NET5501 Boot Problem

[Tom Mazzotta]

After several months of not using my NET5501, I upgraded the BIOS to
1.33c in preparation of installing FreeBSD 7.0-RELEASE. Using PXE, I
installed to a Hard Drive (ATA i/f) that previously ran FreeBSD 6.2 in
the 5501.

The install worked fine, however, after doing a cold restart, I noticed
that the boot process would "hang" for about 20sec before recognizing
the HD, then boot FreeBSD. In this situation, the o/s would ultimately
fail to mount the root fs. If I then hit the RESET switch, I can usually
get the system to fully boot... sometimes I need to hit RESET twice to
get it to work.

Once FreeBSD is running, I can issue a "reboot" command at an o/s
command prompt and the system bounces as expected.

There are a few very similar reports in the list archive, although it
seems that they all were using CF cards. Do all 5501's flashed w/ 1.33c
behave this way, or is this problem limited to a bad batch of system
boards? Does Soekris believe the problems are caused by the storage
devices?

-tm

_______________________________________________
Soekris-tech mailing list
Soekris-tech@...
http://lists.soekris.com/mailman/listinfo/soekris-tech

[Wim Vandeputte-2]

this is a welknown bug in the BIOS

What was your previous version where it did work? 1.32i?

On Wed, Sep 24, 2008 at 10:18:28PM -0400, Tom Mazzotta wrote:

> After several months of not using my NET5501, I upgraded the BIOS to
> 1.33c in preparation of installing FreeBSD 7.0-RELEASE. Using PXE, I
> installed to a Hard Drive (ATA i/f) that previously ran FreeBSD 6.2 in
> the 5501.
>
> The install worked fine, however, after doing a cold restart, I noticed
> that the boot process would "hang" for about 20sec before recognizing
> the HD, then boot FreeBSD. In this situation, the o/s would ultimately
> fail to mount the root fs. If I then hit the RESET switch, I can usually
> get the system to fully boot... sometimes I need to hit RESET twice to
> get it to work.
>
> Once FreeBSD is running, I can issue a "reboot" command at an o/s
> command prompt and the system bounces as expected.
>
> There are a few very similar reports in the list archive, although it
> seems that they all were using CF cards. Do all 5501's flashed w/ 1.33c
> behave this way, or is this problem limited to a bad batch of system
> boards? Does Soekris believe the problems are caused by the storage
> devices?
>
> -tm
>
> _______________________________________________
> Soekris-tech mailing list
> Soekris-tech@...
> http://lists.soekris.com/mailman/listinfo/soekris-tech

_______________________________________________
Soekris-tech mailing list
Soekris-tech@...
http://lists.soekris.com/mailman/listinfo/soekris-tech

[Tom Mazzotta]

Wim,

It's possible I was running 1.32i, but I have since discarded my notes
on the BIOS upgrade so I can't give you a straight answer, sorry! Are
there any work-arounds for FreeBSD-70 systems in the interim until a
BIOS fix is released? Since the box will hang after losing power, this
would obviously be a problem w/ production systems.

Thanks!

-tm


On Thu, 2008-09-25 at 05:47 -0400, Wim Vandeputte Mailing list only
wrote:

> this is a welknown bug in the BIOS
>
> What was your previous version where it did work? 1.32i?
>
> On Wed, Sep 24, 2008 at 10:18:28PM -0400, Tom Mazzotta wrote:
> > After several months of not using my NET5501, I upgraded the BIOS to
> > 1.33c in preparation of installing FreeBSD 7.0-RELEASE. Using PXE, I
> > installed to a Hard Drive (ATA i/f) that previously ran FreeBSD 6.2 in
> > the 5501.
> >
> > The install worked fine, however, after doing a cold restart, I noticed
> > that the boot process would "hang" for about 20sec before recognizing
> > the HD, then boot FreeBSD. In this situation, the o/s would ultimately
> > fail to mount the root fs. If I then hit the RESET switch, I can usually
> > get the system to fully boot... sometimes I need to hit RESET twice to
> > get it to work.
> >
> > Once FreeBSD is running, I can issue a "reboot" command at an o/s
> > command prompt and the system bounces as expected.
> >
> > There are a few very similar reports in the list archive, although it
> > seems that they all were using CF cards. Do all 5501's flashed w/ 1.33c
> > behave this way, or is this problem limited to a bad batch of system
> > boards? Does Soekris believe the problems are caused by the storage
> > devices?
> >
> > -tm
> >
> > _______________________________________________
> > Soekris-tech mailing list
> > Soekris-tech@...
> > http://lists.soekris.com/mailman/listinfo/soekris-tech
>

_______________________________________________
Soekris-tech mailing list
Soekris-tech@...
http://lists.soekris.com/mailman/listinfo/soekris-tech