Posts made by Eli Kelly

Replaced fogservice.zip from a backup prior to my last update. Domain joins are working again.

Haven’t tried this but Tom posted this in the development thread for reverting to earlier version.
[URL=“http://fogproject.org/forum/threads/latest-development-fog.10556/page-22#post-42431”]Latest Development FOG[/URL]

I’m having an issue with AD joins that might be related. My host names are not changing with service starts but do change after image downloads. I’m going to try recompiling hostnamechanger today. I’ll update my thread in the Windows forum as I go.

Ok some progress.

Checkboxes: Appears to be an issue with my Chrome browser. Checkboxes work normally in IE.

Log: Not showing activity related to AD because of Windows Firewall settings.

AD Joins failing: Now see a “padding is invalid message”.

Do I need to recompile the hostnamechanger module and encrypt the AD password again after updating to svn 2868 from svn 1858?

Current SVN 2868. This was working in previous versions.

Join Domain after image checkbox does not stay checked after updating hosts. None of the previously AD registered hosts have checkbox checked any more.

The fog log on the clients does not show any AD activity but that is to be expected based on the checkbox status.

For now I will do some housekeeping. Some of the images can be deleted and some of them can be stored offline.

Thanks!

Searches are relatively responsive but “list all images” takes about 5 minutes and hangs the browser tab until it is done. This isn’t critical but it would be desirable for this to work.

Current: svn 2868 on CentOS 6.6

First noticed this after I updated from svn 1858 to what I think was svn 2759 (or right around).

I bring the following up in case there is a helpful clue:

-No errors found in logs

-FOG_MEMORY_LIMIT=1024 made no detectable difference

-Storage nodes work but dashboard disk info fails to connect to them. Not trying to solve this unless it is a symptom of a more significant issue. Credentials stored work to connect to database through SSH. I did have to change the default image directory to /home/fog. This issue predates image list issue

-I have 43 images taking up just under 1TB.

VMWare is working again for me. There was one important step that I was missing. It was expressed clearly by the OP. There was even a screenshot. I have even done it hundreds of times correctly in the past.

It simply turned out that I have been forgetting to tick the “Generalize” box! Sysprep wasn’t processing anything from the Specialize phase of my unattend.

I have only had luck using Visual Studio 2008. None of the Express versions have worked for me. None of the newer standard versions seem to work either.

[quote=“jmeyer, post: 32641, member: 6537”]Have you reboot your VM before and after sysprep to see if this happen ?[/quote]

VM reboot after sysprep completes normally.

Update: Gave up on VMWare for this. I switched to VirtualBox and the process is working perfectly except that I get prompts for a hostname when windows starts up after deploying to a physical machine. Actually it initially prompted for the hostname and timezone. My unattend (which works correctly with my old VMWare snapshots as well as with MDT deployments) originally declared the Timezone and ComputerName in the specialize pass. Adding a Timezone value in the OOBE pass fixed the prompt for time zone but adding the ComputerName tag there doesn’t make a difference. I have tried leaving the ComputerName tag out of both sections and not surprisingly the hostname prompt still appears when the physical machine boots after deployment.

Admittedly my unattend file is somewhat complex as it does a few things such as configure RDP, enable Administrator, disable WER. Tomorrow I will try a very basic unattend file but thought I would see if any ideas came up that would save me the troubleshooting time.

Thanks

Eli

This is working pretty well for me. Much credit to others on this forum who came up with most of this…

[CODE]login && goto validate || goto return

:validate
iseq ${password} yourpasswordhere && goto MENU || goto return

:MENU
menu
item --gap – ---------------- iPXE boot menu ----------------
item ipxedemo ipxe online boot demo
item shell ipxe shell
item pe86 Generic WinPE x86
item lt86 MDT Lite Touch x86
item lt64 MDT Lite Touch x64
item ghost Ghost Boot
item dban Derek’s Boot and Nuke
item return return to previous menu
item hostinfo details about this computer
choose --default return --timeout 5000 target && goto ${target}

:ipxedemo
chain http://boot.ipxe.org/demo/boot.php ||
goto MENU

:shell
shell ||
goto MENU

:pe86
initrd http://${fog-ip}/ISO/pe86.iso
chain memdisk iso raw ||
goto MENU

:lt86
initrd http://${fog-ip}/ISO/lt86.iso
chain memdisk iso raw ||
goto MENU

:lt64
initrd http://${fog-ip}/ISO/lt64.iso
chain memdisk iso raw ||
goto MENU

:ghost
initrd http://${fog-ip}/ISO/ghost.iso
chain memdisk iso raw ||
goto MENU

:dban
initrd http://${fog-ip}/ISO/dban.iso
chain memdisk iso raw ||
goto MENU

:return
chain http://${fog-ip}/${fog-webroot}/service/ipxe/boot.php?mac=${net0/mac} ||
prompt
goto MENU

:hostinfo
echo This computer : ||
echo MAC address…${net0/mac} ||
echo IP address…${ip} ||
echo Netmask…${netmask} ||
echo Serial…${serial} ||
echo Asset number…${asset} ||
echo Manufacturer…${manufacturer} ||
echo Product…${product} ||
echo BIOS platform…${platform} ||
echo ||
echo press any key to return to Menu ||
prompt
goto MENU

autoboot
item return return to previous menu
choose --default return --timeout 5000 target && goto ${target}

:ipxedemo
chain http://boot.ipxe.org/demo/boot.php ||
goto MENU

:shell
shell ||
goto MENU

:return
chain http://${fog-ip}/${fog-webroot}/service/ipxe/boot.php?mac=${net0/mac} ||
prompt
goto MENU

autoboot[/CODE]

Yes at first boot and before self optimization. Just after “Starting Windows” is displayed just as the logo is starting to show up the computer reboots.

