| View previous topic :: View next topic |
| Author |
Message |
borh
Tux's lil' helper
Joined: 24 Nov 2002
Posts: 89
Location: Ljubljana
|
 Posted: Tue Feb 12, 2008 4:26 pm Post subject: |
 |
|
Just reporting that totally cleaning the /etc/conf.d and init.d directories and remerging everything like micr0c0sm said solved all problems and eveything seems to work now, even with parallel service loading.
Thanks for the great work, my system is booting up faster.  |
|
| Back to top |
|
 |
UberLord
Retired Dev
Joined: 18 Sep 2003
Posts: 6835
Location: Blighty
|
| Posted: Tue Feb 12, 2008 7:58 pm Post subject: |
|
|
| unK wrote: | | I have / on reiser4 and when I set baselayout to run in parallel mode, it hangs at "Remounting root filesystem read-write...". In serial mode remounting works. |
Possibly reiser4fsprogs doesn't understand the -p option to fsck.
| Quote: | | Btw, after each reemerge openrc adds netmount to default runlevel even if it wasn't there before recompilation. |
The git ebuild will do that - released versions won't
_________________
Use dhcpcd for all your automated network configuration needs
Use dhcpcd-ui (GTK+/Qt) as your System Tray Network tool |
|
| Back to top |
|
|
UberLord
Retired Dev
Joined: 18 Sep 2003
Posts: 6835
Location: Blighty
|
| Posted: Tue Feb 12, 2008 7:59 pm Post subject: |
|
|
| borh wrote: | | Just reporting that totally cleaning the /etc/conf.d and init.d directories and remerging everything like micr0c0sm said solved all problems and eveything seems to work now, even with parallel service loading. |
I hope you backed up everything so we can try and find the config setting that is causing this mess!
If you have backups of /etc/conf.d/ and /etc/init.d, please email them to me
| Quote: | | Thanks for the great work, my system is booting up faster. |
Np
_________________
Use dhcpcd for all your automated network configuration needs
Use dhcpcd-ui (GTK+/Qt) as your System Tray Network tool |
|
| Back to top |
|
|
tranquilcool
Veteran
Joined: 25 Mar 2005
Posts: 1179
|
| Posted: Wed Feb 13, 2008 10:39 am Post subject: |
|
|
i just want to ask if any has got baselayout-2 to start spamd.
only baselayout-1 starts the service for me.
_________________
this is a strange strange world. |
|
| Back to top |
|
|
mv
Watchman
Joined: 20 Apr 2005
Posts: 6747
|
| Posted: Wed Feb 13, 2008 2:04 pm Post subject: |
|
|
| tranquilcool wrote: | | i just want to ask if any has got baselayout-2 to start spamd. |
One obvious bug in the init-file is | Code: | | kill -HUP $(< ${PIDFILE}) |
which should actually be something like | Code: | | kill -HUP $(cat -- ${PIDFILE}) |
to be POSIX compliant (maybe you should report this bug). However, this concerns only the "reload" option and if your /bin/sh is not bash. The rest of the script looks sane (but I have never tried). Perhaps it helps to use -L in SPAMD_OPTS? |
|
| Back to top |
|
|
tranquilcool
Veteran
Joined: 25 Mar 2005
Posts: 1179
|
| Posted: Wed Feb 13, 2008 2:39 pm Post subject: |
|
|
| mv wrote: | | tranquilcool wrote: | | i just want to ask if any has got baselayout-2 to start spamd. |
One obvious bug in the init-file is | Code: | | kill -HUP $(< ${PIDFILE}) |
which should actually be something like | Code: | | kill -HUP $(cat -- ${PIDFILE}) |
to be POSIX compliant (maybe you should report this bug). However, this concerns only the "reload" option and if your /bin/sh is not bash. The rest of the script looks sane (but I have never tried). Perhaps it helps to use -L in SPAMD_OPTS? |
tried to modify the init file and reloads works.
the -L option doesn't fix the start script.
very funny that, though it doesn't start, i have the spamd.pid
in /var/run/spamd/
_________________
this is a strange strange world. |
|
| Back to top |
|
|
pholthau
Guru
Joined: 27 Nov 2005
Posts: 361
Location: Bielefeld, Germany
|
| Posted: Wed Feb 13, 2008 5:07 pm Post subject: |
|
|
Hey!
