openrc support thread

Message

ceric35 · Post by **ceric35** » Sat Jan 17, 2009 10:03 am

I forgot to add udev in sysinit runlevel

Phlogiston · Post by **Phlogiston** » Sun Jan 18, 2009 11:59 am

Phlogiston wrote:I can't properly shutdown when using parallel boot. Looks like an endless loop: a lot is waiting for mtab, while mtab waits for root without a change...

Any idea?

I read that this might be a kernel issue. Anyone found out what exactly was the problem? I'm using zen-sources and still can't shutdown properly.

Smart1 · Post by **Smart1** » Mon Jan 19, 2009 1:54 pm

I have probs with OpenRC

Have all Programms the newest

uname -r
2.6.28-0

cat /proc/cpuinfo | grep 'model name' | sed -e 's/model name.*: //' | uniq
AMD Athlon(tm) 64 X2 Dual Core Processor 5000+

OpenRC 0.4.2

rc-status
Runlevel: default
net.eth0 [ started ]
privoxy [ started ]
syslog-ng [ started ]
vsftpd [ started ]
xfs [ started ]
lm_sensors [ started ]
cpudyn [ crashed ]
gpm [ started ]
smartd [ started ]
ddclient [ crashed ]
dbus [ started ]
cupsd [ started ]
sshd [ started ]
denyhosts [ crashed ]
esound [ started ]
hald [ started ]
noip [ crashed ]
tor [ started ]
vixie-cron [ started ]
local [ started ]
Runlevel: hotplugged
udev-postmount [ started ]
Runlevel: needed
udev-dev-tarball [ started ]
Runlevel: manual

rc-status boot
Runlevel: boot
hwclock [ started ]
modules [ started ]
fsck [ started ]
root [ started ]
termencoding [ started ]
mtab [ started ]
localmount [ started ]
hostname [ started ]
sysctl [ started ]
bootmisc [ started ]
alsasound [ started ]
net.lo [ started ]
urandom [ started ]
consolefont [ started ]
udev-postmount [ started ]
procfs [ started ]
keymaps [ started ]
swap

cat /proc/cmdline
root=/dev/sda1

rc
* Caching service dependencies... [ ok ]
* Starting udevd daemon...
* start-stop-daemon: /sbin/udevd is already running [ !! ]
* ERROR: udevd failed to start

net.eth0 works great

/etc/conf.d/net
config_eth0="192.168.1.33 netmask 255.255.255.0 broadcast 192.168.1.255"
routes_eth0="default via 192.168.1.1"

#/etc/init.d/cpudyn status
* status: crashed
#/etc/init.d/cpudyn stop
* Stopping cpudynd... [ ok ]
#/etc/init.d/cpudyn start
* Starting cpudynd... [ ok ]
# /etc/init.d/cpudyn status
* status: crashed

# /etc/init.d/ddclient status
* status: crashed
# /etc/init.d/ddclient stop
* Stopping ddclient...
* start-stop-daemon: no matching processes found [ ok ]
# /etc/init.d/ddclient start
* Starting ddclient... [ ok ]
# /etc/init.d/ddclient status
* status: crashed

# /etc/init.d/denyhosts status
* status: crashed
# /etc/init.d/denyhosts stop
* Stopping DenyHosts daemon... [ ok ]
# /etc/init.d/denyhosts start
* Starting DenyHosts daemon... [ ok ]
# /etc/init.d/denyhosts status
* status: crashed

# /etc/init.d/noip status
* status: crashed
# /etc/init.d/noip stop
* Stopping noip...
* Error stopping the noip daemon, error 1 [ !! ]
* Setting noip addresses to 0.0.0.0... [ ok ]
* ERROR: noip failed to stop
# /etc/init.d/noip start
* WARNING: noip has already been started
# /etc/init.d/noip status
* status: crashed

What is this?

avx · Post by **avx** » Thu Jan 22, 2009 4:22 pm

Little question/wish regarding the handling of dm-crypt/luks with openrc.

I've got the following partitions, which all are encrypted with different keys: /, /usr, /var, /tmp, /opt, /home - so, a total of 5 (/ is handled by the init-script of course).

For convenience, all keyfiles are on the same medium. I don't know exactly, how openrc handles it, but counting the time, it seems that for every partition, the medium gets mounted, the partition decrypted/mapped, medium gets unmounted. Every mount/umount takes around 1s, so that makes a total of around 10s just for mounting. Shouldn't the dmcrypt-file get parsed and checked, if all the keys are in the same location and then the medium gets only mounted/umounted once? Doesn't have to be default, but at least an option would be nice.

