MPT drive not detected under VMware WS5

Bug #112860 reported by Tux
4
Affects Status Importance Assigned to Milestone
Linux
Invalid
Medium
linux-source-2.6.22 (Ubuntu)
Invalid
Medium
Ben Collins
module-init-tools (Ubuntu)
Fix Released
High
Ben Collins

Bug Description

Binary package hint: linux-source-2.6.22-1.5

When I boot with kernel 2.6.22 (even with recovery mode) it drops to initramfs with the following error:
check root= bootarg cat /proc/cmdline
or missing modules, devices: cat /proc/modules ls /dev
ALERT! /dev/disk/by-uuid/119c2980-84b2-409e-b99f-0571f85c2356 does not exist. Dropping to a shell.

I am running an Ubuntu 7.04 Feisty Fawn upgraded to the latest Gutsy Gibbon packages on VMware 5.0.
Booting with kernel 2.6.15 works fine.

Greetings,

Peter Hoogkamer

Tux (peter-hoogkamer)
description: updated
Revision history for this message
Ben Collins (ben-collins) wrote : Re: drop to initramfs with linux-image-2.6.22-1.5

Some of this doesn't make sense. If you are running feisty, upgraded to gutsy, then wouldn't it be 2.6.20 that works, and not 2.6.15?

Also, please boot without "splash quiet" on the command line (booting to recover mode usually does this well enough).

Preferably, use a USB stick/drive that you can mount while in busybox and run this command:

dd if=/proc/kmsg of=/where-you-mounted-usb/dmesg.txt bs=1; sleep 5; sync

Then reboot and attach the file to this bug report (do not paste into comment, use the web interface to actually attach it).

Thanks

Changed in linux-source-2.6.22:
assignee: nobody → ben-collins
status: Unconfirmed → Needs Info
Revision history for this message
Tux (peter-hoogkamer) wrote : Re: [Bug 112860] Re: drop to initramfs with linux-image-2.6.22-1.5

Hello Ben,

You are right about the kernel. It boots with 2.6.20-15 correctly. But
I get the error on kernel 2.6.22, even in recovery mode. I will check
you're suggestion with the USB stick later on.

Greetings,

Peter

2007/5/8, Ben Collins <email address hidden>:
> Some of this doesn't make sense. If you are running feisty, upgraded to
> gutsy, then wouldn't it be 2.6.20 that works, and not 2.6.15?
>
> Also, please boot without "splash quiet" on the command line (booting to
> recover mode usually does this well enough).
>
> Preferably, use a USB stick/drive that you can mount while in busybox
> and run this command:
>
> dd if=/proc/kmsg of=/where-you-mounted-usb/dmesg.txt bs=1; sleep 5; sync
>
> Then reboot and attach the file to this bug report (do not paste into
> comment, use the web interface to actually attach it).
>
> Thanks
>
> ** Changed in: linux-source-2.6.22 (Ubuntu)
> Assignee: (unassigned) => Ben Collins
> Status: Unconfirmed => Needs Info
>
> --
> drop to initramfs with linux-image-2.6.22-1.5
> https://bugs.launchpad.net/bugs/112860
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
Ben Collins (ben-collins) wrote : Re: drop to initramfs with linux-image-2.6.22-1.5

Please retest against 2.6.22-3-generic, which should now be in the archive.

Thanks

Revision history for this message
Tux (peter-hoogkamer) wrote : Re: [Bug 112860] Re: drop to initramfs with linux-image-2.6.22-1.5

Ben,

I have tryed to boot with 2.6.22-3.9 but no luck. Same problem. As far
as my knowledge goes I can't find any strange things in the console
messages before the drop to initramfs. It stalls with the alert that
it can't find my \dev\uuid\[partition code]\ and drops to initramfs.
I will try you're advise with the usb-stick trying to get the dmesg for you.

Peter

2007/5/15, Ben Collins <email address hidden>:
> Please retest against 2.6.22-3-generic, which should now be in the
> archive.
>
> Thanks
>
> --
> drop to initramfs with linux-image-2.6.22-1.5
> https://bugs.launchpad.net/bugs/112860
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
Tux (peter-hoogkamer) wrote :

Ben,

I forgot to mention that I also tryed to boot in recovery mode and got
the same message
ALERT! /dev/disk/by-uuid/[partition-id] does not exist! Dropping to shell
My disk is being detected as a scsi disk at sr0.

