Red Hat Bugzilla – Bug 55681
loading of scsi module fails at boot time
Last modified: 2014-03-16 22:24:14 EDT
From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.2.1) Gecko/20010901 Description of problem: The following message appears on the console, "improperly terminated string argument for ncr53c8xx", and the kernel halts its booting process. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1.Have a system with a scsi controler which needs the ncr53c8xx driver 2.upgrade this system from 6.2 to 7.2 3.watch the boot fail... Additional info: I believe the bug is due to the fact that under Red Hat 6.2, the /etc/modules.conf file has an options entry for the ncr53c8xx.o module. When I commented out this options entry, re-ran the mkinitrd for the kernel I boot, then I was able to boot up fine.
Created attachment 36361 [details] modules.conf config file for problemed system
I've attached the /etc/modules.conf file. Notice that I've commented out the options entry for the ncr53c8xx.o module. This is what I needed to do in order to get the kernel to boot properly (2.4.9-13smp) after doing an mkinitrd. Also, I had the same problem with kernels 2.4.7-10smp.
This is more an upgrade bug; such module options shouldn't be kept during major kernel upgrades (from 2.2 to 2.4)
Anaconda doesn't really mess with this... reassigning to kudzu
kudzu doesn't really have an infrastructure to keep track of what options are good vs. bad for what particular module in what particular kernel version. I'm not sure there's really any way to fix this unilaterally.