On the other hand, even when running in parallel-mode, I've got the feeling, that the cryptsetup-stuff is handled one after the other. That may be safer, but takes a long time and f.e. isn't needed in my case, so would be another option for speed-up.

Am I totally wrong or am I guessing correct and the statements above do make sense?

earthhandbag · Post by **earthhandbag** » Thu Jan 22, 2009 10:12 pm

I am completly underwhelmed.

No, I am pissed.

fsck crashs out on boot - and then nothing goes, because it can not check some fs. WTF? ok, /fastboot works around it, but it is completly and absolutly fucked up.

what is wrong with this fstab - except that the old baselout worked flawlessly with it?
# /etc/fstab: static file system information.
# $Header: /var/cvsroot/gentoo-src/rc-scripts/etc/fstab,v 1.18.4.1 2005/01/31 23:05:14 vapier Exp $
#
# 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.
#
# 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/sda1 /boot ext2 noauto,noatime 1 2
/dev/sda3 / reiserfs data=journal,noatime 0 1
/dev/sda2 none swap sw 0 0

#/dev/disk/by-uuid/7eda1c5e-c53d-4620-9a03-75c0a6e70774 /home reiser4 noatime,tmgr.atom_max_size=32768 0 2
/dev/disk/by-uuid/396a1f32-5f1b-42db-84f0-5ff54ca8ab6d /home reiser4 noatime,tmgr.atom_max_size=32768 0 2

/dev/sda6 /var reiser4 noatime,tmgr.atom_max_size=32768 0 2

/dev/disk/by-uuid/c1a84bd9-91f1-420e-9880-d4290599a399 /mnt/filme reiserfs data=journal,noatime,noauto,rw,users 0 0

/dev/disk/by-uuid/2a2f500a-125c-4205-b510-bd48e7b10c53 /opt reiser4 noatime,tmgr.atom_max_size=32768 0 0

# NOTE: The next line is critical for boot!

proc /proc proc nosuid,noexec 0 0

#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

#/tmp/swapfile none swap sw 0 0

/dev/cdrw1 /mnt/cdrom udf,iso9660 noauto,noatime,rw,users 0 0

tmpfs /var/tmp/portage tmpfs rw,size=2G 0 0
tmpfs /tmp tmpfs rw,size=512M,noexec 0 0

Second thing that REALLY pissed me of was the removal of the net stuff - ok. I was warned, but still.

Boot is not faster than before - no it is slower, same like reboot/shutdown.

And where can I tell the new baselayout/openrc stuff to NOT load modules except the ones explecitly told so?

oh and this:
fancontrol [ crashed ]

Love it! Not!

Phlogiston · Post by **Phlogiston** » Fri Jan 23, 2009 1:59 pm

remove reiser4

kernelOfTruth · Post by **kernelOfTruth** » Fri Jan 23, 2009 2:20 pm

Phlogiston wrote:remove reiser4

it works perfectly with 0.4.2 and reiser4

I don't know why you are all bashing on it

here my working "config" in the run levels:

rc-update show boot
bootmisc | boot
consolefont | boot
device-mapper | boot
dmcrypt | boot
fsck | boot
hostname | boot
hwclock | boot
keymaps | boot
localmount | boot
modules | boot
mtab | boot
net.lo | boot
procfs | boot
root | boot
swap | boot
sysctl | boot
termencoding | boot
udev-postmount | boot
urandom | boot

rc-update show default
acpid | default
alsasound | default
dbus | default
gpm | default
irqbalance | default
lm_sensors | default
local | default
net.eth0 | default
netmount | default
ntp-client | default
ntpd | default
pdnsd | default
postfix | default
syslog-ng | default

rc-update show sysinit
devfs | sysinit
dmesg | sysinit
udev | sysinit

wazze · Post by **wazze** » Fri Jan 23, 2009 3:23 pm

earthhandbag wrote: fsck crashs out on boot - and then nothing goes, because it can not check some fs. WTF? ok, /fastboot works around it, but it is completly and absolutly fucked up....

