Latest FOG 0.33b
-
Hi Tom,
are you using efibootmgr in FOG?
I was testing using partclone manually restoring an image and it wouldn’t boot. I was able to boot a linux live cd, and using efibootmgr repair the boot menu manually.
As far as I am aware, the only way to fix non bootable UEFI from windows tools is BCEDIT, however seeing as you boot a linux kernel to apply the image using partclone, maye you can script efibootmgr afterwards to repair the UEFI on the system and make it bootable.
Or is it a different issue entirely with FOG?
efibootmgr should be scriptable
usage: efibootmgr [options]
-a | --active sets bootnum active
-A | --inactive sets bootnum inactive
-b | --bootnum XXXX modify BootXXXX (hex)
-B | --delete-bootnum delete bootnum (hex)
-c | --create create new variable bootnum and add to bootorder
-d | --disk disk (defaults to /dev/sda) containing loader
-e | --edd [1|3|-1] force EDD 1.0 or 3.0 creation variables, or guess
-E | --device num EDD 1.0 device number (defaults to 0x80)
-g | --gpt force disk w/ invalid PMBR to be treated as GPT
-H | --acpi_hid XXXX set the ACPI HID (used with -i)
-i | --iface name create a netboot entry for the named interface
-l | --loader name (defaults to \elilo.efi)
-L | --label label Boot manager display label (defaults to “Linux”)
-n | --bootnext XXXX set BootNext to XXXX (hex)
-N | --delete-bootnext delete BootNext
-o | --bootorder XXXX,YYYY,ZZZZ,… explicitly set BootOrder (hex)
-O | --delete-bootorder delete BootOrder
-p | --part part (defaults to 1) containing loader
-q | --quiet be quiet
–test filename don’t write to NVRAM, write to filename
-t | --timeout seconds Boot manager timeout
-T | --delete-timeout delete Timeout value
-u | --unicode | --UCS-2 pass extra args as UCS-2 (default is ASCII)
-U | --acpi_uid XXXX set the ACPI UID (used with -i)
-v | --verbose print additional information
-V | --version return version and exit
-w | --write-signature write unique sig to MBR if neededTypical usage:
- Root can use it to display the current Boot Manager settings.
[root@localhost ~]# efibootmgr
BootCurrent: 0004
BootNext: 0003
BootOrder: 0004,0000,0001,0002,0003
Timeout: 30 seconds
Boot0000* Diskette Drive(device:0)
Boot0001* CD-ROM Drive(device:FF)
Boot0002* Hard Drive(Device:80)/HD(Part1,Sig00112233)
Boot0003* PXE Boot: MAC(00D0B7C15D91)
Boot0004* Linux
This shows:
BootCurrent - the boot entry used to start the currently running
system.BootOrder - the boot order as would appear in the boot manager. The
boot manager tries to boot the first active entry on this list. If
unsuccessful, it tries the next entry, and so on.BootNext - the boot entry which is scheduled to be run on next boot.
This superceeds BootOrder for one boot only, and is deleted by the
boot manager after first use. This allows you to change the next boot
behavior without changing BootOrder.Timeout - the time in seconds between when the boot manager appears
on the screen until when it automatically chooses the startup value
from BootNext or BootOrder.Five boot entries (0000 - 0004), the active/inactive flag (* means
active), and the name displayed on the screen.-
An OS installer would call ‘efibootmgr -c’. This assumes that
/boot/efi is your EFI System Partition, and is mounted at /dev/sda1.
This creates a new boot option, called “Linux”, and puts it at the top
of the boot order list. Options may be passed to modify the
default behavior. The default OS Loader is elilo.efi. -
A system administrator wants to change the boot order. She would
call ‘efibootmgr -o 3,4’ to specify PXE boot first, then Linux
boot. -
A system administrator wants to change the boot order for the next
boot only. She would call ‘efibootmgr -n 4’ to specify that the
Linux entry be taken on next boot. -
A system administrator wants to delete the Linux boot option from
the menu. ‘efibootmgr -b 4 -B’ deletes entry 4 and removes it
from BootOrder. -
A system administrator wants to create a boot option to network
boot (PXE). Unfortunately, this requires knowing a little more
information about your system than can be easily found by
efibootmgr, so you’ve got to pass additional information - the ACPI
HID and UID values. These can generally be found by using the EFI
Boot Manager (in the EFI environment) to create a network boot
entry, then using efibootmgr to print it verbosely. Here’s one example:Boot003* Acpi(PNP0A03,0)/PCI(5|0)/Mac(00D0B7F9F510)
ACPI(a0341d0,0)PCI(0,5)MAC(00d0b7f9f510,0)In this case, the ACPI HID is “0A0341d0” and the UID is “0”.
