[Intrepid Ibex] Ubuntu live CD fails to load with Adaptec SCSI controller installed

Bug #262294 reported by Michael
6
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Fix Released
Undecided
Unassigned
Nominated for Intrepid by Walldorf2000

Bug Description

I downloaded the 'live' CD images of the daily 32-bit and 64-bit Ubuntu Intrepid Ibex builds from August 23, 2008.
I burned the images to CD.
I tried using both CDs on my two home desktop computers which each had an Adaptec PCI SCSI controller installed.
Neither CD would boot up with an Adaptec PCI SCSI controller installed, but both computers would boot up with no error messages when there was no Adaptec PCI SCSI controller installed.
I assume this is a problem because Ubuntu does not tolerate the on-board BIOS on PCI SCSI controllers but I have no prior knowledge of Ubuntu (or linux in general).
I downloaded the 'live' CD images of the daily 32-bit and 64-bit Ubuntu Intrepid Ibex builds from August 26, 2008, with similar results.
I downloaded the 'live' CD images of Debian Live Lenny Beta 1 and they would not boot up with the Adaptec PCI SCSI controllers installed either.
I downloaded a 'live' CD of OpenSUSE 11.0 and it would boot up with the Adaptec PCI SCSI controllers installed so I assume the problem is peculiar to Ubuntu and Debian linux distributions.
The screens that result when booting up Ubuntu Intepid Ibex builds with an Adaptec PCI SCSI controller installed are as follows:
---------------------
---------------------
ECS GeForce6100PM-M2 v2.0 motherboard
(nVidia Geforce6100/nForce430)
http://www.ecsusa.com/ECSWebSite/Downloads/ProductsDetail_Download.aspx?detailid=857&DetailName=New&DetailDesc=&CategoryID=1&MenuID=6&LanID=9
AMD Athlon™ X2 Dual-Core 4200+ processor
Adaptec 29160N PCI SCSI controller
=============
error screen with 64-bit Ubuntu:

[0.004000]Aperture beyond 4GB. Ignoring
[0.004000]Your BIOS doesn't leave a aperture memory hole.
[0.004000]Please enable the IOMMU option in the BIOS setup.
[0.004000]This costs you 64MB of RAM.

Loading, please wait...

BusyBox v1.1.0.2 (Ubuntu 1:1.10.2-1ubuntu4)
built-in shell (ash)
Enter 'help' for a list of built-in commands.
(initramfs)(cursor)

Kernel alive
kernel direct mapping tables up to 120000000@8000-e000
============
error screen with 32-bit Ubuntu:

Loading, please wait...

BusyBox v1.1.0.2 (Ubuntu 1:1.10.2-1ubuntu4)
built-in shell (ash)
Enter 'help' for a list of built-in commands.
(initramfs)(cursor)

Kernel alive
kernel direct mapping tables up to 120000000@8000-e000
-------------------
-------------------
ECS 945GCT-M/13333 motherboard
http://www.ecsusa.com/ECSWebSite/Downloads/ProductsDetail_Download.aspx?detailid=830&DetailName=New&DetailDesc=&CategoryID=1&MenuID=6&LanID=9
Intel E2200 processor
Adaptec 29160 PCI SCSI controller
==============
error screen with 32-bit Ubuntu:

vertical white lines - no text
==============
error screen with 64-bit Ubuntu:

Loading, please wait...

BusyBox v1.1.0.2 (Ubuntu 1:1.10.2-1ubuntu4)
built-in shell (ash)
Enter 'help' for a list of built-in commands.
(initramfs)(cursor)

Kernel alive
kernel direct mapping tables up to 100000000@8000-d000

Revision history for this message
Michael (mjb-safe-mail-deactivatedaccount) wrote :

Additional information: The PCI SCSI controllers are only used for A3 size flatbed scanners with SCSI interfaces, Microtek ScanMaker 6400XL and Epson GT-10000+. I do not own any other SCSI devices. The referenced failure to boot a live CD occurs consistently regardless of whether or not the scanners are attached to the PCI SCSI controllers.

description: updated
Revision history for this message
Walldorf2000 (walldorf2000) wrote :

I can confirm this problem.

I have exactly the same symptoms when I try to boot the Live-CD from 9th Sep. 2008.

The boot ends with an (initramfs) prompt.

In my system is an Adaptec 2940 AU SCSI controler. It is a AGP board with an Athlon 500.

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Hi Guys,

Michael, would you be willing to test the latest images. Alpha5 which was released on Sept 5 contains a newer 2.6.27 kernel. For more information refer to http://www.ubuntu.com/testing/intrepid/alpha5 .