Did you set fsck_args="-a" in /etc/conf.d/fsck?

platojones · Post by **platojones** » Sat Jan 24, 2009 6:27 pm

Sorry if this has been addressed before, but I've searched and I can't find the answer. I've just converted one of my linux root partitions to openrc...went very smooth once I figured out what to unmask to get the right versions of udev, hal and cryptsetup. I have a few minor issues, but they aren't openrc related, but one thing I would like is the logging feature of openrc. The docs say they you don't need showconsole to use it, but when I turn on logging in /etc/conf.d/rc, I don't get any rc.log in /var/log. Anybody know the solution to this? Thanks.

gerard27 · Post by **gerard27** » Sat Jan 24, 2009 6:41 pm

Hi platojones,
The openrc config has been moved to /etc/rc.conf!
If you followed the openrc update instructions you
would have known that.
The /etc/conf.d/rc.conf is deprecated and probably
doesn't work anymore.
Gerard.

platojones · Post by **platojones** » Sat Jan 24, 2009 6:45 pm

Gerard van Vuuren wrote:Hi platojones,
The openrc config has been moved to /etc/rc.conf!
If you followed the openrc update instructions you
would have known that.
The /etc/conf.d/rc.conf is deprecated and probably
doesn't work anymore.
Gerard.

Haha, I did follow it...but I can't read so well...thanks, though, you are right:

/etc/conf.d/rc has been deprecated and any settings you have in there will need to be migrated to the appropriate settings in /etc/rc.conf. Please read through /etc/rc.conf and /etc/conf.d/rc and migrate the settings. Once you are complete, delete /etc/conf.d/rc.

I did just the opposite

Robu · Post by **Robu** » Sat Jan 24, 2009 7:43 pm

Hi, I have a small pbm with my keyboard since I switched to openrc/baselayout-2. I use gnome on a desktop amd64 and I am the only user, I have the login auto; the keyboard and mouse are bluetooth wireless. I followed carefully (at least I tried) the how to upgrade and everything at rebooth went smooth apart from my keyboard do not work at all until I log off (the mouse works!). At the prompt of gdm the keyboard start working and then, when I am again log in, the keyboard works perfectly. During the boot process I can stop/start the scrolling (with CRTL s and crtl q) so it seems the keyboard is working, at least since gdm starts. If I disable the auto login, for example, there is no way to log in because at the gdm prompt the keyboard do not work. No one errors in the X.org.0 log, neither in dmesg. Before the upgrade averything was perfectly. The following are the versions of the pgms installed:

sys-apps/sysvinit-2.86-r12
sys-apps/baselayout-2.0.0
sys-apps/openrc-0.4.2
sys-fs/udev-135-r3
sys-apps/hal-0.5.11-r6

I do not know what check, maybe someone else faced this. Any help very appreciated.
Thanks.

sian · Post by **sian** » Sun Jan 25, 2009 8:23 am

hi
did you update X recently ? new xorg uses hal to configure keyboard and mouse via hal.
you can try this :

Section "ServerLayout"
[...]
Option "AllowEmptyInput" "off"
Option "AllowAddDevices" "off"
EndSection

Thus you tell xorg to manage you keyboard and mouse.

Robu · Post by **Robu** » Sun Jan 25, 2009 5:40 pm

@sian:
No i didn't. I checked in the forum many threads about xorg-server and it's upgrade to version 1.5. My version is still the stable x11-base/xorg-server-1.3.0.0-r6 and what I understand is only version greater than 1.4 use hal to manage keyboards and mouse. In any case I did already added in ServerFlags : Option "AllowEmptyInput" "off" as well as I tried to re-compile xorg and all the modules without success. I will try your options and report the results.
Thanks

Robu · Post by **Robu** » Sun Jan 25, 2009 6:21 pm

@sian:

I will try your options and report the results.

Nothing changed.
In any case no errors are reported in Xorg.0.log
If any idea comes up, please let me know, this pbm is not a big issue but I would prefer to solve it. I would discard the idea to downgrade to stable version.

Thanks.

sian · Post by **sian** » Mon Jan 26, 2009 4:12 pm

@Robu : Well it looked like it was something with hal (detecting your input or not launched)...

Robu · Post by **Robu** » Tue Jan 27, 2009 8:06 am

@sian: Thanks sian for your help. I tried to compare the output of

Code: Select all

 lshal

after the bootstrap and after the logout and login process but unfortunately there are no differences at all. Could you suggest any other check? I have no idea.
Thanks.

VinzC · Post by **VinzC** » Fri Jan 30, 2009 8:09 pm

Yiiikes! 666 replies!

(Throwing salt over my left shoulder.)

EDIT: BTW this one makes it the 667th! We've just got passed a terrible accident, probably...

agent_jdh · Post by **agent_jdh** » Sat Jan 31, 2009 2:49 am

VinzC wrote:Yiiikes! 666 replies!