Peter

2007/5/15, Peter Hoogkamer <email address hidden>:
> Ben,
>
> I have tryed to boot with 2.6.22-3.9 but no luck. Same problem. As far
> as my knowledge goes I can't find any strange things in the console
> messages before the drop to initramfs. It stalls with the alert that
> it can't find my \dev\uuid\[partition code]\ and drops to initramfs.
> I will try you're advise with the usb-stick trying to get the dmesg for you.
>
> Peter
>
> 2007/5/15, Ben Collins <email address hidden>:
> > Please retest against 2.6.22-3-generic, which should now be in the
> > archive.
> >
> > Thanks
> >
> > --
> > drop to initramfs with linux-image-2.6.22-1.5
> > https://bugs.launchpad.net/bugs/112860
> > You received this bug notification because you are a direct subscriber
> > of the bug.
> >
>

Revision history for this message
Ben Collins (ben-collins) wrote : Re: drop to initramfs with linux-image-2.6.22-1.5

It may have been the same message, but it's very likely it was preceded by others. The sr0 is your CD, not your hard drive.

Perhaps we can do some quick commands to get more info:

ls -l /dev/sd*
ls -l /dev/hd*
cat /proc/modules | grep ata
cat /proc/modules | grep ide

Please write out the output from these commands. Also, as a final command you can do "cat /proc/kmsg" and see if there's any messages about sd detection in there. Even just a snapshot with a digital camera of this screen would be useful.

Revision history for this message
Tux (peter-hoogkamer) wrote : Re: [Bug 112860] Re: drop to initramfs with linux-image-2.6.22-1.5

Hi Ben,

The attachements shows the output of the commands you gave me.
I will try to get the cat /proc/kmsg to you.
Do you have any clues? The kernel might not be able to detect the
virtual scsi hd from my vmware. It is not the first time I had
troubles with this when booting with new kernels. But it is strange
that the 2.6.20-15 kernel works ok the same vmware image.

2007/5/15, Ben Collins <email address hidden>:
> It may have been the same message, but it's very likely it was preceded
> by others. The sr0 is your CD, not your hard drive.
>
> Perhaps we can do some quick commands to get more info:
>
> ls -l /dev/sd*
> ls -l /dev/hd*
> cat /proc/modules | grep ata
> cat /proc/modules | grep ide
>
> Please write out the output from these commands. Also, as a final
> command you can do "cat /proc/kmsg" and see if there's any messages
> about sd detection in there. Even just a snapshot with a digital camera
> of this screen would be useful.
>
> --
> drop to initramfs with linux-image-2.6.22-1.5
> https://bugs.launchpad.net/bugs/112860
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
Ben Collins (ben-collins) wrote : Re: drop to initramfs with linux-image-2.6.22-1.5

Please paste into comment the output, or attach via the web interface. Attaching via email does not work.

Revision history for this message
Tux (peter-hoogkamer) wrote : Re: [Bug 112860] Re: drop to initramfs with linux-image-2.6.22-1.5

ls -l /dev/sd*
ls: /dev/sd*: No such file or directory
ls -l /dev/hd*
ls: /dev/hd*: No such file or directory
cat /proc/modules | grep ata
ata_piix 15748 0 - Live 0xd0824000
ata_generic 8452 0 - Live 0xd0838000
libata 127132 2 ata_piix,ata_generic, Live 0xd088c000
scsi_mod 147340 6 sg,sr_mod,mptspi,mptscsih,scsi_transport_spi,libata,
Live 0xd0867000
cat /proc/modules | grep ide

My cd-rom is an ide device on the vmware configuration and is
recognised as sr0, which I think is a scsi dev. While the hd is
configured as a scsi disk and is not seen by the kernel.

2007/5/15, Ben Collins <email address hidden>:
> Please paste into comment the output, or attach via the web interface.
> Attaching via email does not work.
>
> --
> drop to initramfs with linux-image-2.6.22-1.5
> https://bugs.launchpad.net/bugs/112860
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
Tux (peter-hoogkamer) wrote :

HI Ben,

I have been analysing the kmsg and saw some things regarding to scsi.
First of all I saw a
scsi1 1:0:0:0: CD-ROM -> this on is later seen as sr1
after that I see a
scsi2: ioc0 : LSI53C1030 - > Could this be my harddisk??
After this entry neighter of the scsi2 data is seen as an dev entry of
some kind. Shortly after this it drops to initramfs.

