Since I updated my baselayout from 1.11.something to 1.12.4-r2 (and now also -r3, -r4 and -r6) I get a segmentation fault in /sbin/rc on boot. After the segfault, when I press control-d to continue, the system restarts and then boots up fine with no hitch whatsoever. If I enter the root password after the segfault and then execute reboot, next boot the system segfaults again exactly the same way. The output after ctrl-d is different from logging in and rebooting: ctrl-d: "* Unmounting filesystems * Rebooting" login & reboot: "INIT: Switching to runlevel: 6 INIT: Sending processes the TERM signal * Unmounting filesystems ... * Remounting remaining filesystems readonly ..." Since I can't copy/paste at that point yet, nor take screenshots, I've taken a picture of the screen (http://someserver.mine.nu/~tomas/rc-segfault.jpg). (Note: where it says "Segmentatie fout", this is the Dutch translation for "Segmentation fault".) A patch (http://dev.gentoo.org/~uberlord/baselayout/rc.patch) was suggested but did not help. It did, however, change the output slightly: above "One or more critical startup scripts failed to start!" I now get "Failed to start /etc/init.d/modules". Someone else with the same problem found out that when adding "rw" to the kernel commandline, the problem does not occur anymore. Additional info: kernel 2.6.16-gentoo-r13 with the reiser4 patches from namesys.com root filesystem is reiser4 reiser4progs 1.0.5 I use BIOS raid (using dmraid). I use the initrd created by "genkernel --dmraid initrd". My current kernel commandline: kernel /kernel-2.6.16-gentoo-r13 root=/dev/ram0 real_root=/dev/mapper/isw_cacaigadei_twins6 init=/linuxrc dodmraid initrd /initramfs-genkernel-x86-2.6.16-gentoo-r13
You are so going to hate us, but can you replicate this on a pure gentoo-sources or vanilla-sources kernel without any patches, such as reiserfs4? So please replicate on ext2/3, reiserfs, xfs, jfs or some other supported filesytem.
(In reply to comment #1) > You are so going to hate us, but can you replicate this on a pure > gentoo-sources or vanilla-sources kernel without any patches, such as > reiserfs4? > > So please replicate on ext2/3, reiserfs, xfs, jfs or some other supported > filesytem. > I also have the same problem using Reiser4. I can move my system on a Reiserfs partition this evening and report back here.
(In reply to comment #1) > You are so going to hate us, but can you replicate this on a pure > gentoo-sources or vanilla-sources kernel without any patches, such as > reiserfs4? > > So please replicate on ext2/3, reiserfs, xfs, jfs or some other supported > filesytem. > I have exams on the 21st and 30th of August, so I have time to test patches and try different kernel commandlines, but not to move over entire partitions :) (I also don't have the free space required.)
(In reply to comment #2) > (In reply to comment #1) > > You are so going to hate us, but can you replicate this on a pure > > gentoo-sources or vanilla-sources kernel without any patches, such as > > reiserfs4? > > > > So please replicate on ext2/3, reiserfs, xfs, jfs or some other supported > > filesytem. > > > I also have the same problem using Reiser4. I can move my system on a Reiserfs > partition this evening and report back here. > So... I have my system on a Reiserfs partition and this bug does not appear :/ It seems to be a Reiser4 bug.
I have exacly the same problem, but I use ReiserFS, not Reiser4. I also only started seeing this after a recent baselayout upgrade. I'm currently running the 2.6.16-gentoo-r9 kernel.
*** Bug 147274 has been marked as a duplicate of this bug. ***
(In reply to comment #4) > > I also have the same problem using Reiser4. I can move my system on a Reiserfs > > partition this evening and report back here. > > > So... I have my system on a Reiserfs partition and this bug does not appear :/ > It seems to be a Reiser4 bug. > +1. When I copied / over to reiserfs (3.6) the error did not occur. Namesys is on it and thinks it could be an r4 issue (although no one really knows what or why at the moment). ITMT, downgrading baselayout to 1.11.15-r3 works for me.
My system does not have Reiser4 so its not a direct issue with only that filesystem.
fwiw same result with baselayout 1.12.5
what gcc version are you using? my prob system has 3.3.6 on it.
It seems this bug is being discussed on the reiserfs mailing list. The start of the thread is http://marc.theaimsgroup.com/?l=reiserfs&m=115790773926650&w=2 . They have not found a solution either, but found out that it could have to do with how the root fs is unmounted. Executing /bin/mount -a -o remount,ro &>/dev/null /sbin/reboot -f makes the problem NOT occur on the next boot. Reboot using reboot.sh or halt.sh makes the problem occur next boot. I have had this problem ever since baselayout 1.12. Various reports indicate moving back to baselayout 1.11 fixes the problem. I have this problem using gcc 3.4.6 + glibc 2.3.6 (profile 2006.0) and gcc 4.1.1 + glibc 2.4 (profile 2006.1/desktop).
sounds more and more like a reiser4 bug: get it fixed upstream, we dont want to hear about "bugs in baselayout"
Given that two people have reported this happening under ReiserFS (*not* Reiser4), and ReiserFS is supposed to be supported under Gentoo, it's not at all clear this is a Reiser4 problem. Isn't it therefore premature to mark it "invalid"? Also, most people are reporting that the problem suddenly started happening after a baselayout update. Something must have changed in baselayout to expose the bug, even if it *is* a filesystem bug in both ReiserFS and Reiser4. Up to now no one has adqeuately explained why everything worked fine in the old baselayout.
(In reply to comment #10) > what gcc version are you using? my prob system has 3.3.6 on it. > 4.1.1 profile 2006.1 x86/desktop. I also found a couple of instances of daemons hanging, which you first reported! I also noted that when that happened and I could not shut down, the error of segfaults did NOT occur on the next boot. I still think something is wacky with the shutdown sequence which (I know vapier does not want to hear it) makes r4 barf. However, I am NOT convinced it is solely an r4 issue at this point.
On all my system I don't use ReiserFS v4 so its defo not a r4 only problem!
(In reply to comment #15) > On all my system I don't use ReiserFS v4 so its defo not a r4 only problem! > well then, maybe the gentoo folks ought to consider taking their thumbs out of..... and reopen this bug. Pretending it's invalid does not seem to be the proper way to proceed now, does it?
(In reply to comment #16) > well then, maybe the gentoo folks ought to consider taking their thumbs out > of..... and reopen this bug. Pretending it's invalid does not seem to be the > proper way to proceed now, does it? > Please keep this discussion/troubleshooting civil. I have great respect for the Gentoo devs and think you should, too. (In reply to comment #15) > On all my system I don't use ReiserFS v4 so its defo not a r4 only problem! > Could you please describe your problem and when and how it occurs (on boot? while running? crashes? things that Just Don't Work?) in great detail? I have a hunch that our problem IS reiser4-related, and you just a have a different problem.
