How to diagnose failed install?

[dvg@jpl.nasa.gov]

I am using Sali because SystemImager BOEL binaries are very out of date and UYOK doesn't work so well.

What I did was drop your initrd.img and kernel files onto my SystemImager server in /usr/share/systemimager/boot/x86_64/<IMAGE NAME> and run si_mkautoinstallcd as always.
This results in a CD that actually boots instead of causing a kernel panic! And it finds my e1000e network card, too! Wonderful! This is much farther than SystemImager can get me with this hardware.

Sali then goes out to the image server and finds the boot script. It says: "Installing image (using the old method). Starting old installer file." It then partitions my disk drive (not entirely correctly according to the way SystemImager would do it) and says:

"An error has occurred, killing off running processes." before printing out "Your autoinstall has failed."

It drops me at a prompt. I have no idea how to diagnose where it failed. How can I see the logs or obtain a more verbose output?

[bas]

Which version do you use and just to make sure you have a console? You can also login on the node with ssh, see:

• ​https://subtrac.sara.nl/oss/sali/wiki/SaliHowto#Test

There must be log files in:

• /var/log

In the log files there must be mentioned where it failed or fund out the error message.