2007/5/16, Peter Hoogkamer <email address hidden>:
> ls -l /dev/sd*
> ls: /dev/sd*: No such file or directory
> ls -l /dev/hd*
> ls: /dev/hd*: No such file or directory
> cat /proc/modules | grep ata
> ata_piix 15748 0 - Live 0xd0824000
> ata_generic 8452 0 - Live 0xd0838000
> libata 127132 2 ata_piix,ata_generic, Live 0xd088c000
> scsi_mod 147340 6 sg,sr_mod,mptspi,mptscsih,scsi_transport_spi,libata,
> Live 0xd0867000
> cat /proc/modules | grep ide
>
> My cd-rom is an ide device on the vmware configuration and is
> recognised as sr0, which I think is a scsi dev. While the hd is
> configured as a scsi disk and is not seen by the kernel.
>
> 2007/5/15, Ben Collins <email address hidden>:
> > Please paste into comment the output, or attach via the web interface.
> > Attaching via email does not work.
> >
> > --
> > drop to initramfs with linux-image-2.6.22-1.5
> > https://bugs.launchpad.net/bugs/112860
> > You received this bug notification because you are a direct subscriber
> > of the bug.
> >
>

Revision history for this message
Ben Collins (ben-collins) wrote : Re: drop to initramfs with linux-image-2.6.22-1.5

Well, I tried to reproduce this, but I can't. Granted, I'm using Workstation 6, but I tried both the IDE and SCSI drives on this guest with latest kernel, and they just worked.

Going to have to rely on you for that dmesg output.

Revision history for this message
dean gaudet (deangaudet) wrote : Re: IDE drive not detected under VMware WS5

i haven't tried your kernel yet -- i just have my own 2.6.22-rc1 based off the feisty /boot/config-foo, and i'm seeing the same problem (vmware 5.5.2)... i've enabled NO_HZ but booting nohz=off doesn't change the result. it looks like the mpt fusion device is recognized but somehow /dev/sda isn't created. i've attached some serial console output.

Revision history for this message
Tux (peter-hoogkamer) wrote : Re: [Bug 112860] Re: IDE drive not detected under VMware WS5

Hi Dean,

That the same this with me. But mine stops after finding the pcnet32
card and then drops to initramfs. How do you extract this output? Then
I can send my dmesg output to the developers. They asked for it.
Thanks,

Peter

2007/5/17, dean gaudet <email address hidden>:
> i haven't tried your kernel yet -- i just have my own 2.6.22-rc1 based
> off the feisty /boot/config-foo, and i'm seeing the same problem (vmware
> 5.5.2)... i've enabled NO_HZ but booting nohz=off doesn't change the
> result. it looks like the mpt fusion device is recognized but somehow
> /dev/sda isn't created. i've attached some serial console output.
>
> ** Attachment added: "dmesg output"
> http://librarian.launchpad.net/7680495/console.txt
>
> --
> IDE drive not detected under VMware WS5
> https://bugs.launchpad.net/bugs/112860
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
dean gaudet (deangaudet) wrote :

stop your vm then go to vm / settings / hardware / serial... then set it
to output to a file.

then start your vm and in grub select the new kernel and press "e" to
edit, then cursor to your kernel command line and again "e" to edit, then
add "console=ttyS0,115200".

this will output to the file you specified, and you can upload the file.

-dean

On Thu, 17 May 2007, phoogkamer wrote:

> Hi Dean,
>
> That the same this with me. But mine stops after finding the pcnet32
> card and then drops to initramfs. How do you extract this output? Then
> I can send my dmesg output to the developers. They asked for it.