@Mods: Feel free to move this post, I don't know if it really belongs here (not trying to hijack thread)
Today, I decided to give openrc a try.
But I am having some troubles with it.
1. If I enable parallel startup, the system won't boot. Instead it crashes with a backtrace after calculating dependencies.
2. I was very pleased to see "rc_depend_strict". With the help of this I thought it might be possible to let net.tap0 depend on
openvpn, no matter to which VPN it is connected (I created two symlinks to /etc/init.d/openvpn called openvpn.home, openvpn.remote).
So I added
rc_net_tap0_need="openvpn"
to /etc/rc.conf
Sadly, it does not work:
| Code: |
/etc/init.d/net.tap0 start
* Starting openvpn ...
grep: /etc/openvpn/openvpn.conf: No such file or directory
grep: /etc/openvpn/openvpn.conf: No such file or directory
grep: /etc/openvpn/openvpn.conf: No such file or directory
grep: /etc/openvpn/openvpn.conf: No such file or directory
Options error: In [CMD-LINE]:1: Error opening configuration file: /etc/openvpn/openvpn.conf
Use --help for more information.
* start-stop-daemon: failed to start `/usr/sbin/openvpn'
* Check your logs to see why startup failed [!!]
* ERROR: openvpn failed to start
* ERROR: cannot start net.tap0 as openvpn would not start
|
3. IRQ Balance fails to start
| Code: |
/etc/init.d/irqbalance start
* Starting irqbalance ...
* start-stop-daemon: did not create a valid pid in `/var/run/irqbalance.pid' [!!]
* ERROR: irqbalance failed to start
|
Maybe you can tell me about these... Maybe I can provide you with more information...
_________________
The message is Feierei, alder! The message is Gude Laune, alder! [Sven] |
|
| Back to top |
|
|
FireBurn
Apprentice
Joined: 19 Sep 2004
Posts: 170
Location: Edinburgh, UK
|
| Posted: Wed Feb 13, 2008 7:21 pm Post subject: |
|
|
Grr
OK cleared out my /etc/init.d and my /etc/conf.d
Installed openrc
No error... Hurrah!!
Hoever X doesn't start (yes I've reinstalled xdm and rc-update add xdm default) I can't login as either root or user and when I reboot using Ctrl Alt Del it stops all the services but doesn't restart the computer!!!
How irritating! |
|
| Back to top |
|
|
micr0c0sm
Tux's lil' helper
Joined: 29 Oct 2005
Posts: 148
Location: New York
|
| Posted: Wed Feb 13, 2008 7:48 pm Post subject: |
|
|
| Yeah shutdown -h now or halt or ctrl-alt-del all don't actually shut down the machine, just stop the runlevels... |
|
| Back to top |
|
|
borh
Tux's lil' helper
Joined: 24 Nov 2002
Posts: 89
Location: Ljubljana
|
| Posted: Wed Feb 13, 2008 8:29 pm Post subject: |
|
|
| FireBurn wrote: | Grr
OK cleared out my /etc/init.d and my /etc/conf.d
Installed openrc
No error... Hurrah!!
Hoever X doesn't start (yes I've reinstalled xdm and rc-update add xdm default) I can't login as either root or user and when I reboot using Ctrl Alt Del it stops all the services but doesn't restart the computer!!!
How irritating! |
Make sure you remerge any packages that install into those directories. Could it be that some other app, apart from xdm, installed something X-related?
An
| Code: | equery belongs /etc/conf.d/*
equery belongs /etc/init.d/*
|
should help you there, if you made a backup, that is. I don't use xdm so I can't help you directly, but maybe remergeing those other apps would solve your problems (admittedly, a long shot).
With the above mentioned method, I found out that 34 (!) packages install into /etc/init.d on my system. |
|
| Back to top |
|
|
FireBurn
Apprentice
Joined: 19 Sep 2004
Posts: 170
Location: Edinburgh, UK
|
| Posted: Wed Feb 13, 2008 10:22 pm Post subject: |
|
|
That's all well and good BUT what the hell is stopping me from logging on!!!
