[syslinux] PXELINUX [BIOS] 6.02 - Upgrade to 6.02 from 4.07 introduces local boot regression

Gene Cumm gene.cumm at gmail.com
Mon Dec 9 17:05:43 PST 2013


On Mon, Dec 9, 2013 at 3:27 PM, Cory Fair <cory.fair at gmail.com> wrote:
> Hello All, this is my first post so I'll try to be as verbose as possible.
>
> We have a PXE deployment server serving mostly fujitsu RX300 rack mount
> servers with Intel 82575EB Gigabit Ethernet Controllers onboard. I recently
> upgraded from version 4.07 to version 6.02 and, without making any edits to
> my default menu, immediately noticed a regression. I use vesamenu.c32 to
> present a menu to the user which starts with something like this:
>
> default vesamenu.c32
>     prompt 0
>
>     menu title Host Provisioning Tool
>     menu INCLUDE pxelinux.cfg/graphics.conf
>     MENU AUTOBOOT Starting Local System in # seconds
>     MENU BACKGROUND splash.jpg
>
>     label bootlocal
>       menu label ^Boot This Host
>       menu default
>       localboot 0
>       timeout 80

Reflowing for clarity of scope:

default vesamenu.c32
prompt 0
timeout 80

menu title Host Provisioning Tool
menu INCLUDE pxelinux.cfg/graphics.conf
MENU AUTOBOOT Starting Local System in # seconds
MENU BACKGROUND splash.jpg

label bootlocal
    menu label ^Boot This Host
    menu default
    localboot 0



Where's pxelinux.cfg/graphics.conf?

> Prior to upgrading to pxelinux 6.02 If you did nothing for 8 seconds the
> system would boot from the local disk. Now the system will hang after the
> timeout. The only visual difference I can discern is that when the system
> properly boots I see a message briefly indicating the PXE boot rom had
> exited properly. Now with version 6.02 It hangs just before that.
>
> Another datapoint, I reverted back to version 6.01 and the problem seems to
> be resolved.
> Final datapoint, if you hit "enter" instead of waiting for the timeout
> (using version 6.02) it will boot properly

Rather curious.  What if you dropped the timeout to 0.8-2.0 seconds or
even 0.2 seconds?  I believe I've seen some quirky behavior in other
PXE stacks where they act up after about 3 seconds.  It's as if
they've sometimes arbitrarily decided that the PXE services won't live
longer (since you should have a real kernel).

> My basic question is what is right place to report this issue to the
> project?

I'd say the list is an excellent place to start.  If it's not a
duplicate report of an existing bug at http://bugzilla.syslinux.com
nor already resolved by any of the patches that aren't pulled into the
code base, then it will likely need to be filed there

-- 
-Gene


More information about the Syslinux mailing list