PXE kernel panic on git trunk
-
after chain loading boot.php and getting the menu, when I select registration and it loads the kernel (I’ve tried both bzImage32 v 4.6.2 and v 4.5.3) I get a kernel panic.
Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(1,0)
Kernel Offset: disabled
—[ end Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(1,0)the system halts and i have to hard shut it down, start it up, hard shut it down, start it up and it will boot to the PXE card again.
(PS - not a FOG novice, been using it since version 0.32 but this is the first time I’m using the trunk version)
-
What model is the problem host? What version of fog?
-
Going to guess, you updated from 0.32?
-
No, unfortunately not. This is a brand new install on a new centos 7 system with trunk - the one you helped fix the install script for. But not an upgrade.
-
Your dhcp server is using undionly.kpxe to load the hosts?
-
Have you tried on a different system?
-
FOG version trunk as of yesterday. Its a Dell Latitude 3450.
-
yes, DHCP is using undionly.kpxe - I don’t actually run any 0.32 versions anymore. The oldest I run is 1.2.0
This might be a problem with the Dell systems I’m trying to image. should i repost in the hardware area?
-
@Chris-Shipley You can try a more recent kernel. I think 4.1.2 is the highest 1.2.0 shows?
-
We only have the single installation at this site with Centos 7, new install, git trunk.
I can confirm this happens on a Lenovo Thinkpad W530 as well as the Dell Latitude 3450.
This is also the installation with the problem for chainloading that I am “getting around” by going into the iPXE system and writing my own “chain http://10.10.2.18/fog/service/ipxe/boot.php” command to get to the boot menu. (see here: https://forums.fogproject.org/topic/7879/pxe-boot-issue-video-included)
-
@Tom-Elliott using git trunk version. OP says I’ve tried both bzImage32 v 4.6.2 and v 4.5.3
-
Fixing the chain bootloading issue on the other thread i posted here also fixed this kernel panic issue. thank you Tom.