server to make life easier.
For example another thing I do is make aliases for cd-ing into the fog dirs I use often, so I can just type cdfog to get to the webroot for example. I realize that you can also do this with global variables, but I’m a fan of aliases.

In case you don’t know, to make the file for your fog user
sudo nano /home/fog/.bash_aliases
Note that you’ll need to restart your terminal/login session for the aliases to take effect.

This is what my .bash_aliases looks like.

alias cdfog=‘cd /var/www/html/fog’
alias editDns=‘sudo nano /etc/dnsmasq.d/ltsp.conf’
alias editSmb=‘sudo nano /etc/samba/smb.cnf’
alias cdipxe=‘cd /var/www/html/fog/service/ipxe’
alias snapins=‘cd /opt/fog/snapins’

Hi friends,

So back in the times of fog 0.33b I posted a nifty script to help with automating the svn dev build update process. Since much has changed in the process since then I figure maybe someone else might find this rather helpful.
The script is attached and it is thoroughly commented. I’ll just post the contents without most the comments in the post too.

Some prereqs for this script

Already have an existing working install/configuration of FOG 1.0 or later

sudo apt-get install subversion
mkdir /home/fog/fogInstalls
sudo apt-get install dos2unix

A encrypted version of your sudo password, create one with this command

echo “your_super_secret_password” | openssl enc -des -a -e -pass pass:PASSWORD

Then just copy the output, including any “=” signs at the end of the string, into the later section of this script

Enter sudo mode and do some quick server maintenance update fun times

echo “$(echo yourOpensslEncryptedPassword | openssl enc -des -a -d -pass pass:PASSWORD)” | sudo -S apt-get update -y
sudo apt-get upgrade -y

Backup stuff

sudo cp /opt/fog/service/etc/config.php /home/fog/fogInstalls # config file
sudo cp /etc/exports /home/fog/fogInstalls/exports # Exports file
sudo cp /var/www/html/fog/service/ipxe/bzImage /home/fog/fogInstalls # 64 bit kernel
sudo cp /var/www/html/fog/service/ipxe/bzImage32 /home/fog/fogInstalls # 32 bit kernel
sudo cp /var/www/html/fog/service/ipxe/bg.png /home/fog/fogInstalls # custom background
mysqldump -u root --all-databases --events > all-Databases-backup.sql # database backup

Maintaining downloaded dev build folder structure

cd /home/fog/fogInstalls
time_stamp=$(date +%Y_%m_%d_%H_%M_%S)
mkdir ${time_stamp}
cd ${time_stamp}

SVN download

sudo svn checkout [url]https://svn.code.sf.net/p/freeghost/code/trunk[/url]
sudo chown -R fog /home/fog/fogInstalls
sudo chgrp -R fog /home/fog/fogInstalls
sudo chmod -R 777 /home/fog/fogInstalls

Update Fog

cd ${PWD}/trunk/bin
sudo yes | ${PWD}/installfog.sh
wget -O - --post-data=“confirm=yes” “[url]http://${FOGWEBIP}/commons/schemaupdater/index.php[/url]” 2>/dev/null

#Optional - Fix webroot
sudo rm -rf /var/www/html/fog
sudo mv /var/www/fog /var/www/html/fog

Restore backed up files

sudo dos2unix /home/fog/fogInstalls/.
sudo rm /opt/fog/service/etc/config.php
sudo cp /home/fog/fogInstalls/config.php /opt/fog/service/etc
sudo rm /home/fog/fogInstalls/bzImage && sudo rm /home/fog/fogInstalls/bzImage32
sudo cp /home/fog/fogInstalls/bzImage /var/www/html/fog/service/ipxe
sudo cp /home/fog/fogInstalls/bzImage32 /var/www/html/fog/service/ipxe
sudo cp /home/fog/fogInstalls/bg.png /var/www/html/fog/service/ipxe
sudo cp /tftpboot/undionly.kpxe /tftpboot/undionly.kpxe.original && mv /tftpboot/undionly.kpxe /tftpboot/undionly.0
sudo cp /tftpboot/undionly.0 /var/www/html/fog/service/ipxe/undionly.0

Make sure Permissions are right on webroot and nfs images path

sudo chown -R fog /var/www/html/fog
sudo chgrp -R fog /var/www/html/fog
sudo chmod -R 775 /var/www/html/fog
sudo chown -R fog /images
sudo chgrp -R fog /images
sudo chmod -R 775 /images

Restart/Refresh server services

sudo /etc/init.d/apache2 restart
sudo service mysql restart

Alternatively, you could also just have it restart the server at this point with

sudo shutdown -r now

[url=“/_imported_xf_attachments/1/1757_fog_SVN_updater.zip?:”]fog_SVN_updater.zip[/url]