Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!

Bug 75902

Summary: lvm on raid not supported
Product: Gentoo Linux Reporter: klavs klavsen <kl>
Component: [OLD] baselayoutAssignee: Gentoo's Team for Core System packages <base-system>
Severity: blocker    
Priority: High    
Version: unspecified   
Hardware: All   
OS: Linux   
Package list:
Runtime testing required: ---

Description klavs klavsen 2004-12-28 04:34:47 UTC
when creating lvm volumes on a software raid device, checkfs fails to initiate it properly and thus drops into "single-user-mode"..

Reproducible: Always
Steps to Reproduce:
1.create /dev/mdx
2.create lvm volumes on /dev/mdx
3.set lvm volumes up in fstab incl. mkfs etc.
4.reboot and watch it fail

Actual Results:  
boot fails

Expected Results:  
proper boot.

I fixed this easily, by simply moving the LVM support for /usr.. section to just
before the if [ -f /fastboot section.
Comment 1 SpanKY gentoo-dev 2004-12-28 05:43:28 UTC
you neglected to provide `emerge info` like the bug report page told you to

re-open once you have
Comment 2 klavs klavsen 2004-12-28 06:19:49 UTC
hmm - missed that bit.

# emerge info
Portage 2.0.51-r8 (default-linux/x86/2004.3, gcc-3.4.3, glibc-, 2.6.9-vs1.9.3 i686)
System uname: 2.6.9-vs1.9.3 i686 Intel(R) Pentium(R) 4 CPU 2.40GHz
Gentoo Base System version 1.4.16
Python:              dev-lang/python-2.3.4 [2.3.4 (#1, Dec 16 2004, 19:19:52)]
dev-lang/python:     2.3.4
sys-devel/autoconf:  2.59-r5
sys-devel/automake:  1.8.5-r1
sys-devel/libtool:   1.5.2-r7
CFLAGS="--param max-unrolled-insns=16 -fsingle-precision-constant -DNO_DEBUG -g0 -funroll-loops -ffast-math -finline-limit=4096 -march=pentium4 -fomit-frame-pointer -pipe -ftracer -O3 -fmerge-all-constants -fprefetch-loop-arrays-fstack-protector"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3/share/config /usr/share/config /var/bind /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-fabi-version=0 -ffunction-sections --param max-unrolled-insns=16 -fsingle-precision-constant -DNO_DEBUG-g0 -funroll-loops -ffast-math -finline-limit=4096 -march=pentium4 -fomit-frame-pointer -pipe -ftracer -O3 -fmerge-all-constants -fprefetch-loop-arrays -fpermissive -fno-enforce-eh-specs  -fstack-protector"
FEATURES="autoaddcvs autoconfig buildpkg ccache distlocks sandbox sfperms userpriv usersandbox"
USE="x86 acl apache2 apm arts authdaemond avi berkdb bitmap-fonts chroot crypt cups curl encode foomaticdb fortrangd gdbm gif gpm hardened imap innodb ipv6 jpeg kerberos ldap libwww mad maildir mcal memlimit mikmod motif mozcalendar moznoirc mpeg mysql mysqli ncurses nls nptl oggvorbis opengl oss pam pdflib perl php pic pie png postgres ppdspython quicktime readline sasl snmp spell ssl svga tcpd tiff truetype usb vda vhosts xml2 xmms xv zlib"
Comment 3 SpanKY gentoo-dev 2004-12-28 07:02:10 UTC
is this Bug 75394 ?
Comment 4 klavs klavsen 2004-12-28 07:09:33 UTC
No. lvm and raid is NOT modules - it's all compiled in. It's simply a matter of the order in which checkfs checks for lvm and raid devices. It checks for lvm BEFORE raid, but should do it the other way around - as you don't put raid on lvm, but do put lvm on raid devices.
Comment 5 Kenton Groombridge 2004-12-31 12:11:33 UTC
This sounds related to my problem.  I have a raid0 configuration.  My /dev directory has the following entries relevant for my raid0 config:

#My two SATA disks

#My raid0 devices

When I had sys-apps/baselayout-1.11.7-r2 all worked fine, and the above devices are present in the dev directory.

After emerging sys-apps/baselayout-1.11.8 and rebooting, mdadm failed.

All the above devices with the exception of /dev/md0 were missing.  Recreated them using mknod in single user mode and then went to runlevel 3 and all was fine until the next reboot where the devices were again deleted.  I think it is related to the file /etc/init.d/ (it looks like it deletes /dev entries).

I masked sys-apps/baselayout-1.11.8 and went back to sys-apps/baselayout-1.11.7-r2 and all is fine.

emerge info
Portage 2.0.51-r8 (default-linux/x86/2004.0, gcc-3.4.3, glibc-, 2.6.10-ck1 i686)
System uname: 2.6.10-ck1 i686 Unknown CPU Typ
Gentoo Base System version 1.6.7
Python:              dev-lang/python-2.3.4 [2.3.4 (#1, Oct 24 2004, 12:02:21)]
dev-lang/python:     2.3.4
sys-devel/autoconf:  2.13, 2.59-r6
sys-devel/automake:  1.6.3, 1.5, 1.8.5-r2, 1.7.9, 1.4_p6, 1.9.3
sys-devel/libtool:   1.5.10-r2
CFLAGS="-O2 -pipe -march=athlon-xp -fomit-frame-pointer"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.3/env /usr/kde/3.3/share/config /usr/kde/3.3/shutdown /usr/kde/3/share/config /usr/lib/X11/xkb /usr/share/config /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O2 -pipe -march=athlon-xp -fomit-frame-pointer"
FEATURES="autoaddcvs autoconfig ccache distlocks sandbox sfperms"
USE="x86 3dnow S3TC X acl acpi acpi4linux alsa apache2 apm avi berkdb bitmap-fonts cddb cdr crypt cups dga divx4linux dvd dvdr dvdread emoticon encode fam ffmpeg flac foomaticdb fortran freetype gdbm gif gimp gimpprint gpm gstreamer gtk2 ieee1394 imagemagick imlib ipv6 java joystick jpeg kde libwww mad mikmod mmx motif mozsvg mpeg mpeg4 mplayer ncurses network nls nptl nptlonly nvidia oggvorbis openal opengl opens pam pdflib perl png python qt quicktime readline scanner sdl slang spell sse ssl svg svga tcltk tcpd tiff transcode truetype type1 unicode usb v4l2 wmf xine xml2 xmms xpm xv xvid zlib zvbi"
Comment 6 klavs klavsen 2005-01-03 00:00:43 UTC
Actually the missing mdx devices is not the same problem as the fact that the current Gentoo /etc/init.d/checkfs does NOT support lvm devices on raid devices.

I too had the problem of everything but /dev/md0 missing - and wrote about that, in some other bug, that I strangely can't find anymore, but the problem is with udev - which does NOT create the devices as it should - and it surfaced, because I can't run with the "tarball"-fix - when on raid (it dies when unpacking because it creates a ramfs for it, which is empty and then dies when trying to untar, due to missing /dev/null on the ramfs which it just created). I "fixed" that, by putting the creation of the /dev/mdx devices into checkfs for now.
Comment 7 Sven Wegener gentoo-dev 2005-01-03 07:38:46 UTC
The order of LVM and raid startup in checkfs has already been changed so that
raid is started first. It's available in unstable baselayouts but not yet in a
stable one. For the md? bug see bug #64750 that's resolved as UPSTREAM.

*** This bug has been marked as a duplicate of 37418 ***