failure with 2.6.15 on vmware 5.5 (scsi disk)

Do you have to manually select the storage driver during install? Can you send
me lsmod and dmesg output from the install after it detects your harddrives?

(In reply to comment #1)
> Do you have to manually select the storage driver during install? Can you send
> me lsmod and dmesg output from the install after it detects your harddrives?

I don't remember I was asked a question about a storage driver. Should this have
been done through debconf during kernel install ?

For lsmod and dmesg, I'm a bit stuck for doing this in a busybox environment.
What's the recommended solution for this ? Otherwise, I will give a try with a
live cd later.

*** Bug 27223 has been marked as a duplicate of this bug. ***

Can I get the output of "lspci -vv" and lspci -vvn"?

The scsi vmware scsi device id is 0x1000/0x0030 mapping to the mptspi driver.

I managed to get it boot without error by adding mptspi AND sd_mod to
/etc/mkinitramfs/modules

I am wondering if this is not the same problem that was solved on Debian:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=341162

Something's got to be wrong with udev here. The mptspi has the proper
module-device-table entry for this device. It is also listed in modules.pcimap.
Somebody is not loading the module.

I don't think it is udev's fault. The mptspi module was not loaded because it
was not in the initrd image.

But when you force to add it to the initrd, and with the sd_mod too (using
/dev/mkinitramfs/modules for instance), udev loads it fine and the system boots
as usual without errors.

This doesn't look like a udev bug to me, the module loads fine in my tests IF
IT'S PRESENT -- it's just not being put in the initramfs.

Adding mptspi to the manual_add_modules list (under SCSI) in /usr/share/initramfs-tools/hook-functions solved this for me. Adam, could you do that?

Adam, initramfs-tools bug

Fixed in initramfs-tools_0.40ubuntu22, just uploaded. Should hit mirrors after the usual archive processing delays.