At boot i get some weird messages : RC_DEPEND_TREE[number]= ... or RC_DEPEND_TREE[number+number]= ... Sometimes when i get this messages at boot, my system works 'fine' and sometimes it doesn't; this is like a random ! I found that unplugging my pcmcia wifi card during the boot fixes the problem, but i never had that before with the wifi card plugged in ! I dont have the whole error message because i get it during the boot ! Reproducible: Sometimes Steps to Reproduce: 1. Boot the system 2. 3. Actual Results: .... RC_DEPEND_TREE[132+10]=1075883249 RC_DEPEND_TREE[143]=local RC_DEPEND_TREE[143+2]= RC_DEPEND_TREE[143+3]=bootsplash RC_DEPEND_TREE[143+4]= RC_DEPEND_TREE[143+5]= RC_DEPEND_TREE[143+6]= RC_DEPEND_TREE[143+7]=acpid alsasound bitlbeed bootmisc canna checkfs checkroot clock consolefont cpufreqd crypto-loop cupsd dictd distccd domainname esound fam famd fetchmail foldingathome gkrellmd gpm hdparm hostname hotplug i8k iptables keymaps localmount mldonkey modules named net.eth0 net.eth1 net.lo netmount nfs nfsmount nscd ntp-client ntp286+8") /sbin/rc: line 1062: syslog-ng emerge info : Portage 2.0.50-r6 (default-x86-1.4, gcc-3.3.3, glibc-2.3.3_pre20040207-r0, 2.6.5) ================================================================= System uname: 2.6.5 i686 Mobile Intel(R) Pentium(R) 4 CPU 3.06GHz Gentoo Base System version 1.4.9 distcc 2.13 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [disabled] ccache version 2.3 [enabled] Autoconf: sys-devel/autoconf-2.59-r3 Automake: sys-devel/automake-1.8.3 ACCEPT_KEYWORDS="x86 ~x86" AUTOCLEAN="yes" CFLAGS=" -g -pipe -march=pentium4 -mcpu=pentium4 -mfpmath=sse,387 -O2 -frename-registers -fomit-frame-pointer -ffast-math -fforce-addr" CHOST="i686-pc-linux-gnu" COMPILER="gcc3" CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3/share/config /usr/lib/mozilla/defaults/pref /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/bind /var/qmail/control"CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS=" -g -pipe -march=pentium4 -mcpu=pentium4 -mfpmath=sse,387 -O2 -frename-registers -fomit-frame-pointer -ffast-math -fforce-addr" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs ccache sandbox" GENTOO_MIRRORS="ftp://sunsite.informatik.rwth-aachen.de/pub/Linux/gentoo ftp://linux.rz.ruhr-uni-bochum.de/gentoo-mirror http://www.ibiblio.org/gentoo" MAKEOPTS="-j4" PKGDIR="/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage" SYNC="rsync://rsync.fr.gentoo.org/gentoo-portage" USE="3ds X X509 aalib acpi acpi4linux alsa ardour-ksi avi berkdb bonobo canna cdr chroot cjk crypt cscope dga directfb dv dvb dvd encode esd evo faad fam fbcon fbdev ffmpeg fftw flac flash freetype gb gd gdbm gif gnome gphoto2 gpm gstreamer gtk gtk2 gtkhtml guile imagemagick imap imlib jabber java javascript jbig jpeg libwww mad matroska mikmod mldonkeypango motif mozilla moznoirc moznomail mozsvg mpeg mpeg4 msn music ncurses nls objc ocaml offensive oggvorbis openal opengl openssh oscar oss pam pcmcia pdflib perl php png pnp python quicktime readline rogue ruby samba sdl slang spell sse ssl svg svga tcltk tcpd tetex tiff transcode truetype unicode usb videos x86 xfs xine xinerama xml2 xmms xosd xv xvid yahoo zlib" rc-status : Runlevel: default local [ started ] net.eth0 [ started ] sshd [ started ] syslog-ng [ started ] pcmcia [ started ] net.eth1 [ started ] xinetd [ started ] domainname [ started ] bootsplash [ started ] gpm [ started ] canna [ started ] xdm [ started ]