If I try to boot to safe mode it restarts right after Loading CLASSPNP.SYS

That’s right. It happens on deployment to a physical machine.

Thanks!

Your vmx file is substantially similar to my working ones with the exception that mine have:

[CODE]virtualHW.version = “8”[/CODE]

An interesting observation is that all of my newer snapshots that are not working have an added line at the bottom of the vmx file

[CODE]monitor.phys_bits_used = “40”[/CODE] I’m not sure if that is significant at all but it is my best lead right now.

[quote=“jmeyer, post: 32396, member: 6537”]Why you don’t build a clean Windows in a VM and use snapshot to revert modification if needed then use SAD in sysprep ? This way, if you have special soft to add to an image, just install them, run sysprep, upload image then revert modification to the VM.
We have HP, DELL, LENOVO, ACER and few other brand of computer and we use ONLY ONE image with sysprep…
Using an image from a computer means that you backup all the specification of the computer for nothing (driver and software).
Maybe you don’t even handle temp files…[/quote]

That is exactly what I am doing and it has been working beautifully until recently. I have a full set of snapshots from every stage of preparing my clean image. From time to time I like to patch the OS, take a new snapshot and then update the stock image on the FOG server. For some reason any new snapshots break this process.

I did try your thought about setting scsi0.present = “FALSE” in the VXD (edited:meant VMX) file. Unfortunately that doesn’t seem to make a difference.

I do set the drives as IDE but I have not edited the VMX file to disable SCSI. It is a good thought that I will try to tomorrow and report results.

P2V is shorthand for the process of capturing a physical machine to a virtual machine. This is not what I’m trying to do but my point was there is plenty of documentation of that process. V2P is essentially what we are doing when we are preparing virtual machine images with VMWare Workstation for deployment to physical machine with FOG.

[quote=“jmeyer, post: 32276, member: 6537”]
There is no problem to build this image under VMware Workstation 10.
Mine was made in 2013 and i keep updating it since.[/quote]

Thanks for confirming this.

I’m thinking this could be a problem with my configuration of VMWare Workstation configuration or the hardware/OS it is running on. I would question the settings I’m using when creating the VM guest images but older snapshots work fine on VMs created in November 2013 or before. I tried starting from scratch with a fresh install of Workstation 9 and still end up with unbootable deployments. The file system transfers fine and partitioning looks good but still won’t boot.

Are there any recent Windows updates known to break VMWare V2P? My workstation 10 is Windows 8 x64(not 8.1) on an HP 6200 Pro while Workstation 9 is Windows 7 x64 on an HP 6000 Pro. P2V is well documented but the reverse not so much.

We are an HP shop with 6000/6200/6300 pro and will be adding 600 G1 shortly. Aside from the 6000 not wanting to step through the boot order when a PXE server is not found these machines have worked well for us until now.

I’m quickly running out of ideas so any thoughts are welcome. Reports of success with similar hardware will also help me feel more hopeful.

Thanks

Eli

Sooo… I have used the instructions from the initial post with great success many times though apparently I haven’t done so since November. My VMWare Workstation snapshots from November work fine but if I upload an image after taking a new snapshot and then deploy it to any machine I get a blue screen stop error and the computer boot loops at the “Starting Windows” screen.

Unfortunately I can’t recall exactly what changed around that time. I first noticed trouble when I started to experiment creating 64bit images last winter. I mistakenly thought it was an issue with my x64 imaging process which wasn’t critical so I didn’t think much of it. As I am trying to update my 32 bit image from last November I finally realized that the issue extends to all VMWare Workstation images EXCEPT those with snapshots that go back to November and before.

I did upgrade from VMWare Workstation 9 to VMWare workstation 10. I thought that was in September or October but it might have been around the time of the problem. My next plan is to roll back to VMWare Workstation 9.

Anyone experienced anything like this and found a solution?

Rhythmtone- What worked for me was “root” for the user values and “localhost” for the host values. I didn’t set my SQL password so I set it to an empty string (“”). My guess is that it needs to be the one you picked for the SQL root user which is probably the one you entered in config.php

Brilliant!

Editing the entries in Config.class.php fixed it. I also corrected the entries in .fogsettings. I ran the installer again. The database update ran and the logs look clean.

I was not able to determine how to apply the information from the blog link to my situation. At this point everything including MySQL appears to be working as expected. It is not clear that creating a symbolic link to a MySQL socket file will solve anything for me. The instructions are written for Ubuntu but if there is a pressing reason you advise this I will work through how to translate them for use with CentOS.

Thanks

Eli

Just upgraded to 1.1.1 svn 1858 and am now having the same problem as OP.

I didn’t ever set the MySQL root password but /var/log/httpd/error_log shows this:
[CODE][Tue Jun 17 08:17:24 2014] [error] [client 10.1.10.192] PHP Warning: mysqli::mysqli(): (28000/1045): Access denied for user ‘root’@‘localhost’ (using password: YES) in /var/www/html/fog/lib/db/MySQL.class.php on line 64[/CODE]
followed by a number of lines of PHP warnings … Couldn’t fetch mysqli in …

There is no config.php in /var/www/fog/commons (/var/www/html/fog/commons for my installation). In addition the ipxe boot menu no longer shows up when clients boot.

FOG 1.1.1 svn 1858
CentOS 6.5
foginstall.log indicates “* Starting FOG Multicast Management Server…Failed!”