For the zx2000 gigE, the HID is “222F” and the UID is “500”.
For the rx2000 gigE, the HID is “0002” and the UID is “100”.
You create the boot entry with:
‘efibootmgr -c -i eth0 -H 222F -U 500 -L netboot’
- Root can use it to display the current Boot Manager settings.
-
r1308 released.
Rewrote the task finding data on Dashboard page. Also removes the setMessage on all hooking pages. May not be nice for things within true table format (header and data) but it should stop reporting messages like (75 Hosts Found) at the top of the page.
-
efibootmgr is built into the init.xz, but it can’t be used unless you’re booted into a UEFI environment (UEFI PXE BOOT).
-
[quote=“warp, post: 24094, member: 22860”]The all this is to stabilize something overcharged as changing constantly as the PXE boot, which worked very well, for having changed?
You still need to know that the 0.33 version is expected for almost 2 years![/quote]FOG is not a paid service and you still need to know that nothing should be “expected” the devs spend their own personal time working very hard to improve and stabilize FOG and do not make any money from doing so. Tom has worked extremely hard and is adding excellent addtions to FOG and the reason for the changes are either to move away from products either outdated or no longer being actively developed i.e. partimage or their are big benefits and potential in doing so i.e. ipxe
-
[quote=“warp, post: 24094, member: 22860”]The all this is to stabilize something overcharged as changing constantly as the PXE boot, which worked very well, for having changed?
You still need to know that the 0.33 version is expected for almost 2 years![/quote]@Warp Before Tom came on board 0.33 development had ground to a ground to a halt. The project has come far under his wing. If he never took over as main developer. 0.33 probably would never ever have been realised. You should thank you lucky stars we are at this stage now.
FOG could have died like many over good open source projects have.
-
r1309 released.
Fixes the selector for the storage node on the dashboard page. Basically places it at the bottom of the field.
Removes the PXEFile class files as we no longer need them.
Minor updates to other files. Still working on hooking elements. As you all know, this may take a little bit.
-
r1310 released.
Finishes adding the hooking stuff, for the most part, in HostManagementPage. All things are pulled from classes as well now, rather than actually calling the SQL statements to do the work.
-
Whatever has changed between 1303 and 1310 has broken my imaging process.
When I pushed the same process yesterday, everything was working fine. Now, it gets to the step right before part clone launches to copy the image, and the machine just reboots, cycles back, and does the same thing over again.
-
Found it…
This is the last 2 lines before reboot:
*Checking Mounted File System…/bin/fog.download: line 394: syntax error near unexpected token ‘fi’
/bin/fog.download: line 394: ’ fi’ -
r1311 released.
Should fix the syntax error you’re getting there.
Also, I’ve changed the undionly.kpxe file to the one I’m personally using and haven’t had any issues with. Hopefully this will fix the issues you were reporting.
-
Yes, that fixed it. Everything is back in is zero-touch glory!
-
r1312 released.
Just updates undionly.kpxe and snponly.efi files to the latest ipxe git repository version.
-
r1313 released.
Just a simple release to fix the Dashboard and allow it to display the tasks from today.
-
r1314 and 1315 released.
r1314 Add’s hooking through more files. Starts removing the excessive <?php ?> calls which should help improve performance as you’re not calling the pre-processing elements multitudes of times throughout each of the files. Adds a GMT timezone setting for those who have not set the date.timezone property in the php.ini file. This way, all still displays appropriately.
r1315 fixes an issue in the HostManagementPage from the Location. Also adds more removals of <?php ?> calls. Hooking is complete through ReportManagementPage if reading alphabetically. On to the S’s
Thank you,
-
r1316 released.
Nearly all pages are complete with Hooking and excessive php calls. Only ones left to complete are StorageManagementPage and TaskManagementPage. Will be working on, just wanted to get the info out there.
Thank you,
-
r1317 released.
Fixes display issue on ServerInfo.class.php where the ethernet interfaces wouldn’t show the proper data. Also adds element to tell the user which interface is being displayed. (Fully hooked of course)
-
r1318 released.
Fixes an issue in FOGCore.class.php for the FOGScheduler tasking. Fixes, again, the size display issue on the Dashboard Page. If it reached 50% (50%-50% = 0% available, 100% used) it displayed as if they entire file system was completely used.
-
r1319 released.
Fixes the Dashboard Tasks display to show the appropriate dates. The data itself was correct, but the dating was one day into the future. How is it possible if the date hasn’t happened yet?
-
r1320 released.
Just adds multicast deploy for raw images.
-
r1321 released.
Just removes the raw image need. Basically using partclone.imager for the raw image types as well. This allows partclone.restore to be the predominant use for the, well, restore parts. Hopefully all don’t mind. It makes coding that much easier.