I have a similar problem. It appeared after I did a world update last weekend. I investigated and found that /var/lib/init.d/deptree is corrupted. Here are two examples: RC_DEPEND_TREE[154+9]= RC_DEPEND_TREE[154+10]="1082752405" RC_DEPEND_TREE[165]="slurpd" RC_DEPEND_TREE[16rc_type_ineed=2 rc_type_needsme=3 rc_type_iuse=4 rc_type_usesme=5 rc_type_ibefore=6 rc_type_iafter=7 rc_type_broken=8 .... RC_DEPEND_TREE[319]="checkfs" RC_DEPEND_TREE[319+2]="checkroot modules" RC_DEPEND_TREE[319+3]="localmount" RC_DEPEND_TREE[319+4]= RC_DEPEND_TREE[319+5]= RC_DEPEND_TRE5+2]="net" RC_DEPEND_TREE[165+3]= RC_DEPEND_TREE[165+4]= RC_DEPEND_TREE[165+5]= RC_DEPEND_TREE[165+6]="local Further investigation showed that /sbin/depscan.sh, which creates deptree, is being called twice at almost the same time. One call is associated with root, the other with /etc/hotplug. I don't know more than this at present. Bug 47111 refers to a similar problem except that depcache was affected rather than deptree. Some information about my system: Portage 2.0.50-r6 (default-x86-1.4, gcc-3.3.2, glibc-2.3.2-r9, 2.6.5-gentoo) ================================================================= System uname: 2.6.5-gentoo i686 Intel(R) Pentium(R) 4 CPU 3.00GHz Gentoo Base System version 1.4.10 Autoconf: sys-devel/autoconf-2.58-r1 Automake: sys-devel/automake-1.8.3 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" I am using baselayout-1.8.12
Still further investigation revealed that the second depscan.sh was triggered by an ethernet module -- sk98lin in my system -- that I had put in /etc/modules.autoload.d/kernel-2.X. I deleted the module from there and put it in /etc/conf.d/local.start (as modprobe sk98lin). On my system, scripts in the latter location are executed after the scripts in the former location. Things now seem to be working properly. I suspect that this will not be a satisfactory solution for many people because it requires that they understand the obscure problems that can occur in the booting process. A better solution might be to have depscan.sh check to see if there are any files in the /var/lib/init.d directory. If there are, then another depscan.sh process has already started generating the files. For those who want to run depscan.sh manually, depscan could print out a message saying that files exist in the /var/lib/init.d directory and they need to be deleted before new files can be generated.
I had to insert a "sleep 1" command between "modprobe sk98lin" and "ifconfig eth0 192.168.XXX.XXX" in /etc/conf.d/local.start. Without it, ifconfig would use 192.168.0.2 as the eth0 address rather than 192.168.XXX.XXX.
I added the following hack to /sbin/functions.sh #------------------------- get_lock () { local filename="$1"; if [ -z $filename ]; then echo "Empty file" > 2 fi filename=/tmp/$filename touch $filename /bin/ln $filename $filename.lk 2>/dev/null # try to grab the lock while [ $? -ne 0 ] # failed... do sleep 1 # wait a while and try again. /bin/ln $filename $filename.lk 2>/dev/null done } rel_lock () { local filename="$1"; filename=/tmp/$filename if [ -z $filename ]; then echo "empty file" > 2 fi /bin/rm $filename.lk } #--------------------- And then added the following to /sbin/depscan.sh:10 get_lock .depscan #Get lock And /sbin/depscan.sh:55 rel_lock .depscan #release lock
it looks like kernel 2.6.6 and baselayout 1.9.1 fixed that for me !
This problem recently showed up on my system as well, starting with an emerge -u world. baselayout-1.9.4 gentoo-sources-2.4.25-r3
Corroboration: My system also has the corrupt /var/lib/init.d/deptree, as mentioned by Ivar Ylvisaker in comments 1 and 2. Commenting out my tulip ethernet driver from /etc/modules.autoload got rid of the RC_DEPEND_TREE messages at boot, so I'm thinking my system is behaving as his did. I wonder, is this triggered only by certain ethernet drivers when compiled as modules, or perhaps just more common on certain hardware (like my dual CPU system)? As a workaround, I rebuilt my kernel with the tulip ethernet driver linked directly in (rather than as a module), and removed tulip from /etc/modules.autoload. This solved my problem, but it seems to me that the problem shouldn't exist in the first place. Thanks to Ivar Ylvisaker for posting the information I needed.