It's a complete pain at the moment having to boot with options splash=off init=/bin/bash
then mount -o remount,rw /
set PATH="/bin:/usr/bin/:/sbin:/usr/sbin"
change stuff or reemerge stuff
reboot
to realise that sshd and xdm still aren't being started (samba is)
despite xdm and sshd being placed in boot or default (requiring several restarts) |
|
| Back to top |
|
|
mdeininger
Veteran
Joined: 15 Jun 2005
Posts: 1740
Location: Emerald Isles, observing Dublin's docklands
|
| Posted: Thu Feb 14, 2008 10:31 am Post subject: |
|
|
assuming the files are actually in /etc/init.d/, maybe they 'need' something that isn't around anymore...
_________________
"Confident, lazy, cocky, dead." -- Felix Jongleur, Otherland
( Twitter | Blog | GitHub ) |
|
| Back to top |
|
|
UberLord
Retired Dev
Joined: 18 Sep 2003
Posts: 6835
Location: Blighty
|
| Posted: Wed Feb 20, 2008 12:51 pm Post subject: |
|
|
| pholthau wrote: | | 1. If I enable parallel startup, the system won't boot. Instead it crashes with a backtrace after calculating dependencies. |
I think I've fixed this now - valgrind was helpful in exposing an off-by-1 error when parallel was enabled.
| Quote: |
2. I was very pleased to see "rc_depend_strict". With the help of this I thought it might be possible to let net.tap0 depend on
openvpn, no matter to which VPN it is connected (I created two symlinks to /etc/init.d/openvpn called openvpn.home, openvpn.remote).
So I added
rc_net_tap0_need="openvpn"
to /etc/rc.conf
Sadly, it does not work:
|
That won't work
Also, I'm not sure what you're trying to achieve as openvpn will create a tap0 anyway ......
| Quote: |
3. IRQ Balance fails to start
| Code: |
/etc/init.d/irqbalance start
* Starting irqbalance ...
* start-stop-daemon: did not create a valid pid in `/var/run/irqbalance.pid' [!!]
* ERROR: irqbalance failed to start
|
Maybe you can tell me about these... Maybe I can provide you with more information... |
I'll try and look into this, but as I don't have an smp machine it's doubtful I could test any potnetial fixes.
You could try removing the --pidfile directive in the start() part of the init script.
_________________
Use dhcpcd for all your automated network configuration needs
Use dhcpcd-ui (GTK+/Qt) as your System Tray Network tool |
|
| Back to top |
|
|
UberLord
Retired Dev
Joined: 18 Sep 2003
Posts: 6835
Location: Blighty
|
|
| Back to top |
|
|
Insanity5902
Veteran
Joined: 23 Jan 2004
Posts: 1228
Location: Fort Worth, Texas
|
| Posted: Thu Feb 21, 2008 1:18 am Post subject: |
|
|
Well, one thing lead to another, and I ended up purchasing a new and bigger HDD (woohoo for perpinduclar recording)
Anyways, just emerge the openrc-9999 and all seems to work now, I've got parrell startup enabled also.
_________________
Join the adopt an unanswered post initiative today |
|
| Back to top |
|
|
ryker
Guru
Joined: 28 May 2003
Posts: 412
Location: Portage, IN
|
| Posted: Thu Feb 21, 2008 1:19 am Post subject: |
|
|
| pholthau wrote: | Hey!
3. IRQ Balance fails to start
| Code: |
/etc/init.d/irqbalance start
* Starting irqbalance ...
* start-stop-daemon: did not create a valid pid in `/var/run/irqbalance.pid' [!!]
* ERROR: irqbalance failed to start
|
|
Do you have a multiple socket system? According to this thread, they have disabled startup on unicore/dualcore systems.
