1. Updated to new baselayout 1.11.12-r4 2. did etc-update and update all files in /etc/init.d 3. restarted system * Starting eth0 "/etc/init.d/net.eth0: line 114: find: command not found" also multiple warnings "DO NOT USE EXIT IN INIT.D SCRIPTS" in /var/log/ everything/current BUT: manually running net.eth0 works fine Reproducible: Always Steps to Reproduce: see "Details" Actual Results: No network connection anymore because eth0 is down Expected Results: Start eth0 Portage 2.0.51.19 (default-linux/x86/2005.0, gcc-3.3.5-20050130, glibc-2.3.4. 20041102-r1, 2.6.11-gentoo-r7 i686) ================================================================= System uname: 2.6.11-gentoo-r7 i686 Pentium III (Katmai) Gentoo Base System version 1.6.12 Python: dev-lang/python-2.3.5 [2.3.5 (#1, May 3 2005, 23:20:25)] dev-lang/python: 2.3.5 sys-apps/sandbox: [Not Present] sys-devel/autoconf: 2.59-r6, 2.13 sys-devel/automake: 1.7.9-r1, 1.8.5-r3, 1.5, 1.4_p6, 1.6.3, 1.9.5 sys-devel/binutils: 2.15.92.0.2-r10 sys-devel/libtool: 1.5.16 virtual/os-headers: 2.6.8.1-r2 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CFLAGS="-O3 -march=pentium3 -pipe" CHOST="i686-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.4/env /usr/kde/3.4/ share/config /usr/kde/3.4/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="-O3 -march=pentium3 -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs autoconfig ccache distlocks sandbox sfperms strict" GENTOO_MIRRORS="http://gentoo.inode.at" LANG="en_US" LINGUAS="de en en_GB pl" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" SYNC="rsync://localhost/gentoo-portage" USE="x86 kdeenablefinal pam readline samba usepackagedmakefiles linguas_de linguas_en linguas_en_GB linguas_pl userland_GNU kernel_linux elibc_glibc" Unset: ASFLAGS, CBUILD, CTARGET, LC_ALL, LDFLAGS, PORTDIR_OVERLAY
These are the repeating lines in /var/log/everything/current [rc-scripts] DO NOT USE EXIT IN INIT.D SCRIPTS [rc-scripts] This IS a bug, please fix your broken init.d
The failing line in the rc script net.eth0 (actually symlinked to net.lo) is: MODULES=( $( find ${MODULES_DIR}/ -maxdepth 1 -type f ! -name ".*" | sort ) )
sounds like you have net in your boot runlevel ... can you post the output of: rc-update -s | grep -v '| *$'
Here it is: bootmisc | boot checkfs | boot checkroot | boot clock | boot consolefont | boot domainname | default firewall | default hdparm | default hostname | boot keymaps | boot lisa | default local | default nonetwork localmount | boot metalog | default modules | boot net.eth0 | default net.lo | boot netmount | default rmnologin | boot samba | default serial | boot urandom | boot vmware-tools | default xinetd | default
The installation is from November last year and has continously been updated since. I didn't move any services from the boot to the default runlevel (or vice versa).
do you have /usr on a sep partition from / ? does /usr/bin/find exist ?
I don't have any partitions - and find is in /usr/bin. The net.eth0 script works fine when run manually after booting as root. I guess /usr/bin is not in the "rc script path".
I skipped through the net.lo script and the only two other external programs I found were modprobe and sort. These were called by the full path (/sbin/modprobe and /bin/sort). So maybe it's just as trivial as replacing "find" with "/usr/bin/find" in the script.
This should already be handled by functions.sh, btw... if [ -z "${EBUILD}" ] then # Setup a basic $PATH. Just add system default to existing. # This should solve both /sbin and /usr/sbin not present when # doing 'su -c foo', or for something like: PATH= rcscript start PATH="/bin:/sbin:/usr/bin:/usr/sbin:/usr/local/sbin:${PATH}" ... Seems like "unset EBUILD" would be a good idea in runscript.sh, but that's probably not the problem here.
I have been very afraid to etc-update net.eth0 and net.lo with the sys-apps/baselayout-1.11.12-r4 update. The update pretty much wipes out net.eth0 and replaces it with this header: + # $Header: /home/cvsroot/gentoo-src/rc-scripts/init.d/net.lo,v 1.10 2004/04/21 17:09:18 vapier Exp $ and then deals with nothing but net.lo as far as I can tell. Doing a glance at both seems to lead me to think that the files got misnamed and that the net.eth0 in sys-apps/baselayout-1.11.12-r4 should actually be for net.lo and vice-versa.
baselayout should only have one net script, net.lo all other interfaces need to be symlinks to that /etc/init.d/net.lo /etc/init.d/net.eth0 -> net.lo
Okay, here is how I solved it myself: 1. I moved the net.eth0 service from the default to the boot runlevel. Can someone verify that this is the recommended runlevel for net.eth0? 2. I changed RC_PARALLEL_STARTUP from "yes" to "no" in /etc/conf.d/rc. Note that this configuration worked in the previous baselayout.
Correction: 'I changed RC_PARALLEL_STARTUP from "yes" to "no"' - this is independent of the error. Now I get an error saying "net.eth0: cannot start until the runlevel boot has completed" but after the boot runlevel is executed net.eth0 start without error
the only net script that is supposed to be in the boot runlevel is net.lo if you try to put any other net script into the boot runlevel, the script will detect that and issue an error (as you can see)
Okay, so the right place for the net.eth0 symlink would be the default runlevel. Anybody with an idea about why net.eth0 doesn't work in the default runlevel, but in the boot runlevel and when run manually?!
Is the error still "net.eth0: cannot start until the runlevel boot has completed"? If so, what is the output of rc-update -s | grep net.eth0 Can you also verify the status of net.eth0 immediately after booting? /etc/init.d/net.eth0 status
>Is the error still "net.eth0: cannot start until the runlevel boot has completed"? Yes, sure. But this is "normal" because net.eth0 is now in the boot runlevel. The error is no problem because actually net.eth0 is started at the end of the boot runlevel. > If so, what is the output of rc-update -s | grep net.eth0 > Can you also verify the status of net.eth0 immediately after booting? /etc/init.d/net.eth0 status I will do that when I have again access to the virtual machine (on Wednesday). But please: the bug/problem is *NOT* the error message of net.eth0 in the boot level (because it /does/ get started). The bug/problem is that net.eth0 doesn't start at all in the default runlevel.
Please test baselayout-1.12.0_pre1-r1 - it should fix your issue Re-open if not.
Problem solved. Thanks to all who worked on it. So what has actually been the cause of the problem? There's nothing about it in the Changelog...
You'll have to look waaaaay down as it was fixed in the dev branch months ago Azarah fixed it on May 10th
