Latest FOG 0.33b
-
r1008 contained 64 bit and 32 bit binaries.
r1009 updates the fog.mk script to know which binaries to copy.
-
Hello!
I have updated fog 0.33b to the latest revision on my test server, then I started to test a few things. I tried to upload the image from the client to the server and got this error: “FOG requires your Windows XP’s partition start sector to be 63 but is 1”. I checked the start sector from a linux live cd and it shows me 63, tested with vmware and winxp, same problem. To be sure that this problem has nothing to do with the update, I did a complete reinstall of ubuntu 12.04.03 and from the fog server, the error still appears.
If you need more informations or want me to test me something, write it down here and I will test it.
-
Can you run in upload debug and run this line from the FOG init.gz file?
[code]fdisk -l | grep /dev/sda2 | awk ‘{print $3}’ [/code]It sounds like the -a opt part of parted doesn’t like doing this for Windows XP partitions.
[COLOR=#000000]fdisk -l | grep $part | awk '{print $3}'fdisk -l | grep $part | awk ‘{print $3}’[/COLOR] -
I think this is because it’s reading /dev/sda1 as well, and of course the start sector for the first partition is going to be 1.
I’ll see if I can figure out another method.
-
I started from debug mode, and typed in fdisk -l, it shows me start sector 1. The disk has only 1 partition, there is only /dev/sda1.
-
[quote=“Albatros, post: 20692, member: 16710”]I started from debug mode, and typed in fdisk -l, it shows me start sector 1. The disk has only 1 partition, there is only /dev/sda1.[/quote]
I get exactly the same issue using REV 1009.
Also I’m unable to upload a Windows 7 Image, it flashes on the blue upload screen then restarts.
Deploying ‘looks’ like its working, until it restarts then just crashes before the windows logo (note its not the image as using the exact same image of previous revisions 999 and below it worked fine)
Maybe the problem is linked in someway? -
[quote=“Albatros, post: 20692, member: 16710”]I started from debug mode, and typed in fdisk -l, it shows me start sector 1. The disk has only 1 partition, there is only /dev/sda1.[/quote]
It sounds like the issue for you is related to it creating the partition. To my knowledge, XP only uses the one partition. I correct the start sector, if it’s blank or 1 to set it 63.
[quote=“Joe Butler, post: 20698, member: 3637”]I get exactly the same issue using REV 1009.
Also I’m unable to upload a Windows 7 Image, it flashes on the blue upload screen then restarts.
Deploying ‘looks’ like its working, until it restarts then just crashes before the windows logo (note its not the image as using the exact same image of previous revisions 999 and below it worked fine)
Maybe the problem is linked in someway?[/quote]Joe, can you start yours as debug (Download or Upload?) and run the command:
[code]parted -s /dev/sda2 -a opt u kB mkpart primary ntfs 105906176s 100%[/code]Tell me what it spits out at you.
It sounds for your particular issue, is it’s not recreating the second partition and I need to find out why. I have two commands:
[code]parted -s ${part} -a opt u kB mkpart primary ntfs ${defaultpart2start}B ${layoutSize}
parted -s ${part} -a opt u kB mkpart primary ntfs ${defaultpart2start}B ${diskSize}kB[/code]Back to back. One way or another, it’s supposed to create the second partition. Some drives don’t mind 100%, others need the size specified directly. I don’t know why. On an 80GB drive, it wants percents. On a 50GB drive, it wants the actual disk size. It’s dependent upon the drive itself.
${layoutSize} is set to 100%
${diskSize} reads the max size the drive can contain. -
r1010 released, hopefully to address start sector 1 for windows xp images. Please test. I don’t have XP to play around with.
Thank you Albatros for the info.
Joe, I’ll try to come up with a solution for your issue as well, but on the systems I have to test with everything seems to work. So I’ll need your help if at all possible.
Thank you,
-
I upgraded the fog server, tried to upload an image from the client to the server and get this error:
[IMG]http://i.imgur.com/a6XlaEh.png[/IMG]Tried this wth other computers, same problem. I am not sure what’s now wrong there, sorry Tom.
-
First of all, thank you! You’re doing a great job
I want to ask for UEFI support, when I try to register a PC it throws an “unsupported BIOS version” error and later a Kernel Panic.
There are any news about of that?
Thanks!
Joan
P.D: I was writing this message at the same time as Albatros
-
The issue you’re seeing is the init.gz file you’re using, seems to not have lzma support. So it can’t mount the rootfs which will, of course, cause a kernel panic. You’re sure you’re on revision 1010?
I don’t have any UEFI systems to test with, but could you provide more information about the kernel panic? Ultimately, I don’t care so much about the error. That’s fairly typical as the kernel has many bios options available to it, but not all will be supported.
Thank you,
-
Well, I thought the error was caused by UEFI but it is the same kernel panic that Albastro has posted.
I’m running the r1010 revision.
Thank you!
-
I believe I’ve found out my idiotness of 1010. I gunzipped where I shoulda lzma’d. Right now I’m rebuilding init.gz so it should be fixed shortly, sorry about all of that.
-
r1011 released to address gzip/lzma issue. Also adds resize2fs program with Gilou’s suggestion.
Rsync is also included.
-
r1012 released to fix the FOG script again. FOG Script did not contain the sector 1 fix. Sorry about that.
-
Ty, Tom.
I tested it and got syntax error on line 837. Had to change
[CODE]if [ “$partitionStart” == “” || “$partitionStart” == “1” ]; then[/CODE]
to
[CODE]if [ “$partitionStart” == “” ] || [ “$partitionStart” == “1” ]; then[/CODE]After I fixed that , I get this error
[IMG]http://i.imgur.com/ifmTWSi.png[/IMG]Sorry that I have now the next error.
Ty very much for your great work, will try to test snapins tomorrw and will will report back. -
You don’t have to be sorry for the error. It looks like it didn’t create the partition as required. Probably because it’s not actually resizing the partition in the code base. Rebuilding as we speak.
-
r1013 is out. Please try this and let me know.
Thanks,
-
[quote=“Tom Elliott, post: 20740, member: 7271”]r1013 is out. Please try this and let me know.
Thanks,[/quote]
I can not test this now, will test it tomorrow. Thanks for your quick response.
Edit:
I looked into the current revision. You didn’t fixed the syntax error, here:
[CODE]if [ “$partitionStart” == “” || “$partitionStart” == “1” ]; then[/CODE]Is in file src/buildroot/package/fog/scripts/bin/fog on line 837. Please fix.
Ty,
Albatros -
Yes I did fix that, but not quite on the src/buildroot side yet. I just created a couple of patches and updated a few of the FOG components. Once tested and working, I’ll update the src directories with all the proper information.
In the init.gz file, however, this fix is already in place. If it didn’t take I don’t know why not, but the build I’m setting up right now will, for sure, have this in place.