(Throwing salt over my left shoulder.)

EDIT: BTW this one makes it the 667th! We've just got passed a terrible accident, probably...

RATS. I WANTED THAT ONE

VinzC · Post by **VinzC** » Sat Jan 31, 2009 10:48 am

VinzC wrote:Yiiikes! 666 replies!

(Throwing salt over my left shoulder.)

EDIT: BTW this one makes it the 667th! We've just got passed a terrible accident, probably...

agent_jdh wrote:RATS. I WANTED THAT ONE

Wooohooo! Busted!

You'll get a second chance on the 999th; don't miss it, mate!...

jowr · Post by **jowr** » Sat Jan 31, 2009 11:36 am

Yummy - openrc took booting on my Dell Latitude d430 from about 50 seconds to about 20.

I'm impressed with how easy the upgrade was, especially with the automatic porting of things like my autoloaded modules. I wonder when this will become the default, as it sure is impressive.

makism · Post by **makism** » Sat Jan 31, 2009 6:02 pm

i just upgraded udev/sysvinit/cryptsetup and hal to solve the blockers of openrc-0.4.2.
Now, automounting does not work in gnome. Any thoughts?

(i think it`s faster than 0.2.5? :S)

edit:
my bad, i was right about running revdep-rebuild... something is broken

i`ll edit again later

edit2:
revdep-rebuild fixed hal. but the issue was fixed in this comment:
http://bugs.gentoo.org/show_bug.cgi?id=247025#c1

UberLord · Post by **UberLord** » Thu Feb 05, 2009 11:49 am

ph030 wrote:Little question/wish regarding the handling of dm-crypt/luks with openrc.

OpenRC treats dm-crypt/luks no differently from any other services.

KejPi · Post by **KejPi** » Fri Feb 06, 2009 9:19 pm

Hi, I want to change the order of soundcards. I have added following to /etc/conf.d/modules but is doesn't work. But maybe this is not openrc problem:

Code: Select all

modules_2_6="snd_emu10k1 snd_hda_intel"
module_snd_emu10k1_args_2_6="index=0"  
module_snd_hda_intel_args_2_6="index=1"

The desired behaviour is snd_emu10k1 -> card 0 (default) and snd_hda_intel -> card 1, but the default card is always snd_hda_intel.

What am I doing wrong? Thanks for any help.

HekMek · Post by **HekMek** » Sat Feb 07, 2009 11:46 am

ph030 wrote:Try to remove "-d ${key}" from /lib/rcscripts/addons/dm-crypt-start.sh - that should work. If you've got a combination of key-files and passphrases, try with an if-else-thingy depending on your partition.
If I read the script correct, than it always wants a keyfile and only can distinguish between gpg-protected/not-protected-keyfile.

The way I read the script and the output it produces, it does in fact do the correct system call, however, apparently OpenRC does not allow the script to ask for input, so it just receives EOF and thus opening the devices fails.

Here is what I did to get a little more convenience with DM-Crypt. Encrypted partitions are /home and /tmp, everything but some dirs in /var are on the rootfs.
This method does produce a lot of error messages at boot (could not start X because local did not start...), but did not do appear to do any harm otherwise. Use at your own risk and/or post your suggestions:

Code: Select all

# remove dmcrypt from runlevel(s)
rc-update del dmcrypt default

# Add the respective calls to cryptsetup in /etc/conf.d/local
nano /etc/conf.d/local
-> start section
cryptsetup luksOpen /dev/sda6 tmp -T 3
cryptsetup luksOpen /dev/sda8 home -T 3

-> stop section
cryptsetup luksClose tmp
cryptsetup luksClose home

# You will want cryptsetup to be called before local filesystems are being mounted, so we will also modify /etc/rc.conf
echo "rc_localmount_need="local"" >> /etc/rc.conf

# In my case, this did not stop XDM from being started, so I also added
echo -e "rc_xdm_need="local"\nrc_xdm_after="local"" >> /etc/rc.conf

This approach does create a lot of warnings when I boot (as indicated above), but generally produces the desired result, i.e. I get prompted for the keyphrases on the console when booting and the mapper-drives will be mounted automatically by localmount. If you do not use an US-keyboard, you should be aware that keymaps are not yet loaded when you type your keyphrase - this might be fixeable with some extra editing of rc.conf, but I did not care to do that

.
This is certainly a very blunt hack. Use at your own risk and don't forget to remove all the changes once OpenRCs dmcrypt allows keyphrases, again.

openrc support thread

Keyphrases with DM-crypt