Walldorf2000 - after you get dropped into the busybox shell, would you be able to capture the dmesg output from that machine and attach it to this bug report? I'm told that you can type 'httpd' at the busybox prompt and then point another machines web browser to the ip address of that box and you'll see the content of /var/log . Thanks.

Revision history for this message
Michael (mjb-safe-mail-deactivatedaccount) wrote :

I tested the latest images, alpha5 and the latest daily build on September 11, on one of my two home computers.
Neither image would boot up.

ECS GeForce6100PM-M2 v2.0 motherboard
(nVidia Geforce6100/nForce430)
http://www.ecsusa.com/ECSWebSite/Downloads/ProductsDetail_Download.aspx?detailid=857&DetailName=New&DetailDesc=&CategoryID=1&MenuID=6&LanID=9
AMD Athlon™ X2 Dual-Core 4200+ processor
Adaptec 29160N PCI SCSI controller
=============
The error screen with 64-bit Ubuntu was exactly the same except the last line says 'Kernel really alive' now.
The error screen with 32-bit Ubuntu was exactly the same except the last two lines are missing, with the cursor blinking after (initramfs), as described by Walldorf2000.

Revision history for this message
Walldorf2000 (walldorf2000) wrote :

I apologize for high jacking this bug. It is not Adaptec related. I tested the DVD from 9th of September with three machines.

I got the (initramfs) prompt on the old Atholon 500 machine described above and a brand new machine with AMD64 X2 with an AMD 770 chipset and SATA HDD. The new machine failed for normal SATA mode and AHCI mode. There is no SCSI device in it. The Ubuntu 8.04 DVD boots up fine with both modes.

The only machine I found where I could boot this DVD is an four year old Intel machine (mobo and CPU).

Revision history for this message
Michael (mjb-safe-mail-deactivatedaccount) wrote :

I tested the 32-bit and 64-bit Intrepid Ibex alpha5 build on my other home computer and neither image would boot up on that computer with the referenced Adaptec PCI SCSI controller either.
-------------------
ECS 945GCT-M/13333 motherboard
http://www.ecsusa.com/ECSWebSite/Downloads/ProductsDetail_Download.aspx?detailid=830&DetailName=New&DetailDesc=&CategoryID=1&MenuID=6&LanID=9
Intel E2200 processor
Adaptec 29160 PCI SCSI controller
==============
The error screen with 32-bit Ubuntu is now identical on both my home desktop computers with an Adaptec PCI SCSI card installed.
The error screen with 64-bit Ubuntu is now identical on both my home desktop computers with an Adaptec PCI SCSI card installed.

Revision history for this message
Michael (mjb-safe-mail-deactivatedaccount) wrote :

I downloaded 32-bit and 64-bit DVD images of the daily September 20, 2008 build of Ubuntu 8.10.
Neither build would boot up on either of my two desktop computers with the Adaptec PCI SCSI controllers installed when using their respective internal optical drives.
I had been using the internal DVD-RW drives on my desktop computers to try to boot up, so as an experiment I tried booting Ubuntu 8.10 from my (I/O Magic brand) USB 2.0 external DVD-RW drive.
To my surprise, both 32-bit and 64-bit DVD images of the daily September 20, 2008 build of Ubuntu 8.10 would boot on both computers with Adaptec PCI SCSI controllers using the USB 2.0 external DVD-RW drive.
The bug which I am experiencing appears to occur only when booting Ubuntu from an internal optical drive when Adaptec PCI SCSI controllers are installed.
The DVD-RW drive in the system with the ECS 945GCT-M/13333 motherboard is an Optiarc AD-7190S drive with serial ATA interface.
The DVD-RW drive in the system with the ECS GeForce6100PM-M2 v2.0 motherboard is a Pioneer DVR-1810 drive with ATAPI-5 (IDE) interface.

Revision history for this message
Michael (mjb-safe-mail-deactivatedaccount) wrote :

I downloaded 32-bit and 64-bit DVD images of the beta build of Ubuntu 8.10.
Both builds would boot up on either of my two desktop computers with the Adaptec PCI SCSI controllers installed when using their respective internal optical drives (and the external DVD-RW drive also).
It appears that something was changed from the alpha builds which has fixed the bug I was experiencing.
I guess this bug report can be closed.

There was another odd bug with all the builds: either an approximate three-second burst of music would play when Ubuntu was booting and the sound would not function after Ubuntu was booted to the desktop; or no burst of music would play when Ubuntu was booting and the sound functioned after Ubuntu booted to the desktop. I did not check if the sound bug is related to the Adaptec PCI SCSI controllers but I will do more testing later and file a new bug report if necessary.

Revision history for this message
Michael (mjb-safe-mail-deactivatedaccount) wrote :

The above referenced audio sound bug has been fixed in the released version of Ubuntu 8.10.

Changed in linux:
status: New → Fix Released
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.