Created attachment 33428 [details] Corrupt /var/lib/init.d/deptree
I too an having the same problem with a corrupt /var/lib/init.d/deptree at boot time and have attached the corrupt file. I'm including this just in case it adds any additional information you may find useful. Portage 2.0.50-r8 (default-x86-1.4, gcc-3.3.3, glibc-2.3.3.20040420-r0, 2.4.25-gentoo-r4) ================================================================= System uname: 2.4.25-gentoo-r4 i686 Mobile Intel(R) Pentium(R) 4 - M CPU 1.80GHz Gentoo Base System version 1.4.16 Autoconf: sys-devel/autoconf-2.59-r3 Automake: sys-devel/automake-1.8.3 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CFLAGS="-march=pentium4 -Os -fforce-addr -fomit-frame-pointer -pipe" CHOST="i686-pc-linux-gnu" COMPILER="gcc3" CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3.2/share/config /usr/kde/3/share/config /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS="-march=pentium4 -Os -fforce-addr -fomit-frame-pointer -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs ccache sandbox" GENTOO_MIRRORS="http://gentoo.oregonstate.edu http://distro.ibiblio.org/pub/Linux/distributions/gentoo" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="" SYNC="rsync://rsync.namerica.gentoo.org/gentoo-portage" USE="X apm arts avi berkdb cdr crypt cscope cups encode esd foomaticdb gdbm gif gpm gtk gtk2 imlib java jpeg kde libg++ libwww lirc mad mbox mikmod mmx motif mpeg mpeg3 ncurses nls offensive oggvorbis opengl oss pam pcmcia pdflib perl png pnp python qt quicktime readline sdl slang spell ssl svga tcltk tcpd tetex threads truetype trusted usb x86 xml2 xmms xv zlib" I'm using: sys-apps/baselayout-1.9.4-r2 sys-kernel/gentoo-sources-2.4.25-r4 And here's the state of what gets started at boot time: aumix | default bootmisc | boot checkfs | boot checkroot | boot clock | boot consolefont | boot crypto-loop | cupsd | domainname | esound | fam | famd | gkrellmd | gpm | default hdparm | default hostname | boot hotplug | default ifplugd | iptables | keymaps | boot lircd | lisa | local | default nonetwork localmount | boot modules | boot net.eth0 | default net.lo | boot netmount | default nscd | ntp-client | default ntpd | default numlock | pcmcia | portmap | reslisa | rmnologin | boot rsyncd | samba | serial | boot shorewall | default sshd | switch | synclient | sysklogd | boot urandom | boot vixie-cron | default winbind | xdm | default xfs | xinetd | default Please let me know if there's any additional information I can provide that you can use to solve this problem.
Yesterday, I upgraded to the 2.6.7 kernel (skipping 2.6.6). When I start my ethernet module from the /etc/modules.autoload.d directory, the RC_DEPEND_TREE error messages are not longer seen in the boot messages. However, when I look at the /var/lib/init.d/deptree file, I still see evidence that two /sbin/depscan.sh processes are running. With the 2.6.5 kernel, the deptree file contained two overlapping copies of the desired deptree file, the second mangling the first. With the new kernel, the deptree file still contains two copies but they follow one another. Either (or both?) can be read successfully by the boot software. However, this suggests that if my hardware or software changed (e.g., a faster or slower or different CPU or more or fewer or rewritten modules, etc.), the errors could reappear. Putting my ethernet module in /etc/conf.d/local.start still eliminates the problem for me. But is this assured if other things change?
I've two ethernet cards. Today I unmerged coldplug and put the Ethernet kernel modules e1000 and 8139too into /etc/modules.autoload.d/kernel-2.6. Now I've the same problem. Depending on if I press a key during boot I see different error messages. So this is definitly a race condition. ... hmmm, maybe this explains some other strange errors (e.g. bug # 71025) I see on my system for quite a while ...
Can you try depscan.patch from bug 70009? it should fix the problem. just: cd / patch -p0 < /tmp/depscan.patch
I am experiencing this bug with baselayout 1.9.4-r6. I'm sure someone's working out something right now, meanwhile I need a system that boots reliably. So here my working bugfix: gendepends.patch. Basically, this is the same fix as formerly applied to cachedepends.awk (Thanks to whoever did that, my solution is a case of simple copy/paste).
Created attachment 51956 [details, diff] Patch using a temporary file
Added to CVS, thanks. Should be in next unstable baselayout.
After emerging additional services I got tons of weired "RC_DEPEND_TREE" messages. I gave Stefan's patch http://bugs.gentoo.org/attachment.cgi?id=51956&action=view a try and now the issue is gone. :) Thanx!
I too can corroborate, removing my network card from modules.autoload and compiling it into the kernel fixed it, but I'd sure like to be able to have my network card be a module... Anyone have any ideas for a more permanent fix? (I tried the patch and could not get the patch to apply sucessfully) Tommy
should be released