_________________
Athlon 64 3200+, 80G WD sata hd + 200G IDE, 1G Geil DDR400, MSI K8T Neo
IntelCore2Duo 2.0Ghz MSI laptop,100G SATA hd, 2G RAM |
|
| Back to top |
|
|
pholthau
Guru
Joined: 27 Nov 2005
Posts: 361
Location: Bielefeld, Germany
|
| Posted: Thu Feb 28, 2008 5:33 pm Post subject: |
|
|
| UberLord wrote: |
I think I've fixed this now - valgrind was helpful in exposing an off-by-1 error when parallel was enabled.
|
Okay, I will try it again.
EDIT: Works now, thanks!
| UberLord wrote: |
| Quote: |
2. I was very pleased to see "rc_depend_strict". With the help of this I thought it might be possible to let net.tap0 depend on
openvpn, no matter to which VPN it is connected (I created two symlinks to /etc/init.d/openvpn called openvpn.home, openvpn.remote).
So I added
rc_net_tap0_need="openvpn"
to /etc/rc.conf
Sadly, it does not work:
|
That won't work
Also, I'm not sure what you're trying to achieve as openvpn will create a tap0 anyway ......
|
I was trying to let nfsmount depend on net.tap0. net.tap0 should trigger any openvpn to start up.
| UberLord wrote: |
I'll try and look into this, but as I don't have an smp machine it's doubtful I could test any potnetial fixes.
You could try removing the --pidfile directive in the start() part of the init script. |
I will try it out tonight.
EDIT: I commented the --pidfile-line. The init script starts but the interrupts are not disributed between the two cores.
| ryker wrote: |
Do you have a multiple socket system? According to this thread, they have disabled startup on unicore/dualcore systems. |
No, I don't. It's a Core2 Duo. I emerged irqbalance because of this thread: https://forums.gentoo.org/viewtopic-t-594074-highlight-irqbalance.html
And thanks for all your answers!
_________________
The message is Feierei, alder! The message is Gude Laune, alder! [Sven] |
|
| Back to top |
|
|
UberLord
Retired Dev
Joined: 18 Sep 2003
Posts: 6835
Location: Blighty
|
| Posted: Thu Feb 28, 2008 9:34 pm Post subject: |
|
|
| pholthau wrote: |
2. I was very pleased to see "rc_depend_strict". With the help of this I thought it might be possible to let net.tap0 depend on
openvpn, no matter to which VPN it is connected (I created two symlinks to /etc/init.d/openvpn called openvpn.home, openvpn.remote).
So I added
rc_net_tap0_need="openvpn"
to /etc/rc.conf
Sadly, it does not work:
|
That won't work
Also, I'm not sure what you're trying to achieve as openvpn will create a tap0 anyway ......
| Quote: |
I was trying to let nfsmount depend on net.tap0. net.tap0 should trigger any openvpn to start up. |
So the correct solution is to remove all net.* as net dependencies.
This is done by only having net.tap0 in the default runlevel and allowing your other interfaces to be coldplugged.
Or add rc_need="net.tap0" to /etc/conf.d/nfsmount
| Quote: | | EDIT: I commented the --pidfile-line. The init script starts but the interrupts are not disributed between the two cores. |
They're not - it's for multiple CPU's, not multiple cores.
_________________
Use dhcpcd for all your automated network configuration needs
Use dhcpcd-ui (GTK+/Qt) as your System Tray Network tool |
|
| Back to top |
|
|
kernelOfTruth
Watchman
Joined: 20 Dec 2005
Posts: 6111
Location: Vienna, Austria; Germany; hello world :)
|
|
| Back to top |
|
|
pholthau
Guru
Joined: 27 Nov 2005
Posts: 361
Location: Bielefeld, Germany
|
| Posted: Thu Feb 28, 2008 10:24 pm Post subject: |
|
|
| UberLord wrote: |
So the correct solution is to remove all net.* as net dependencies.
This is done by only having net.tap0 in the default runlevel and allowing your other interfaces to be coldplugged.
Or add rc_need="net.tap0" to /etc/conf.d/nfsmount
|
I don't understand exactly what you are trying to tell me.
