Posts made by george1421

There is… (said with a zest of a 5 year old girl)

http://en.community.dell.com/techcenter/enterprise-client/w/wiki/7532.dell-command-configure

cctk for linux. So in theory we could use a FOG custom task to configure the bios. But then we would need to reboot after the configuration for the firmware to reconfigure for the new settings… that may add more complications then fix. But it IS interesting for post imaging ensuring that the bios settings are right.

@bedcruncher It should image ok without raid mode enabled.

We use a boot usb flash drive to install a consistent bios in a winpe environment. During image we use cctk in the winpe environment to adjust the bios settings. This way we have a consistent configuration when the device leaves our buildup area.

let me check on something…

@bedcruncher Sweet. lsblk now sees the nvme disk.

The skylake kabylake was an after thought as I was looking up the specs on the 7720.

Now that lsblk can see the disk give it a shot again with imaging.

Can you confirm of this has a kaby lake chipset or sky lake?

@Developers it looks like the release date on the 7720 was in May 2017.

@bedcruncher said in Registration Issues with Dell Precision 7720 and M.2 Sata:

RAID bus controller [0104]: Intel Corporation SATA Controller [RAID mode] [8086:2822] (rev 31)

Well this is a good sign the rest are bad news. But lets work with this.

Is this a uefi system? I see that raid is enabled on this sata controller. If uefi = yes take the system out of raid mode and hopefully the lsblk command will show us the block devices. When you switch out of raid mode the line I commented above will change from [raid mode] to something else.

Here you go.

1. Login to the database with this command
   mysql -u root fog
2. Use this command to list all of your images. You will need the exact image name for the next command.
   select imageID,imageName,imageProtect from images;
3. To change the setting run this command. imageProtect=0 to turn it off imageProtect=1 to enable, <exact_imageName> is taken from the select statment.
   update images set imageProtect=0 where imageName="<exact_imageName>";

@jim-graczyk Just so you know the developers are on an extended holiday right now so their response is going to be a bit slow right now.

Let me see if I can work up the sql commands you need. Its pretty easy I just need to find the right fields.

This is a really new system, its highly possible that linux doesn’t support it just yet.

Can you try this, manually register this device. Then schedule a manual deployment. It doesn’t matter what you deploy, the debug deployment will allow us to get access to the FOS console. Once at the command prompt on the target computer key in the following commands.

1. lspci -nn
2. lsblk
3. df -h

Now I’ll give you a tip.

Before you do the above.

1. At the linux prompt give root a password, just key in passwd and give it any password hello will work.
2. key in the following command to get the ip address of the FOS (target) computer. ip addr show
3. With those two bits of information you can use putty to connect to the target computer over the network. With putty its much easier to copy and paste screen shots then on the FOS console directly.

Ubuntu 16.04 configuration.

WARNING: These instructions to not work as of now. Everything is configured correctly but I can't seem to login to the FOG site. There has to be something I'm missing. This configuration is close, very close since the login page appears.

1. Change to the root user
   sudo su -
2. Install the apache fast cgi module
   apt-get -y install libapache2-mod-fastcgi
3. Enable the apache fast cgi module
   a2enmod proxy_fcgi
4. Edit the php configuration file.
   vi /etc/apache2/mods-available/php7.1.conf
5. Comment out the following line
   #SetHandler application/x-httpd-php
6. Insert the following line just below the commented out line
   SetHandler "proxy:fcgi://127.0.0.1:9000"
7. Save and exit the php-fpm.conf file
8. Change to the php-fpm pool directory
   cd /etc/php/7.1/fpm/pool.d
9. Remove the default www.conf in the pool.d directory (this file is created by the fog installer and not used in this setup)
   rm /etc/php/7.1/fpm/pool.d/www.conf
10. Create a new file called fog.conf
    vi /etc/php/7.1/fpm/pool.d/fog.conf
11. Paste in the following

[fog]
user = www-data
group = www-data

listen = 127.0.0.1:9000

;listen.owner = www-data
;listen.group = www-data
;listen.mode = 0660
;listen.acl_users = apache
;listen.acl_groups =

listen.allowed_clients = 127.0.0.1

pm = dynamic
pm.max_children = 50
pm.start_servers = 5
pm.min_spare_servers = 5
pm.max_spare_servers = 35
;pm.process_idle_timeout = 10s;
pm.max_requests = 500
;pm.status_path = /status
;ping.path = /ping
;ping.response = pong
 
access.log = /var/log/php-fpm/$pool.access.log
;access.format = "%R - %u %t \"%m %r%Q%q\" %s %f %{mili}d %{kilo}M %C%%"
slowlog = /var/log/php-fpm/$pool-slow.log

;php_admin_value[sendmail_path] = /usr/sbin/sendmail -t -i -f www@my.domain.com
;php_flag[display_errors] = off
php_admin_value[error_log] = /var/log/php-fpm/fog-error.log
php_admin_flag[log_errors] = on
;php_admin_value[memory_limit] = 128M
php_value[session.save_handler] = files
php_value[session.save_path]    = /var/lib/php/sessions

; we will use these settings when memcache (d) is configured.
;php_value[session.save_handler] = memcached
;php_value[session.save_path] = "127.0.0.1:11211"

php_value[soap.wsdl_cache_dir]  = /var/lib/php/wsdlcache

12. Save and exit fog.conf
13. Create the php-fpm log directory
    mkdir /var/log/php-fpm
14. Create a php information page in the fog web root
    vi /var/www/html/fog/info.php
15. Paste in the following

