[Voyage-linux] grub 17 errors

Sun Mar 25 16:10:40 HKT 2012

Hello Julie and list,

there is a relative precise docu about all grub errors at
http://www.gentoo.org/doc/en/grub-error-guide.xml

There are a lot docus with technical aspects of Linux on Gentoo's
website, which are worth reading and can also be applied on nearly every
other Linux distribution.

Enjoy

On 03/24/2012 03:35 PM, jfh at greenhousepc.com wrote:
> Frank,
>
> Neither of those is the problem.  The ALIX.3D2 doesn't recognize a USB
> drive at boot time as a "BIOS Drive", so even if they had a USB drive
> plugged in, that wouldn't cause the problem.
>
> Also, these systems boot multiple times with the same partition table
> and flash drive.  Then, all of the sudden, they wind up with a grub 17
> error.  When I look at the partition table later, nothing has changed
> and nothing is wrong.
>
> What I need is something that can read the grub information from disk
> and determine why grub would fail.  I've used debugfs to make sure the
> filesystem is in proper order, and various fdisk tools to make sure
> the partition table is in proper order, but there's no tool that can
> indicate what might be wrong with the data grub has written to the MBR.
>
> If someone has a map of which variables / data grub writes to where in
> the MBR, that would be very helpful.  I'm trying to get some new
> hardware products developed (I have a nice DIN rail mountable board
> that can be used with a 9.6 volt NiMH battery pack as a UPS, and
> several other goodies that are likely less interesting) and trying to
> debug grub at the same time has finally reached my limit of "things I
> can do all at once."
> --
> Julie Haugh
> Senior Design Engineer
> greenHouse Computers, LLC // jfh at greenhousepc.com
> <http://greenhousepc.com> // greenHousePC on Skype
>
>
>     -------- Original Message --------
>     Subject: Re: [Voyage-linux] grub 17 errors
>     From: Frank Hoeflich <frankhoeflich at yahoo.com
>     <mailto:frankhoeflich at yahoo.com>>
>     Date: Sat, March 24, 2012 1:47 am
>     To: voyage-linux at voyage.hk <mailto:voyage-linux at voyage.hk>,
>     jfh at greenhousepc.com <mailto:jfh at greenhousepc.com>
>
>     Julie:
>
>         Most of the Googling on this turns up partition table disk
>     order errors:
>
>     http://stringofthoughts.wordpress.com/2009/05/24/grub-error-17-debianubuntu/
>     http://iknowkungfoo.com/blog/index.cfm/2008/6/16/Stupid-solution-for-Grub-error-17
>
>     Note in the second one that people who have other drives plugged
>     in sometimes get relief by disconnecting them.  Any chance that in
>     the field your units have USB drives plugged/unplugged into the
>     Alix that werent/were present at the factory?  Does simply
>     reinstalling grub (regular, not debug version) do the trick as well?
>
>     --Frank
>
>     --- On *Fri, 3/23/12, jfh at greenhousepc.com
>     <mailto:jfh at greenhousepc.com> /<jfh at greenhousepc.com
>     <mailto:jfh at greenhousepc.com>>/* wrote:
>
>         From: jfh at greenhousepc.com <mailto:jfh at greenhousepc.com>
>         <jfh at greenhousepc.com <mailto:jfh at greenhousepc.com>>
>         Subject: [Voyage-linux] grub 17 errors
>         To: voyage-linux at voyage.hk <mailto:voyage-linux at voyage.hk>
>         Date: Friday, March 23, 2012, 10:25 AM
>
>         Greets,
>
>         I have a client who has been plagued by Grub 17 errors for the
>         longest time.  I'm using grub 0.97 ("legacy") on a version 0.6
>         version of Voyage.  Please don't tell me to upgrade -- it
>         takes months to verify that a change of that nature is going
>         to be safe.
>
>         I've done everything I can to debug the problem, but the error
>         path in fsys_ext2fs.c doesn't seem applicable -- the partition
>         sizes have been verified to be correct, there are no errors
>         from fsck (other than the expected ones because the ALIX.3D2
>         was power cycled just before it happened) indicating the
>         superblock had a bad magic number.
>
>         I'm not able to recreate the problem here, and updating the
>         filesystem so it has a debugging version of grub fixes the
>         problem.
>
>         The systems are cloned from an image using "dd", but grub is
>         then re-installed manually.  All of the system I have in
>         house, and all of the tests that I perform on systems I ship,
>         are successful.  When they receive the systems they perform
>         the exact same test, which is also successful.  It isn't until
>         the systems are installed in the field that they fail.
>
>         I'm using Kingston "Elite Pro" 8GB compact flash for the O/S
>         image and application.  Although it isn't "industrial" grade
>         flash, the application was written so it minimizes the number
>         of modified filesystem blocks as much as possible.
>
>         Help?
>         --
>         Julie Haugh
>         Senior Design Engineer
>         greenHouse Computers, LLC // jfh at greenhousepc.com
>         <http://greenhousepc.com> // greenHousePC on Skype
>
>         -----Inline Attachment Follows-----
>
>         _______________________________________________
>         Voyage-linux mailing list
>         Voyage-linux at list.voyage.hk
>         </mc/compose?to=Voyage-linux at list.voyage.hk>
>         http://list.voyage.hk/mailman/listinfo/voyage-linux
>
>
> _______________________________________________
> Voyage-linux mailing list
> Voyage-linux at list.voyage.hk
> http://list.voyage.hk/mailman/listinfo/voyage-linux
>   

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.voyage.hk/pipermail/voyage-linux/attachments/20120325/67fa5b69/attachment.html>