Kernel panic with 2.4.9 as patched from "martel" on SuSE FTP
Good afternoon! I've just built several variants of a 2.4.9 kernel for SuSE 7.2, and am having the same kernel panic no matter what I try in the configuration options. I started with a generic kernel tree from ftp.kernel.org, and applied (with no error messages) all the patches for 2.4.9 from ftp://ftp.suse.com/...../martel/next/ on the SuSE FTP site. Everything seemed fine, and the build went smoothly. In the build, I started with the default SuSE configuration file from /boot, did a "make oldconfig" to bring it up to 2.4.9, and then ran "make xconfig" to remove some stuff like Amateur Radio, ISDN, and MTD support, that I don't plan to use. I ran make dep, make bzImage, make modules, and make modules_install, in that order, after first making sure that the directory /lib/modules/2.4.9 was empty. Everything seemed to work fine. I copied the bzImage file to a new filename under /boot, configured /etc/lilo.conf as needed, then ran lilo. When I try to boot the new kernel, I get a panic with the following rather cryptic information: Process init (pid: 1, stackpage=c1409000) Stack: <<omitted for brevity>> Call Trace: <<omitted for brevity>> Code: 0f 0b eb 26 8d b6 00 00 00 00 83 7b 04 00 75 1d 68 b6 00 00 Kernel panic: Attempted to kill init! I know what init does, but I have no idea how it came to be that some process tried to kill it. Rebooting from the original kernel still works, so my system isn't hosed. I also tried rebooting from the rescue CD, immediately after the crash, to recover any useful log entries, but alas there is nothing being logged (too early in the boot process, I guess). The console seems to indicate that the crash is happening just after Reiserfs is loaded, but it scrolls by too fast for me to be sure. Has anyone seen this before, and if so, what should I be checking in my kernel config to isolate the problem? Thanks for any help someone can offer. Scott -- ------------------------------------------------------------ "I don't mind Microsoft making money. I mind them having a bad operating system." -- Linus Torvalds, in the NY Times Scott Courtney http://www.4th.com/
Mantel's kernels have had problems for a couple of weeks. I tried them and got oopses and other problems. I'd wait till 2.2.10 is released. Rafael
Well, the kernels in Huberts directory are for testing only and not for production. If a production certified kernel is needed..please look under your version on the ftp site for the certified 2.4.7 kernel. I would bet that Hubert will will have a stable..patched 2.4.9/2.4.10 sometime in the near future. * Rafael E. Herrera (raffo@neuronet.pitt.edu) [010831 15:43]: ->Mantel's kernels have had problems for a couple of weeks. I tried them ->and got oopses and other problems. I'd wait till 2.2.10 is released. -> ->Rafael -> ->-- ->To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com ->For additional commands send e-mail to suse-linux-e-help@suse.com ->Also check the FAQ at http://www.suse.com/support/faq and the ->archives at http://lists.suse.com -> -- Ben Rosenberg mailto:ben@whack.org cat CE|ME|NT|XP > /dev/null
On Friday 31 August 2001 06:55 pm, Ben Rosenberg wrote:
Well, the kernels in Huberts directory are for testing only and not for production. If a production certified kernel is needed..please look under your version on the ftp site for the certified 2.4.7 kernel.
I would bet that Hubert will will have a stable..patched 2.4.9/2.4.10 sometime in the near future.
Understood...I'm intentionally going to the bleeding edge in hopes of helping to find bugs, and because I have more fun with Linux that way. In both of these regards, I'm succeeding. :-) Seriously, at the moment I have the luxury of a brand-new laptop that isn't yet set up for production, so if anything is going to go bump in the night, it won't cost me important data. This kernel is *not* going on my desktop machine yet. At the same time, I thought it made sense to ask if someone else had already seen this before reporting it to Mr. Mantel. Apparently I've found a new one, so I sent him an e-mail to let him know. BTW...I now have some further information. It is *not* the Reiserfs failing, as I had suspected. The real error was scrolling off the screen, but this time I caught it! It turns out it's a PCI bus error at line 180 in .../include/asm/pci.h. The error report is "invalid operand: 000". At least I'm getting a useful diagnostic now. Scott -- ------------------------------------------------------------ "I don't mind Microsoft making money. I mind them having a bad operating system." -- Linus Torvalds, in the NY Times Scott Courtney http://www.4th.com/
On Friday 31 August 2001 06:40 pm, Rafael E. Herrera wrote:
Mantel's kernels have had problems for a couple of weeks. I tried them and got oopses and other problems. I'd wait till 2.2.10 is released.
Please see my other note...this is not (yet) a production machine, just a new laptop that I'm setting up and currently no important data on it. Scott -- ------------------------------------------------------------ "I don't mind Microsoft making money. I mind them having a bad operating system." -- Linus Torvalds, in the NY Times Scott Courtney http://www.4th.com/
participants (3)
-
Ben Rosenberg -
Rafael E. Herrera -
Scott Courtney