<html>
<body>

<?php
 phpinfo();
?>

</body>
</html>

16. Save and exit info.php page.
17. Change the owner of that page to apache
    chown www-data:www-data /var/www/html/fog/info.php
18. From a web browser call the info page with http://<fog_server_ip>/fog/info.php. The page should look similar to below
    <insert_info_page_image>
19. Now lets restart both apache2 and php-fpm
    systemctl restart php7.1-fpm apache2
20. Give it a few seconds for both services to initialize.
21. Now call the same info.php page. The page should look similar to below. NOTE: that the Server API variable is now ‘FPM/FastCGI’
    <insert_info_page_image>
22. Now access the FOG management page console
23. Access a few of the managment pages, you will note that after a few page clicks the pages will respond faster to your page selections.
24. To confirm that php-fpm is working you can inspect the logs files being created in /var/log/php-fpm directory. The php errors will now be listed there instead of the default apache error log (because php-fpm is now handling the php code).
25. You may remove the info.php page in the fog web root directory. Its no longer needed.
26. Done

If you have time to help us test this new configuration, please let me know and I will provide instructions.

@sebastian-roth Isn’t that iPXE menu line managed by these WebUI settings?

1. FOG Configuration->FOG System Settings->FOG Boot Settings->FOG_PXE_MENU_HIDDEN
2. FOG Configuration->FOG System Settings->FOG Boot Settings->FOG_PXE_HIDDENMENU_TIMEOUT

@ljedi said in Fog Exit Windows 10:

I have not yet created an image for this version yet. 1703 is a clean install and not an upgrade from 1607

This confuses me a bit. Just so I understand what I think you said and didn’t say. You created a fresh install of 1703 that works. But when you add pxe booting first in the boot order, iPXE is not exiting correctly to boot windows? But when you have the hard drive first in the boot order the pc boots correctly??

I just trying to understand the flow here. I want to see if I can duplicate that in our lab, I just need to know how you did what you did.

@ljedi What hardware are you deploying to?

Did you use single disk resizable?

Did you upgrade windows to get to 1703 or did you install it from a newly created reference image?

There has to be something in common here.

Simple answer is don’t use a slash [ / ] or other special characters in the image name. The slash character is allowed in the file name to setup special file grouping on the storage node.

For example if your image name is “town1/image31” the fog server will put the image name [image31] inside a subfolder [/images/town1] on the storage node. This is done to allow image grouping on the fog server. If this isn’t what you intended to do, take the slash out of the image name.

if 10.0.88.42 is your FOG server then 10.2.102.49 is a defined storage node. That is the only place where that IP can come from.

@aksl For uefi, rEFInd is really your best option. Make sure you have the exit mode for the specific computer and uefi set to rEFInd and then lets tell refind to look in more places for a boot drive.

1. On the fog server from a linux console prompt edit the following file: /var/www/html/fog/service/ipxe/refind.conf
2. Search for the line scanfor internal and comment it out to look like this

#scanfor internal

3. Search for the line #scanfor internal,hdbios,external,biosexternal and uncomment it to look like this

scanfor internal,hdbios,external,biosexternal

4. Save and exit the editor.
5. Now pxe boot the uefi system and see if that mode finds the hard drive.

rEFInd “should” find the boot disk if it’s installed.

@richard-wise Right that is usually the case when windows is involved. There are a couple of ways to side step the issue.

1. Not use windows at all.
2. Use Notepad++ on windows
3. Once the file gets to linux unaltered, use the linux command dos2unix to scrub out any remaining Windows bits left behind.

With that said, well done on the debugging!!.

My initial reaction is that you have 2 dhcp servers on your network.

If your fog server, dhcp server, and pxe booting client are on the same subnet, we can use the FOG server to eavesdrop on the pxe booting process to tell us what is going on.

Please follow these instructions: https://forums.fogproject.org/topic/9673/when-dhcp-pxe-booting-process-goes-bad-and-you-have-no-clue

Upload the pcap to a google drive or drop box and post the link here. If you don’t want to post the link in a public forum, send me an IM and we’ll take a look at it. The captured file will tell us the actors in your pxe booting play.

@natsufiree Here are the instructions for building a usb boot into FOG. https://forums.fogproject.org/topic/7727/building-usb-booting-fos-image

There is a sequence you must execute to image with usb.
You must schedule a capture or deployment task before you usb boot your target computer or you will get an error about an invalid task type “Null”

You can register a computer by usb booting and picking registration. But you must remember to when capturing or deploying images to schedule them on the fog web gui first then usb boot the target computer.

@natsufiree I wrote a tutorial on how to compile dnsmasq 2.76. If you need to support both bios (legacy mode) computers and uefi computers then you will need to use dnsmasq 2.76 or newer. Not all linux distributions have an updated version of dnsmasq. If your linux distribution does then the setup is pretty easy. Just use the ltsp.conf config file I have posted in the thread. Update the <fog_server_IP> directives with the IP address of your fog server and restart the dnsmasq service.

If your linux distribution does not have version 2.76 or newer then you must follow the instructions in the tutorial on compiling your own version of dnsmasq. The instructions I created for ubuntu, but should work for debian.

https://forums.fogproject.org/topic/8725/compiling-dnsmasq-2-76-if-you-need-uefi-support/6

Once you have the current version of dnsmasq installed then update the config file and start the dnsmasq service.