Thanks for quashing my hasty conclusion about the timing of the kernel crash. I now see that the crash occurred after the ramdisk stage succeeded. Probably, this may have something to do with memory paging to disk. I could boot to the shell, disable swapping, launch udev, resolvconf, dnsmasq, networking, submit the bug, update packages with aptitude.
I could not find a work-around. I cannot remember the version of my last successful image as the upgrade deleted it. Perhaps, I can look up my logs to find the version. My online search showed only the current unstable and stable packages, both of which failed to boot for different reasons. The earlier (oldstable?) package 2.6.26 bailed to a ramdisk shell prompt after failing to run a newer udev daemon. Perhaps, the package information for that earlier image could require a ramdisk make tool and/or ramdisk components relevant to that version. I sent screenshots of the stack trace just before receiving your email requesting them. I see the screen shots in the log of this bug report. Cheers. --
