HPT370 infinite reboot loop as GRUB loads

Bug #529363 reported by Shannon VanWagner
16
This bug affects 3 people
Affects Status Importance Assigned to Milestone
grub2 (Ubuntu)
Expired
Undecided
Unassigned

Bug Description

Binary package hint: grub2

Ubuntu 9.10 Karmic installs perfectly until rebooted, when it enters an infinite reboot loop just as GRUB loads.
Have tried repairing grub with LiveCD - everything reports as being fixed, but then after reboot - same reboot loop.
Fedora 12 installs and works without issue.
Ubuntu 10.04 Lucid Alpha 3(02-28-10) doesn't even detect the drive.
SATA/RAID controller is Highpoint 370, motherboard is Shuttle AS45GT.
This is an old motherboard, but it's just irritating that Fedora 12 with legacy grub installs without issue.
Uwe Zeisberger describes a similar problem along with a fix (for legacy GRUB) at http://lists.gnu.org/archive/html/bug-grub/2004-07/msg00113.html . The posting describes something about how the HPT370 overwrites the default GRUB storage area on the hdd, and about moving it.. But I'm not sure how to do this with GRUB2. I tried to email Uwe, but the email address doesn't work.

Thanks for any answers you can provide.
Sincerely,
Shannon VanWagner

ProblemType: Bug
Architecture: i386
Date: Sun Feb 28 07:39:05 2010
DistroRelease: Ubuntu 9.10
LiveMediaBuild: Ubuntu 9.10 "Karmic Koala" - Release i386 (20091028.5)
Package: grub-pc 1.97~beta4-1ubuntu3
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-14.48-generic
SourcePackage: grub2
Uname: Linux 2.6.31-14-generic i686
XsessionErrors:
 (gnome-settings-daemon:3039): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (gnome-settings-daemon:3039): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (nautilus:3140): Eel-CRITICAL **: eel_preferences_get_boolean: assertion `preferences_is_initialized ()' failed
 (polkit-gnome-authentication-agent-1:3157): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed
 (gnome-panel:3139): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to allocate widget with width -4 and height 24

Revision history for this message
Shannon VanWagner (shannon-vanwagner) wrote :
Revision history for this message
Shannon VanWagner (shannon-vanwagner) wrote :

I've attached lspci -v output.

Revision history for this message
Rod B. (rbrbrb2004) wrote :

I am having the same issue on Abit KG7-RAID which has a on-board HPT370. My HPT370's BIOS are version 2.351 which is the latest from Highpoint.

Revision history for this message
Shannon VanWagner (shannon-vanwagner) wrote :

I read somewhere that installing a /boot partition on a separate EIDE hard drive may fix the problem but I haven't had a chance to test it yet. When I get around to testing it, I'll post the results here.
Cheers! Shannon

Revision history for this message
Rod B. (rbrbrb2004) wrote :

This might have something to do with the issue. Here is what I find in dmesg while attempting to access a drive connected to the hpt370. I see the same timeout when trying to install grub on the drive. I booted Ubuntu from the 9.10 i386 install CD live session. The driver the live session is using is pata_hpt37x.ko.

[ 383.000071] ata3: lost interrupt (Status 0x58)
[ 383.004009] ata3: drained 32768 bytes to clear DRQ.
[ 383.043334] ata3.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
[ 383.043350] ata3.00: cmd c8/00:80:af:10:00/00:00:00:00:00/e0 tag 0 dma 65536 in
[ 383.043352] res 40/00:00:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 383.043356] ata3.00: status: { DRDY }
[ 383.043486] ata3: soft resetting link
[ 383.220495] ata3.00: configured for UDMA/100
[ 383.220505] ata3.00: device reported invalid CHS sector 0
[ 383.220523] ata3: EH complete
[ 414.000507] ata3: lost interrupt (Status 0x58)
[ 414.004009] ata3: drained 32768 bytes to clear DRQ.
[ 414.043771] ata3.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
[ 414.043786] ata3.00: cmd c8/00:80:af:10:00/00:00:00:00:00/e0 tag 0 dma 65536 in
[ 414.043788] res 40/00:00:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 414.043792] ata3.00: status: { DRDY }
[ 414.043922] ata3: soft resetting link
[ 414.220469] ata3.00: configured for UDMA/100
[ 414.220479] ata3.00: device reported invalid CHS sector 0
[ 414.220491] ata3: EH complete

Revision history for this message
Shannon VanWagner (shannon-vanwagner) wrote :

I've installed a PATA drive and am using it for the root "/" and swap partitions and now Ubuntu 10.04 installed perfectly, and everything works fine.