Maybe I was kind of unclear about what I am trying to set up:
nfsmount needs tap0 because it tries to access mounts that are available via this interface.
tap0 needs openvpn because openvpn creates that device. If openvpn is not started, tap0 refuses to start, too.
the openvpn connection can be established in two different scenarios: at home (openvpn.home) or remote (openvpn.remote)
openvpn needs another net (eth0 or wlan0) to be up and running.
So only having net.tap0 in the default runlevel would not work because I neither have eth0/wlan0 nor an openvpn connection. I don't know if it is possible to have the nfs shares mounted automatically. At the moment, I am doing it like this:
1. wlan0 is started automatically (runlevel default or udev-triggered)
2. starting openvpn.home or remote (dependent on where I am)
3. starting net.tap0
4. starting nfsmount
There are 2 drawbacks of this approach:
1. I have to open a root shell every time I boot and type in these commands.
2. If I forget to stop nfsmount before shutting down, the laptop cannot shut down cleanly because it is unable to unmount the shares (openvpn is stopped before nfsmount or net.tap0)
_________________
The message is Feierei, alder! The message is Gude Laune, alder! [Sven] |
|
| Back to top |
|
|
n3Cre0
Tux's lil' helper
Joined: 16 Jan 2007
Posts: 129
|
| Posted: Mon Mar 24, 2008 11:45 am Post subject: |
|
|
Sorry if I'm hijacking this thread but it also falls under 'baselayout issues'
Yesterday I updated and I was "forced" (because the update process wouldn't continue otherwise) to pull openrc-9999 and baselayout-2. I updated, updated all my config files (there were ALOT of them) and rebooted... Then the trouble began.
Between each startup service it displayed errors about splash.so (which I got rid of later by uninstalling splashutils). This splash wasn't that bad, worse was the fact that it never mounted my / rw, it was ro, causing virtually every service to fail.
| Sample of the splash errors wrote: | /sbin/runscript: relocation error: /lib/rc/plugins/splash.so
symbol rc_strlist_free
version RC_1.0 not defined in librc.so.1 with link time reference |
When I logged in (console obviously - X refused to start - /tmp isn't even writable), I managed to get rw support back by using
| Code: | mount / -o remount,rw
mount -t proc proc /proc
mount -t sysfs sys /sys
mount /dev/sda3 /home
mount /dev/sda7 /boot/
mount /dev/sda6 /media/backupos/
env-update
/etc/init.d/gpm restart
echo \'source .bashrc\' now |
Right now I uninstalled splashutils causing the nasty bootup warnings/errors to disappear, but it still boots / ro...
I just completely updated my system (world) again, updated every config file, reinstalled openrc, baselayout and stuff (like udev) but it still mount ro. Can someone help me out here? 
_________________
OS: Linux 2.6.26 i686 Gentoo v2.0.0 Sound: HDA Intel
CPU: Intel(R) Pentium(R) M processor 1.73GHz (GenuineIntel)
Disk: 95.23GB Video: ATI Radeon Mobility X700 XL with Catalyst 8.08 RAM: 1011MB
Last edited by n3Cre0 on Mon Mar 24, 2008 1:32 pm; edited 1 time in total |
|
| Back to top |
|
|
asturm
Developer
Joined: 05 Apr 2007
Posts: 8936
|
| Posted: Mon Mar 24, 2008 12:14 pm Post subject: |
|
|
| Which udev version have you installed? If > udev-115 (I think they made the switch back then, using udev-118 myself), have you added a udev tmpfs entry to your /etc/fstab? |
|
| Back to top |
|
|
n3Cre0
Tux's lil' helper
Joined: 16 Jan 2007
Posts: 129
|
| Posted: Mon Mar 24, 2008 1:22 pm Post subject: |
|
|
| genstorm wrote: | | Which udev version have you installed? If > udev-115 (I think they made the switch back then, using udev-118 myself), have you added a udev tmpfs entry to your /etc/fstab? |
| Quote: | [I] sys-fs/udev Available versions: 104-r12 104-r13 (~)111 (~)111-r1 (~)111-r3 (~)112 (~)112-r1 (~)113 (~)113-r1 (~)113-r2 114 (~)114-r1 (~)114-r2 (~)115 115-r1 (~)115-r5 (~)115-r6 (~)116 (~)116-r1 (~)117 (~)118 (~)118-r1 (~)118-r2 (~)118-r3 (~)119 {selinux}
Installed versions: 119(01:15:37 03/24/08)(-selinux)
Homepage: http://www.kernel.org/pub/linux/utils/kernel/hotplug/udev.html
Description: Linux dynamic and persistent device naming support (aka userspace devfs) |
I guess not:
| cat /etc/fstab wrote: | # /etc/fstab: static file system information.