Revision history for this message
Tux (peter-hoogkamer) wrote : Re: [Bug 112860] Re: drop to initramfs with linux-image-2.6.22-1.5
Download full text (28.1 KiB)

GHi Ben,

With the help of somebody else (Dean Gaudet) I got to extract the
dmesg from my vm. He told me to redirect the output of the console to
a file. Here is my dmesg:

[ 0.000000] Linux version 2.6.22-3-generic (buildd@rothera) (gcc
version 4.1.3 20070429 (prerelease) (Ubuntu 4.1.2-5ubuntu3)) #1 SMP
Tue May 15 05:47:48 GMT 2007

[ 0.000000] BIOS-provided physical RAM map:

[ 0.000000] BIOS-e820: 0000000000000000 - 000000000009f800 (usable)

[ 0.000000] BIOS-e820: 000000000009f800 - 00000000000a0000 (reserved)

[ 0.000000] BIOS-e820: 00000000000ca000 - 00000000000cc000 (reserved)

[ 0.000000] BIOS-e820: 00000000000dc000 - 0000000000100000 (reserved)

[ 0.000000] BIOS-e820: 0000000000100000 - 000000000fef0000 (usable)

[ 0.000000] BIOS-e820: 000000000fef0000 - 000000000fefc000 (ACPI data)

[ 0.000000] BIOS-e820: 000000000fefc000 - 000000000ff00000 (ACPI NVS)

[ 0.000000] BIOS-e820: 000000000ff00000 - 0000000010000000 (usable)

[ 0.000000] BIOS-e820: 00000000fec00000 - 00000000fec10000 (reserved)

[ 0.000000] BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved)

[ 0.000000] BIOS-e820: 00000000fffe0000 - 0000000100000000 (reserved)

[ 0.000000] 0MB HIGHMEM available.

[ 0.000000] 256MB LOWMEM available.

[ 0.000000] found SMP MP-table at 000f6ce0

[ 0.000000] Zone PFN ranges:

[ 0.000000] DMA 0 -> 4096

[ 0.000000] Normal 4096 -> 65536

[ 0.000000] HighMem 65536 -> 65536

[ 0.000000] early_node_map[1] active PFN ranges

[ 0.000000] 0: 0 -> 65536

[ 0.000000] DMI present.

[ 0.000000] ACPI: RSDP 000F6C70, 0014 (r0 PTLTD )

[ 0.000000] ACPI: RSDT 0FEF7C02, 0030 (r1 PTLTD RSDT 6040000
LTP 0)

[ 0.000000] ACPI: FACP 0FEFBF14, 0074 (r1 INTEL 440BX 6040000
PTL F4240)

[ 0.000000] ACPI: DSDT 0FEF7C32, 42E2 (r1 PTLTD Custom 6040000
MSFT 100000D)

[ 0.000000] ACPI: FACS 0FEFCFC0, 0040

[ 0.000000] ACPI: APIC 0FEFBF88, 0050 (r1 PTLTD APIC 6040000
LTP 0)

[ 0.000000] ACPI: BOOT 0FEFBFD8, 0028 (r1 PTLTD $SBFTBL$ 6040000
LTP 1)

[ 0.000000] ACPI: PM-Timer IO Port: 0x1008

[ 0.000000] ACPI: LAPIC (acpi_id[0x00] lapic_id[0x00] enabled)

[ 0.000000] Processor #0 6:15 APIC version 17

[ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x00] high edge lint[0x1])

[ 0.000000] ACPI: IOAPIC (id[0x01] address[0xfec00000] gsi_base[0])

[ 0.000000] IOAPIC[0]: apic_id 1, version 17, address 0xfec00000, GSI 0-23

[ 0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 high edge)

[ 0.000000] Enabling APIC mode: Flat. Using 1 I/O APICs

[ 0.000000] Using ACPI (MADT) for SMP configuration information

[ 0.000000] Allocating PCI resources starting at 20000000 (gap:
10000000:eec00000)

[ 0.000000] Built 1 zonelists. Total pages: 65024

[ 0.000000] Kernel command line:
root=UUID=faba4d18-e6e2-435b-b507-2663b6696949 ro single
console=ttyS0,115200

[ 0.000000] Enabling fast FPU save and restore... done.

[ 0.000000] Enabling unmasked SIMD FPU exception support... done.

[ 0.000000] Initializing CPU#0

[ 0.000000] PID hash table en...

Revision history for this message
Ben Collins (ben-collins) wrote : Re: IDE drive not detected under VMware WS5

Peter, please, in the future, use the web interface to attach large files, else it really makes it hard to read the file, and surrounding comments.

