Syslinux - Aug 2013 - Intel DX79TO localboot problem with CentOS

On 08/05/2013 11:29 PM, Jonas Keidel wrote:
> 2013/8/6 H. Peter Anvin <hpa at zytor.com>
> 
>> What about "chain.c32 hd1" or "chain.c32 hd1 swap"?
>>
> 
> That's right, but due to our pxe environment we need a static entry
which
> handles both, booting from first and second hdd if one fails. localboot
> does so.
> 
Actually this does highlight an issue with chain.c32... there really
should be support for "try this, check X, then that".

	-hpa

2013/8/6 H. Peter Anvin <hpa at zytor.com>
> On 08/05/2013 11:29 PM, Jonas Keidel wrote:
> > 2013/8/6 H. Peter Anvin <hpa at zytor.com>
> >
> >> What about "chain.c32 hd1" or "chain.c32 hd1
swap"?
> >>
> >
> > That's right, but due to our pxe environment we need a static
entry which
> > handles both, booting from first and second hdd if one fails.
localboot
> > does so.
> >
>
> Actually this does highlight an issue with chain.c32... there really
> should be support for "try this, check X, then that".
>
Can that be implemented? Because i think that is a must-have feature of
chain.c32 to test if there is a bootable sector on first hdd --> boot from
that.
If not it should try to look for a bootable sector on the second hdd and
try that.
if both don't work it should abort. At current state it always aborts if
the given hdd (hd0 or hd1) is not bootable.

>
>         -hpa
>
>
>

2013/8/7 Jonas Keidel <jonas at jonas-keidel.de>
>
>
> 2013/8/6 H. Peter Anvin <hpa at zytor.com>
>
>> On 08/05/2013 11:29 PM, Jonas Keidel wrote:
>> > 2013/8/6 H. Peter Anvin <hpa at zytor.com>
>> >
>> >> What about "chain.c32 hd1" or "chain.c32 hd1
swap"?
>> >>
>> >
>> > That's right, but due to our pxe environment we need a static
entry
>> which
>> > handles both, booting from first and second hdd if one fails.
localboot
>> > does so.
>> >
>>
>> Actually this does highlight an issue with chain.c32... there really
>> should be support for "try this, check X, then that".
>>
>
> Can that be implemented? Because i think that is a must-have feature of
> chain.c32 to test if there is a bootable sector on first hdd --> boot
from
> that.
> If not it should try to look for a bootable sector on the second hdd and
> try that.
> if both don't work it should abort. At current state it always aborts
if
> the given hdd (hd0 or hd1) is not bootable.
>
>
So what about this topic? Do you think it can be implemented?

>
>>         -hpa
>>
>>
>>
>