#
# noatime turns off atimes for increased performance (atimes normally aren't
# needed; notail increases performance of ReiserFS (at the expense of storage
# efficiency). It's safe to drop the noatime options if you want and to
# switch between notail / tail freely.
#
# The root filesystem should have a pass number of either 0 or 1.
# All other filesystems should have a pass number of 0 or greater than 1.
#
# See the manpage fstab(5) for more information.
#
# <fs> <mountpoint> <type> <opts> <dump/pass>
# NOTE: If your BOOT partition is ReiserFS, add the notail option to opts.
#/dev/mapper/crypt-swap none swap sw 0 0
/dev/sda2 / ext3 noatime 0 1
/dev/sda7 /boot ext2 defaults 1 2
/dev/sda3 /home ext3 defaults 1 2
/dev/mapper/crypt-tmp /tmp ext2 noatime 0 1
/dev/sda1 /media/windows ntfs-3g auto,users,uid=1000,gid=100 0 0
/dev/sda5 /media/documents vfat auto,users,exec,uid=1000,gid=100 0 2
/dev/sda6 /media/backtrack ext3 noatime 0 2
/dev/cdrom /media/dvdrw iso9660 noauto,ro,users 0 0
# ** INFO **
/dev/sda9 none swap sw 0 0
#/dev/sda8 /tmp ext2 noatime 0 1
# NOTE: The next line is critical for boot!
proc /proc proc defaults 0 0
# glibc 2.2 and above expects tmpfs to be mounted at /dev/shm for
# POSIX shared memory (shm_open, shm_unlink).
# (tmpfs is a dynamically expandable/shrinkable ramdisk, and will
# use almost no memory if not populated with files)
shm /dev/shm tmpfs nodev,nosuid,noexec 0 0
|
What should I add?
--
I did some more booting and this is what I noticed:
It start erroring at
| Quote: | | * Invalidating stale software suspend images |
with
| Quote: | ERROR: cannot start localmount as fsck would not start
ERROR: cannot start bootmisc as fsck would not start
ERROR: cannot start alsasound as fsck would not start
ERROR: cannot start keymaps as fsck would not start
ERROR: cannot start net.lo as fsck would not start
ERROR: cannot start udev-postmount as fsck would not start
ERROR: cannot start urandom as fsck would not start
|
and so on..
Then dbus fails to start because / is ro, and then multiple services fail because of dbus isn't running.
Domino effect.
_________________
OS: Linux 2.6.26 i686 Gentoo v2.0.0 Sound: HDA Intel
CPU: Intel(R) Pentium(R) M processor 1.73GHz (GenuineIntel)
Disk: 95.23GB Video: ATI Radeon Mobility X700 XL with Catalyst 8.08 RAM: 1011MB
Last edited by n3Cre0 on Mon Mar 24, 2008 2:52 pm; edited 1 time in total |
|
| Back to top |
|
|
asturm
Developer
Joined: 05 Apr 2007
Posts: 8936
|
| Posted: Mon Mar 24, 2008 1:46 pm Post subject: |
|
|
| n3Cre0 wrote: | | What should I add? |
You need to put the following line into fstab:
| Code: | | udev /dev tmpfs size=10M,nr_inodes=4k 0 0 |
|
|
| Back to top |
|
|
tranquilcool
Veteran
Joined: 25 Mar 2005
Posts: 1179
|
| Posted: Mon Mar 24, 2008 2:07 pm Post subject: |
|
|
anybody got spamd to start?
_________________
this is a strange strange world. |
|
| Back to top |
|
|
|
Unsupported Software