I'm going to have to get WS5 obviously and see what I can figure out here.

Changed in linux-source-2.6.22:
importance: Undecided → High
status: Needs Info → In Progress
Revision history for this message
Tux (peter-hoogkamer) wrote : Re: [Bug 112860] Re: IDE drive not detected under VMware WS5

Ok, I am quite a newbe in this. I will use the webinterface in the future.

Peter

2007/5/17, Ben Collins <email address hidden>:
> Peter, please, in the future, use the web interface to attach large
> files, else it really makes it hard to read the file, and surrounding
> comments.
>
> I'm going to have to get WS5 obviously and see what I can figure out
> here.
>
> ** Changed in: linux-source-2.6.22 (Ubuntu)
> Importance: Undecided => High
> Status: Needs Info => In Progress
>
> --
> IDE drive not detected under VMware WS5
> https://bugs.launchpad.net/bugs/112860
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
Tux (peter-hoogkamer) wrote : Re: [Bug 112860] Re: MPT drive not detected under VMware WS5

Ben,

Could you tell me what an mpt driver is? Is it a specific virtual
driver from vmware?
Or something else?

Peter

2007/5/17, Ben Collins <email address hidden>:
> ** Summary changed:
>
> - IDE drive not detected under VMware WS5
> + MPT drive not detected under VMware WS5
>
> ** Bug watch added: Linux Kernel Bug Tracker #8492
> http://bugzilla.kernel.org/show_bug.cgi?id=8492
>
> ** Also affects: linux (upstream) via
> http://bugzilla.kernel.org/show_bug.cgi?id=8492
> Importance: Unknown
> Status: Unknown
>
> --
> MPT drive not detected under VMware WS5
> https://bugs.launchpad.net/bugs/112860
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
Ben Collins (ben-collins) wrote :

On Thu, 2007-05-17 at 17:41 +0000, phoogkamer wrote:
> Ben,
>
> Could you tell me what an mpt driver is? Is it a specific virtual
> driver from vmware?
> Or something else?

It's the scsi driver in the kernel that operates the virtual controller
from vmware. It's not specific to vmware at all.

Revision history for this message
Ben Collins (ben-collins) wrote :

I seem to have found a workaround:

echo options mptbase mpt_channel_mapping=1 | sudo tee /etc/modprobe.d/mpt_workaround
sudo update-initramfs -u

I'll post the results of my bisect to the kernel.org bug report.

Revision history for this message
Ben Collins (ben-collins) wrote :

Added a workaround for this to module-init-tools. Leaving bug open until a proper fix is in place.

Changed in linux-source-2.6.22:
assignee: ben-collins → ubuntu-kernel-team
importance: High → Low
status: In Progress → Confirmed
Revision history for this message
Ben Collins (ben-collins) wrote :

Uploaded module-init-tools with default of mpt_channel_mapping=1 for mptbase.

Changed in module-init-tools:
assignee: nobody → ben-collins
importance: Undecided → High
status: Unconfirmed → Fix Released
Changed in linux:
status: Unknown → Confirmed
Revision history for this message
Tux (peter-hoogkamer) wrote :

He Ben,

This works. Great! It boots perfectly into gutsy now. Do you have my
dmesg again to see what happens now?

Peter

2007/5/18, Bug Watch Updater <email address hidden>:
> ** Changed in: linux (upstream)
> Status: Unknown => Confirmed
>
> --
> MPT drive not detected under VMware WS5
> https://bugs.launchpad.net/bugs/112860
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
Ben Collins (ben-collins) wrote :

According to the kernel bug, this is a bug in vmware's emulation.

Changed in linux-source-2.6.22:
assignee: ubuntu-kernel-team → ben-collins
importance: Low → Medium
status: Confirmed → Rejected
Changed in linux:
status: Confirmed → Rejected
Revision history for this message
Fabio Massimo Di Nitto (fabbione) wrote :

Hi Ben,

be aware that this workaround breaks all .20 kernels that people can use as backup because the mpt_channel_mapping option does not exists in .20.

That also include breaking boots on all Niagara machines :)

Fabio

Changed in linux:
importance: Unknown → Medium
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Bug attachments

Remote bug watches

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