> So, I recently built (and rebuilt) svn-20060220. Both times resulted in > the same kernel panic. Any tips as to how to interpret the panic so as > to point to a specific cause would be much appreciated. > > The first time through, I was working (and installing to) a single ext3 > partition; that was the only deviation from the book. When I got the > kernel panic, I figured I had made a dumb mistake like misconfiguring > the kernel (perhaps I didn't compile in ext3) or somesuch. I had done > the final work on it late at night, and perhaps had been sloppy. > > The second time through, I was much more careful. I started completely > from scratch - new disk, newly formatted (ext3), built toolchain from > scratch, etc. I (foolishly, in retrospect) did not run *every* make > check, but I did on several (ie, glibc, gcc) and I did do the check > compiles to make sure I wasn't linking against tools, that SSP was > working correctly, etc. In particular, in configuring the kernel, I > left it at all the defaults (ext3 appeared to be compiled in, not as a > module, by default) except the grsec and PaX options. Same panic. I > then did a cp -pr to an ext2 partition, remounted dev proc etc, chrooted > in and reran grub. Rebooted, same result. > > One thing I thought of was hotplug. I read that it was needed to > populate /dev. I assumed that that meant for new devices, after boot > (ie, usb), but... In make menuconfig, the section for hotplug listed > "---" instead of the "<*>" I would have expected, but looking at the > .config file showed hotplug set to y, so that seems OK (?) > > I can't see all of the data from the panic, because it scrolls of the > screen. Basically, I see the tail end of a column of hex numbers, then > some other data, as follows: > > [<00000000>] > Code: 57 56 53 83 ec 2c 8b 44 24 44 85 c0 0f 88 e5 04 00 00 8b 74 24 40 > 8b 44 24 44 8d 6c 06 ff 89 f0 83 e8 01 39 e8 77 43 8b 4c 24 48 <80> 39 > 00 74 25 0f b6 01 3c 25 74 42 39 ee 77 06 8b 4c 24 > <0>Kernel panic - not syncing: Attempted to kill init! > > I googled the last phrase, but came up mostly with references to a > problem in Red Hat that didn't seem relevant - but I admit that I > probably don't have the knowledge to know what's relevant and what's not... > > In particular, I'd appreciate not only ideas about what the problem > might be, but advice on how I could debug it myself. Meanwhile I'll be > embarking on yet another build, with 'make check' at every point along > the way... Thanks! > > -jps >
before goiung any further, stop. Download something like the ultimate boot cd: http://www.ultimatebootcd.com/ And run memtest86+ (any/all of their memory testers depending on yopur level of paranoia) I have seen some very strange things happen with compilation and memory issues. And yes, I have successfully compiled an entire scratch system on a machine with bad ram..when it comes to runtime or kernel compile, many strings things happen. -- http://linuxfromscratch.org/mailman/listinfo/hlfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