My /home drive is ext4 and is on the sata drive that is connected to the highpoint controller.

I still haven't tested whether putting just /boot on the pata drive would work, but I assume it would.

Revision history for this message
Niels tijssen (tijssenn) wrote :

Hi

I can reproduce this bug again with the following setup.
Ubuntu Karmic 09-10 stock 32 bit installation disk.

Mainboard: Abit KR7A - Raid
CPU AMD Athlon 1900+
Highpoint HPT 370/372 external ATA controller.
Via VT8366A southbridge

Hard disk setup

Primary master: DVD rom drive.
Secondary Master: DVD Rewriter.
External highpoint HPT 370 controller: hard disk.

Start situation:

PC did boot to windows XP normally before Kubuntu Installation.
Installation of Kubuntu Karmic 09-10 went normally.
After reboot PC gets in infinite reboot loop as described in parent post.
No grub errors visible on console.

Revision history for this message
Niels tijssen (tijssenn) wrote :

Here is the generated grub.cfg after installation.

Revision history for this message
Niels tijssen (tijssenn) wrote :

Here is an lspci dump when the boot CD has started up. It shows that the highpoint controller has been detected.

Revision history for this message
Mossroy (mossroy) wrote :

I experience the exact same problem.
I have an old computer with an Abit KR7A Raid motherboard, which has an HPT372 controller (see http://file.abit.com.tw/pub/download/manual/english/kr7a-raid.zip)
It worked correctly for years under Ubuntu 8.04.
I did a fresh install of Ubuntu 10.04 (not an upgrade) and had the infinite reboot loop too.

A new fresh install of Ubuntu 8.04.3 from the liveCD works well.
The same reboots occur if I install from the liveCD or from the alternate CD.
If I boot on the liveCD, it works well and I can access the hard disk without any problem. I can see that nothing has been written in /var/log/dmesg or other log files : the reboot happens inside Grub2, before booting the OS.
I suppose that the reason why it works with ubuntu 8.04 is that it uses Grub, and not Grub2

My hard disk is an IDE one, plugged in the IDE3 port (the one controlled by the HPT372). If I plug it in IDE4 (also controlled by HPT372), and reinstall Ubuntu, I have the same behavior.
My IDE1 and IDE2 ports are used by CD/DVD drives and are slower than the HPT372 (U-DMA-133 vs U-DMA-100). I'll try to plug my hard disk on IDE1 or IDE2 (which are not controlled by HPT372), to check if the bug is related to this controller

Changed in grub2 (Ubuntu):
status: New → Confirmed
Revision history for this message
Mossroy (mossroy) wrote :

Plugging the hard disk to IDE1 or IDE2 worked as a workaround for me.
So I confirm that it looks like a bug of Grub2, related to HPT controllers.
It is a regression as it is working fine with Grub

Niels : this workaround should probably work for you too. I had to disconnect any drive of the HPT ports (and put both CD/DVD drives on a single port with master/slave)
The sad news is that the hard disk is probably slower as it is Ultra-DMA-100 instead of Ultra-DMA-133. But at least it works.

Revision history for this message
Marcus Tomlinson (marcustomlinson) wrote :

This release of Ubuntu is no longer receiving maintenance updates. If this is still an issue on a maintained version of Ubuntu please let us know.

Changed in grub2 (Ubuntu):
status: Confirmed → Incomplete
Revision history for this message
Launchpad Janitor (janitor) wrote :

[Expired for grub2 (Ubuntu) because there has been no activity for 60 days.]

Changed in grub2 (Ubuntu):
status: Incomplete → Expired
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.